Non-destructive rapid diagnostic evaluation system and method for safety of existing building structures

Abstract

The application provides a nondestructive rapid diagnosis evaluation system and method for safety of existing building structures, relates to the technical field of building pile detection, and utilizes outer ring control to determine termination of evaluation, continuation of detection in the current stage or entering of a higher precision evaluation stage according to the comparison result of the confidence index of the evaluation conclusion and the confidence target threshold value, forms stage progressive control in the macroscopic level, the inner ring control carries out statistical redundancy evaluation on the collected detection data set in the current evaluation stage, then carries out adaptive expansion strategy, forms adaptive adjustment of the detection strategy in the microscopic level, and divides the evaluation conclusion into four levels of Au, Bu, Cu and Du according to the grading principle of the foundation safety appraisal in the national standard, so that the detection evaluation result can be directly marked against the engineering acceptance and safety appraisal standard, the dynamic balance between the detection cost and the data sufficiency is realized, the cost is reduced to avoid resource waste, and the evaluation report has usability.

Description

Technical Field

[0001] This invention relates to the field of building foundation pile testing technology, specifically to a non-destructive rapid diagnostic assessment system and method for the safety of existing building structures. Background Technology

[0002] As critical components that transfer the load of the superstructure to the foundation, the foundation piles in existing buildings directly affect the overall service safety and remaining lifespan of the building. For foundation piles in existing buildings, since the superstructure has already been constructed and the pile tops are covered by beams, slabs, columns, and other superstructure elements, traditional pile foundation testing methods, such as the low-strain reflected wave method, require applying excitation signals to the pile tops, and the acoustic transmission method requires pre-embedding acoustic logging tubes before concrete pouring. These methods cannot be directly applied in the context of existing buildings. Currently, the testing of foundation piles in existing buildings typically yields only limited information from finite drilling, partial excavation, or structural response inversion, resulting in highly sparse data. Furthermore, the testing of foundation piles in existing buildings often requires destructive sampling or large-scale excavation, leading to high costs and disruption to the normal use of the building. Existing technologies suffer from the following technical problems: Problem 1: In the process of inspecting the foundation piles of existing buildings, the accessibility of inspection data is poor and the available inspection data is highly sparse because the pile top is covered by the superstructure. Moreover, the cost of inspection methods with different accuracies varies greatly. The existing technology adopts a preset fixed inspection process and cannot adaptively and dynamically adjust the subsequent inspection strategy according to the statistical characteristics of the acquired data. This makes it difficult to achieve a dynamic balance between inspection cost and evaluation accuracy, and it is easy to have problems such as inspection cost overruns and insufficient data sufficiency or excessive data collection leading to cost waste. Question 2: Existing methods for testing and evaluating existing building foundation piles often output test results that are merely physical parameters such as defect location, defect size, and integrity category, as well as binary judgments of pass or fail. These assessment results lack quantitative confidence indicators, making it difficult for project decision-makers to determine the credibility of the assessment results. Furthermore, there is a lack of step-by-step communication channels between the test and evaluation results and subsequent handling recommendations, making it difficult to directly use the assessment reports for project acceptance, safety assessment, and reinforcement decisions. Summary of the Invention

[0003] To achieve the above objectives, the present invention provides the following technical solution: a non-destructive rapid diagnostic assessment method for the structural safety of existing buildings, the method comprising: For the evaluation of existing building configurations, a confidence target threshold, cost ceiling, and combination of detection methods are set for each evaluation stage. Upon entering the current assessment phase, the detection data of existing building foundation piles are collected and stored in the current phase dataset according to the combination of detection methods configured for the assessment phase. The statistical redundancy of the current dataset is evaluated, and the statistical redundancy index is calculated. Based on the data sufficiency threshold preset for the current evaluation stage, it is determined whether the statistical redundancy index meets the requirements. If not, the inner loop adaptive expansion strategy is executed and the statistical redundancy evaluation is performed again until the requirements are met or the cumulative detection cost reaches the cost limit preset for the current evaluation stage. For the current stage dataset after statistical redundancy assessment, calculate the security level assessment conclusion and the corresponding confidence index, and compare the confidence index with the preset confidence target threshold for the current stage: If the confidence index is greater than or equal to the confidence target threshold, the security level assessment conclusion is output and the assessment process is terminated; otherwise, it is determined whether the detection cost has reached the cost limit of the current assessment stage. If not, the step of performing statistical redundancy assessment on the dataset of the current stage is returned; if so, the next assessment stage is entered.

[0004] On the other hand, this invention proposes a non-destructive rapid diagnostic assessment system for the safety of existing building structures, used to implement the aforementioned non-destructive rapid diagnostic assessment method for the safety of existing building structures. The system includes: The information collection and stage configuration module is used to configure the evaluation stage sequence for the existing building to be evaluated. For each evaluation stage, the confidence target threshold, cost limit and detection method combination are set. When entering the current evaluation stage, the detection data of the existing building foundation piles are collected according to the detection method combination configured for the evaluation stage and then stored in the current stage dataset. The inner loop control module is used to evaluate the statistical redundancy of the dataset in the current stage, calculate the statistical redundancy index, and determine whether the statistical redundancy index meets the requirements based on the preset data sufficiency threshold in the current evaluation stage. If not, the inner loop adaptive expansion strategy is executed and the statistical redundancy evaluation is performed again until the requirements are met or the cumulative detection cost reaches the preset cost limit in the current evaluation stage. The outer loop control module is used to calculate the safety level assessment conclusion and the corresponding confidence index, and compare the confidence index with the preset confidence target threshold for the current stage. If the confidence index is greater than or equal to the confidence target threshold, the security level assessment conclusion is output and the assessment process is terminated; otherwise, it is determined whether the detection cost has reached the cost limit of the current assessment stage. If not, it returns to the inner loop control module; if so, it proceeds to the next assessment stage. The assessment output module is used for the final assessment conclusion, generating a comprehensive assessment report that includes the security level assessment conclusion, confidence index, statistical redundancy assessment results, and handling recommendations.

[0005] This invention provides a non-destructive rapid diagnostic assessment system and method for the structural safety of existing buildings. It offers the following advantages: This invention achieves adaptive cost optimization of the detection and evaluation process by constructing a two-layer collaborative architecture of outer-loop control and inner-loop control. The outer-loop control uses the comparison between the confidence index of the evaluation conclusion and the target confidence threshold to determine whether to terminate the evaluation, continue detection at the current stage, or enter a higher-precision evaluation stage, forming a macro-level stage-progressive control. The inner-loop control performs a statistical redundancy assessment on the collected detection dataset within the current evaluation stage. When indicators such as the coefficient of variation and sample size sufficiency index do not reach the data sufficiency threshold, a spatial expansion strategy of increasing detection points is prioritized. Only when spatial expansion cannot effectively reduce the coefficient of variation is a strategy of upgrading the accuracy of detection methods implemented, forming a micro-level adaptive adjustment of the detection strategy. Through the collaborative operation of the inner and outer loops, while ensuring the data sufficiency of the evaluation conclusion, a dataset meeting the statistical redundancy requirements is obtained with the fewest number of detections and the lowest detection cost. The evaluation process automatically terminates when the confidence threshold is met, achieving a dynamic balance between detection cost and data sufficiency, avoiding cost waste or insufficient data caused by preset fixed processes.

[0006] This invention establishes a direct mapping channel from test data to national standard safety levels and outputs evaluation conclusions with quantified confidence indices. By calculating on a dataset that has undergone statistical redundancy assessment and adaptive expansion, the evaluation conclusions are divided into four levels—Au, Bu, Cu, and Du—according to the grading principles of foundation safety assessment in national standards. This allows the test evaluation results to directly align with engineering acceptance and safety assessment standards. Simultaneously, the confidence index of the evaluation conclusions comprehensively considers data sufficiency, test consistency, and model determinism, quantifying the credibility of the evaluation conclusions as a percentage. Engineering decision-makers can determine whether supplementary testing or improved testing accuracy is needed based on the confidence level. Differentiated treatment strategies are automatically matched according to the final safety level, prioritizing pile foundations requiring treatment according to their urgency. This makes the evaluation report readily usable, forming a step-by-step direct connection between the test evaluation results and subsequent engineering decisions such as reinforcement, monitoring, and re-inspection. This solves the problem of the disconnect between test conclusions and engineering applications in existing technologies. Attached Figure Description

[0007] Figure 1 This is a flowchart of the non-destructive rapid diagnostic assessment method for the safety of existing building structures according to the present invention; Figure 2 This is an architecture diagram of the non-destructive rapid diagnostic assessment system for the structural safety of existing buildings according to the present invention. Detailed Implementation

[0008] 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, and 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.

[0009] like Figure 1 A non-destructive rapid diagnostic assessment method for the structural safety of existing buildings, the method including: Step S100: Obtain the importance level and foundation design level of the existing building to be evaluated. Configure an evaluation stage sequence for the existing building to be evaluated, and set the confidence target threshold, cost ceiling, and combination of testing methods for each evaluation stage. Obtain the design documents, construction records, engineering geological survey reports, and historical testing data of the existing building to be evaluated, and extract basic information such as the type, quantity, design bearing capacity, construction technology, and service life of the foundation piles. Based on the building's importance level and foundation design level, and referring to the grading principles of foundation safety appraisal in national standard GB55021-2021, configure the evaluation stage sequence and corresponding stage parameters to obtain the evaluation stage sequence. The assessment phase sequence pre-defines a macro-level progressive path for the entire assessment process, from extensive to intensive, and from low-cost to high-cost. It clarifies the various stages that the assessment work may go through and their order. It does not require the use of the highest precision and highest cost testing methods for all buildings and all situations. Instead, it starts from a basic assessment stage based on the safety level requirements of existing buildings. Only when the confidence level of the assessment conclusion of the current stage is insufficient to meet the requirements will it automatically trigger the next higher precision stage. This approach obtains assessment conclusions with sufficient credibility to meet decision-making needs at the lowest possible cost, avoiding the waste of resources caused by directly using complex methods and preventing unreliable conclusions due to overly simplistic methods.

[0010] Step S200: Enter the current evaluation stage. Following the configured combination of detection methods for this evaluation stage, collect the detection data of existing building foundation piles and store it in the current stage dataset. The current stage dataset serves as the sole data basis for all subsequent analyses, redundancy assessments, and conclusion calculations within a specific evaluation stage. It is a dynamic and expandable data container, initially loaded with the detection data of representative pile foundations selected according to rules, and continuously incorporating data collected again after new acquisition methods or upgrades during the execution of the inner-loop adaptive expansion strategy.

[0011] Step S300: Statistical redundancy assessment is performed on the current stage dataset. A statistical redundancy index is calculated, and based on the preset data sufficiency threshold for the current assessment stage, it is determined whether the statistical redundancy index meets the requirements. If yes, proceed to step S400; otherwise, execute the inner-loop adaptive expansion strategy and perform another statistical redundancy assessment until the statistical redundancy index meets the data sufficiency threshold or the cumulative detection cost reaches the preset cost limit for the current assessment stage. The statistical redundancy index reflects the representativeness and sufficiency of the current dataset. Representativeness indicates whether the few detected pile foundations can represent all pile foundations of the entire building, reflected by the coefficient of variation. Sufficiency indicates whether the number and spatial distribution of the detected samples are sufficient to cover all possible situations of the total pile foundations and the resulting estimation accuracy, reflected by the sample size sufficiency index and confidence interval width. If all three parameters of representativeness and sufficiency meet the data sufficiency threshold, it indicates that the statistical redundancy of the current dataset is sufficient, and the conclusions drawn based on it are reliable.

[0012] Step S400: For the current stage dataset after statistical redundancy assessment and inner loop adaptive expansion strategy, calculate the current stage security level assessment conclusion and corresponding confidence index, and compare the confidence index with the current stage preset confidence target threshold. If the confidence index is greater than or equal to the confidence target threshold, the safety level assessment conclusion for the current stage is output and the assessment process terminates. Otherwise, it is determined whether the detection cost has reached the cost limit for the current assessment stage. If not, the process returns to the step of evaluating the statistical redundancy of the dataset for the current stage and continues to execute the inner-loop adaptive expansion strategy within the current stage. If so, the process returns to the step of entering the current assessment stage and proceeds to the next assessment stage. The calculated safety level assessment conclusion reflects the final qualitative or grading judgment of the current safety status of the existing building foundation, transforming the raw and complex detection data into decision-making basis that engineering and management personnel can directly understand and use. The calculated confidence index reflects a quantitative evaluation of the credibility of this safety level conclusion. The process only terminates when the credibility of the conclusion is sufficiently high, thus avoiding drawing conclusions when evidence is insufficient or continuing unnecessary detection when evidence is sufficient.

[0013] In this embodiment, the configured evaluation stage sequence includes a rapid screening stage, a routine detailed investigation stage, and a high-precision verification stage, which correspond to the stage indexes of the evaluation stage sequence as stage 0, stage 1, and stage 2, respectively. Phase 0 is the rapid screening phase, which uses a combination of methods including tapping, rebound hammer testing, and crack observation, with a target confidence threshold of 70% to 80%. This rapid screening phase aims to conduct a preliminary inspection of building foundation piles at the lowest cost and fastest speed, to rule out obvious and serious safety issues or to quickly provide a preliminary, low-confidence safety assessment.

[0014] Phase 1 is the routine detailed investigation phase, employing a combination of low-strain reflected wave method or side-hole transmitted wave method as detection techniques, with a target confidence threshold of 80% to 90%. Following the routine detailed investigation phase, if the conclusions of the rapid screening phase are unclear or lack sufficient confidence, industry-standard detection methods are used for more in-depth data collection to obtain assessment conclusions with higher confidence.

[0015] Phase 2 is the high-precision verification phase, which uses a combination of detection methods such as acoustic tomography, core drilling, or static load testing, with a confidence threshold of 90% to 99%. For complex or important buildings where high-confidence conclusions cannot be reached in the first two phases, the high-precision verification phase employs the most accurate, reliable, but also the most expensive method for final verification.

[0016] The confidence thresholds for each stage increase in a stepwise manner, with a minimum difference of five percentage points between adjacent stages. Stage 0 has the lowest cost ceiling, while Stage 2 has the highest. Following the stage index, the parameters for the cost ceiling, confidence threshold, and detection methods are arranged in a stepwise manner, forming a natural and gradual upgrade path. This allows the evaluation system to automatically find the most economical evaluation endpoint that meets the required confidence level at minimal cost, avoiding over- or under-detection caused by pre-set fixed procedures.

[0017] The cost cap sets a limit on the testing expenditure for each evaluation stage, preventing unlimited investment in any one stage and reflecting control over the economics of the evaluation. It is set based on the market price and estimated workload of the pre-set testing methods for different stages. The confidence target threshold defines the credibility threshold that the evaluation conclusions of each stage need to reach, serving as a decision-making boundary between terminating the evaluation and proceeding to the next stage. It is set based on the general understanding of the reliability of conclusions achievable by different precision testing methods in engineering practice. The testing method combination is the specific technical toolkit for achieving the evaluation objectives of this stage, determining the type and precision of raw data that can be collected. It is set based on the physical principles, applicable scope, precision level, and cost of different testing methods.

[0018] In this embodiment, the specific implementation steps of step S200 include: Before starting the inspection, based on the evaluation stage sequence configured in step S100, the first evaluation stage of the building to be evaluated is directly determined as the first stage in the sequence, i.e., the rapid screening stage corresponding to stage index k=0. This is a fixed initial setting that starts from the most basic stage, ensuring that any evaluation should begin with the lowest-cost and most convenient preliminary work. Only when the conclusions of the preliminary work fail to meet the requirements is it necessary to start more complex and costly subsequent stages. This avoids using expensive methods directly on buildings that do not require high-precision evaluation. Initialize the current stage index k=0, enter stage k, and collect inspection data on the existing building foundation piles to be evaluated according to the combination of inspection methods configured in stage k. The specific content of the inspection data collection includes one or more of the following combinations: The test data of accessible parts of the foundation and superstructure include acoustic waveform data obtained by the impact method, concrete surface strength data obtained by the rebound hammer, crack width observation data, and steel reinforcement protective layer thickness measurement data. Core sample integrity data, pile concrete strength data, and pile bottom sediment thickness data obtained by drilling method; pile wave velocity curve data, pile impedance change data, and pile top displacement response data obtained by low strain reflection wave method; Data on the first arrival time and amplitude variation of seismic waves obtained by the side-hole transmission wave method; multi-section ultrasonic transmission waveform data obtained by the acoustic CT method; core sample photographs and core sample integrity records obtained by the core drilling method.

[0019] The collected data is stored according to a pre-defined data structure, with each data entry including a data source identifier, a collection timestamp, and collection location coordinates. The initial construction of the detection dataset focuses on representative pile foundations, selected according to the following rules: priority is given to pile foundations located at corners, pile foundations under the highest stress (such as those corresponding to the column bases of the superstructure), pile foundations with questionable construction records, and pile foundations in areas with significant variations in foundation soil conditions. The initial number of collected pile foundations should not be less than 10% of the total number of building pile foundations, and should not be less than 3. The collected detection data is then stored in the current stage dataset.

[0020] Among these methods, the impact method involves striking the pile top or foundation with a hand hammer or vibratory hammer, using sensors installed on the structure to receive the acoustic waveforms and preliminarily assess the pile integrity. The rebound hammer method estimates the surface strength of concrete by measuring its rebound value. Crack observation uses a crack width observer to directly measure and record the width, length, and direction of cracks in accessible parts of the foundation or superstructure. Reinforcement cover thickness measurement uses a reinforcement detector based on electromagnetic induction to non-destructively determine the cover thickness. The drilling method involves drilling holes in or around the pile to obtain core samples for visual observation of core integrity and to measure concrete strength. The low-strain reflected wave method applies a transient impact force to the pile top and analyzes the reflected wave curves to determine changes in pile impedance and the location of defects. The side-hole transmission wave method involves drilling holes in the soil beside the pile, vibrating the pile top, and evaluating pile integrity and pile bottom conditions by receiving the first arrival time and amplitude changes of seismic waves at different depths in the hole. The acoustic CT method involves pre-embedding multiple acoustic transmitting and receiving probes within the pile or drilling holes, collecting a large number of cross-hole transmission waveforms to invert the wave velocity distribution inside the pile. Core drilling involves directly drilling concrete core samples from the pile shaft. By observing the core sample surface and conducting compressive strength tests, it is one of the most intuitive and reliable testing methods. Static load testing involves applying loads to the pile top in stages and measuring the pile top settlement to directly determine the vertical compressive (or tensile) bearing capacity of a single pile. It is a recognized benchmark method for bearing capacity testing. The data acquisition methods for each stage in the evaluation sequence are mature technologies widely used in my country's building foundation pile testing field and supported by national or industry standards. The specific equipment and sensors used are clearly defined in existing technologies and will not be elaborated upon here. In practice, selection should be made according to specific needs.

[0021] When selecting representative piles, piles at the corners typically bear greater overturning moments and shear forces from wind loads, seismic forces, or eccentric loads, and their constraints are weaker than those of internal piles. These are sensitive areas of the structure and prone to damage first; therefore, piles at the corners are given priority as representative piles. Pile foundations bearing the greatest stress directly bear the maximum vertical load from the superstructure, and their safety margin is relatively small. Damage to these piles would have the most severe impact on the overall building safety; therefore, they are also given priority as representative piles. Construction issues such as borehole collapse, necking, excessive sediment buildup, and concrete segregation are the main causes of pile foundation defects. The probability of defects in these questionable pile foundations is much higher than in pile foundations constructed under normal conditions; therefore, pile foundations with questionable construction records are given priority as representative pile foundations. Due to complex or changing geological conditions, such as alternating layers of soft and hard soil, the presence of lenses, and large undulations in the rock surface, the bearing capacity and deformation characteristics of pile foundations at different locations on the same site can vary significantly. Pile foundations located in unfavorable geological areas are at higher risk. Therefore, pile foundations in areas with significant variations in foundation soil conditions are preferred as representative pile foundations.

[0022] In this embodiment, the implementation steps for evaluating the statistical redundancy of the current stage dataset and calculating the statistical redundancy index include: Step S301: Calculate the coefficient of variation, sample adequacy index, and confidence interval width based on the current stage dataset to form a statistical redundancy index. The coefficient of variation is used to assess the consistency and dispersion of the test data among the tested pile foundations. It is obtained by calculating the ratio of the standard deviation to the mean of the current stage dataset, quantifying the difference between the test results of different pile foundations. The smaller the coefficient of variation, the closer the performance of each pile foundation is, and the stronger the representativeness of the test results of the current small number of pile foundations to the untested pile foundations; conversely, a larger coefficient of variation indicates a large difference in pile foundation performance, requiring more samples to be tested. The sample adequacy index is a composite index of the number of detected pile foundations and spatial distribution coverage, ranging from 0 to 1. It includes two parts: the degree to which the current detected sample size meets the preset minimum sample size, and the degree of matching between the spatial distribution density of the detected pile foundations and the overall spatial distribution of all pile foundations. It is used to determine whether the existing sample is sufficient to statistically and spatially represent the total number of pile foundations. When calculating the sample adequacy index: First, the basic sample size score is calculated by dividing the current number of detected pile foundations by the preset minimum sample size, i.e., the initial number of pile foundations collected in step S200. If the result is greater than 1, it is counted as 1. Then, the spatial coverage score is calculated by dividing the entire building pile foundation plan into multiple grid areas and dividing the number of grids occupied by the detected pile foundations by the total number of grids. Finally, the basic sample size score and the spatial coverage score are weighted and summed, for example, each with a 50% weight, to obtain the final sample adequacy index. The closer the value is to 1, the more adequacy the sample size.

[0023] The confidence interval width is a quantitative expression of the accuracy of the current assessment conclusion (such as the average bearing capacity of pile foundations). The smaller the confidence interval width, the more accurate the estimate and the greater the confidence in the assessment conclusion; the larger the confidence interval width, the higher the uncertainty. To calculate the confidence interval width, firstly, based on the current stage dataset, select a core assessment indicator, such as average bearing capacity, and calculate its sample mean. Then, calculate the standard deviation and standard error of this indicator; the standard error is the standard deviation divided by the square root of the number of tested pile foundations. Next, based on a preset confidence level (e.g., 95%), obtain a coefficient from a statistical distribution table, such as the critical value corresponding to a normal distribution. Multiply the standard error by this coefficient to obtain the marginal error. Finally, subtract the marginal error from the sample mean to obtain the lower confidence limit, and add the marginal error to obtain the upper confidence limit. The difference between the two is the confidence interval width.

[0024] Step S302: The calculated coefficient of variation, sample adequacy index, and confidence interval width form a statistical redundancy index. This index is used to determine whether the existing data simultaneously meets the three conditions of data consistency, sample adequacy, and accurate estimation. The coefficient of variation, sample adequacy index, and confidence interval width are compared with the data adequacy threshold for the current stage. Based on the preset data adequacy threshold for stage k, it is determined whether the statistical redundancy of the dataset in the current stage meets the requirements. The data adequacy threshold includes the preset coefficient of variation threshold, sample adequacy index threshold, and confidence interval width threshold for the current stage. The calculated statistical redundancy index is compared with the corresponding data adequacy threshold. Only when the coefficient of variation is less than the coefficient of variation threshold, the sample adequacy index is greater than or equal to the sample adequacy index threshold, and the confidence interval width is less than the confidence interval width threshold, is the dataset in the current stage deemed to meet the data adequacy threshold requirements. After determining that the statistical redundancy assessment requirements are met, step S500 is executed. Conversely, if any condition is not met, it is determined to be unmet, and the inner-loop adaptive expansion strategy steps of steps S302 and S303 are executed.

[0025] The data sufficiency threshold is a set of preset critical values ​​used in each evaluation stage to determine whether the statistical redundancy of the current dataset meets the standard. The coefficient of variation threshold sets the upper limit for data consistency; exceeding this limit indicates excessive differences in pile foundation performance, requiring additional testing. It is set based on the inherent accuracy of the testing methods at that stage and the maximum acceptable dispersion based on engineering experience. The threshold is more lenient in the rapid screening stage and more stringent in the high-precision stage. The sample size sufficiency index threshold sets the lower limit for sample coverage; below this limit indicates insufficient sample size or spatial distribution. It is set based on sampling theory and spatial analysis requirements; the higher the corresponding stage index, the more stringent the representativeness requirement, and the higher the sample size sufficiency index threshold. The confidence interval width threshold sets the upper limit for estimation accuracy; exceeding this limit indicates the conclusion is too coarse. It is set based on the maximum uncertainty range tolerated by engineering decisions. High-precision stages require narrower confidence intervals; that is, stages with higher stage indices have smaller confidence interval width thresholds. Only when data simultaneously meets the requirements of consistency, representativeness, and accuracy can it be considered to have sufficient data redundancy.

[0026] In this embodiment, the adaptive expansion strategy includes a first priority sub-step and a second priority sub-step, and the execution process of prioritizing the first priority sub-step includes: Step S303: Based on the spatial coordinates of the detected points in the current stage dataset within the overall distribution map of the building foundation piles, identify the current spatial sampling blank area. The current spatial sampling blank area is the area where the spatial distance between adjacent detected points exceeds a preset spatial threshold. The spatial distance between adjacent detected points in the overall distribution map of the foundation piles is calculated using the Euclidean distance formula, which is to calculate the square root of the sum of the squares of the differences in the horizontal and vertical coordinates between the two points. The spatial coordinates of the overall pile distribution map are directly obtained from the as-built drawings or foundation layout plans of the existing building to be evaluated. A Cartesian coordinate system is established using a fixed reference point on the building, such as the intersection of the exterior wall axes at the southwest corner or the intersection of axis A and axis I, as the coordinate origin (0,0). The horizontal distance (X-coordinate) and horizontal longitudinal distance (Y-coordinate) of each pile relative to this origin are measured on the drawings or calculated based on the dimensions marked on the drawings, forming a coordinate list containing the location information of all piles. For existing buildings without electronic drawings, the projected position of each pile is measured and assigned coordinates using a total station or laser rangefinder, with the building's exterior walls or existing permanent features as references. This coordinate list is only used to describe the relative spatial relationships between piles within the building. A preset spatial threshold is a pre-defined distance threshold. When the spatial distance between adjacent detected points exceeds the preset spatial threshold, the area between these two points is considered to lack data coverage and requires supplementary detection. The preset spatial threshold is set based on the spatial variability of pile foundation testing and engineering experience. Within the same building site, geological conditions and construction quality may change significantly over short distances, such as 5 to 10 meters. Therefore, it is necessary to ensure that there are testing points within this scale. It is usually set according to the column grid spacing and pile spacing of the building to ensure that there is at least one testing sample in each major structural unit or geological zone.

[0027] Based on preset point selection rules, expanded testing points are selected from the spatial sampling gap area. These rules include selecting untested piles located between already tested piles, selecting piles whose geological conditions differ from those of the tested piles by a preset value, selecting piles located at building corners or with special stress conditions, and selecting piles with questionable construction records. The point selection rules are set based on the understanding of high-risk pile foundation locations in engineering practice and the statistical needs for spatial filling. By selecting untested piles located between already tested piles, spatial sampling gaps are filled, directly reducing the maximum distance between adjacent testing points. By selecting piles whose geological conditions differ from those of the tested piles by a preset value, the sample ensures coverage of different geological units. By selecting piles located at building corners or with special stress conditions, performance data of the most unfavorable structural stress locations are captured. By selecting piles with questionable construction records, high-risk targets are targeted for investigation.

[0028] Based on the selected expanded testing sites, supplementary testing is conducted using the same testing methods and combinations as the current stage. After each round of expanded testing, the number of piles tested in each round should not be less than 30% of the currently tested piles. The newly collected testing data is then incorporated into the current stage dataset, and the statistical redundancy index is recalculated. By expanding the testing sites, the sample size of the current stage dataset is directly increased, improving the spatial distribution uniformity of the samples and more comprehensively capturing the overall performance variability of the pile foundations. This reduces the coefficient of variation of the dataset, increases the sample sufficiency index, and narrows the confidence interval width, thereby ensuring that the statistical redundancy index meets the data sufficiency threshold requirements.

[0029] Repeat the process of the first priority sub-step until the termination condition of the first priority sub-step is met. The first priority sub-step terminates when the termination condition is met. The termination condition includes: Condition 1: The statistical redundancy index of the current dataset has met the data sufficiency threshold, indicating that sufficient data can be obtained simply by increasing the number of detection points. Condition 2: The cumulative testing cost has reached more than 90% of the preset cost limit for this stage, indicating that the budget has been exhausted and it is impossible to continue to expand testing to more locations. Condition 3: After three rounds of augmentation testing, the coefficient of variation of the dataset decreased very little compared to the previous round of augmentation (less than the preset improvement threshold, such as 1% or 2%), indicating that simply increasing the number of detection points can no longer effectively improve the consistency of the data. The data dispersion may be due to some special outliers or systematic measurement errors. In this case, adding more points is a waste of resources and other strategies need to be used.

[0030] The first priority sub-step is terminated when any one of the termination conditions is met.

[0031] In this embodiment, the execution process of the second priority sub-step includes: Step S304: If, when the first priority sub-step is executed, the cumulative detection cost has reached more than 90% of the preset cost limit for the current stage, and the statistical redundancy index of the dataset in the current stage still does not meet the data sufficiency threshold requirement, or the reduction in the coefficient of variation after three consecutive rounds of execution of the first priority sub-step is less than the preset improvement threshold, then the second priority sub-step is triggered. Upgrade detection methods for key detection points. Identify piles with the largest contribution to the coefficient of variation, piles at the edge of spatial distribution, and piles with special stress conditions from the current dataset as key detection points. By identifying key detection points, select the data in the current dataset that contributes the most to uncertainty or discreteness for secondary detection to evaluate the accuracy of this type of data.

[0032] For key testing points, a combination of testing methods with higher precision than the currently preset combination is employed. For example, this could be upgraded from impact testing to low-strain reflected wave testing, from shallow core drilling to full-section core drilling, or from low-strain testing to acoustic CT. The higher-precision combination of testing methods is used to re-collect testing data and supplement the corresponding pile records in the current stage dataset. By re-collecting data at selected key testing points using more precise and reliable methods than the currently preset methods, performance parameters of the pile that more closely approximate its true state are obtained. This replaces or corrects inaccurate data that may have been introduced by low-precision methods, directly reducing the coefficient of variation of the dataset and improving the confidence level of the conclusions.

[0033] Perform a statistical redundancy assessment on the supplemented current stage dataset, and repeat the process of upgrading detection methods for key detection points until the statistical redundancy of the current stage dataset meets the data sufficiency threshold or the cumulative detection cost reaches the stage's preset cost limit. The second priority sub-step then terminates.

[0034] If, during the execution of the first or second priority sub-step, the cumulative detection cost has reached the pre-set cost limit for the stage, but the statistical redundancy of the current stage dataset still does not meet the data sufficiency threshold requirement, then the inner loop adaptive expansion strategy is terminated, the current stage dataset is used as the final dataset for this stage, a prompt message indicating insufficient data redundancy is output, and the specific gap indicators for the current stage not meeting the data sufficiency threshold are recorded. Then, the process proceeds to step S400 to calculate the stage evaluation conclusion.

[0035] In this embodiment, the process of calculating the security level assessment conclusion and corresponding confidence index for the current stage dataset after statistical redundancy assessment includes: Step S401: Obtain the current stage dataset after statistical redundancy assessment, calculate the safety level assessment conclusion for the current assessment stage, and classify the safety level assessment conclusion into Au, Bu, Cu, and Du grades according to the grading principle of foundation safety appraisal in national standards; Au grade indicates safety, meeting the safety requirements of current national standards; Bu grade indicates basic safety, slightly below the current standard requirements but not affecting the overall load-bearing capacity; Cu grade indicates the existence of safety hazards, and reinforcement or other measures should be taken; Du grade indicates the existence of serious safety hazards, and immediate risk mitigation measures should be taken. When generating safety level assessment conclusions based on the current stage dataset, the process begins by using the test data in the current stage dataset, such as low-strain wave velocity curves and core sample photographs, to classify each tested pile into four categories based on the pile integrity classification standards specified in the national standard "Technical Specification for Testing of Building Foundation Piles" (JGJ106-2014). Category I piles are considered intact; Category II piles have minor defects that do not affect bearing capacity; Category III piles have significant defects requiring further verification; and Category IV piles have severe defects and are unusable. The number and proportion of each type of integrity pile are then statistically analyzed across all tested pile foundations. Next, soil parameters (such as pile side friction and standard values ​​of pile end resistance) are extracted from the engineering geological survey report for each tested pile, and bearing capacity calculation parameters are extracted from the design documents. Combined with the pile integrity category, the reduction in the current actual bearing capacity of each pile relative to the design bearing capacity is estimated. Finally, based on the grading principles for foundation safety assessment in the national standard GB55021-2021, a comprehensive evaluation is conducted, taking into account factors such as the proportion of Class III and Class IV piles, the overall reduction in bearing capacity, and the building's importance level. If all tested piles are Class I or II piles and their bearing capacity meets the requirements, the grade is Au; if a small number of Class III piles exist but the overall bearing capacity is still basically met, the grade is Bu; if the proportion of Class III piles is high or there is a significant deficiency in bearing capacity, the grade is Cu; if any Class IV piles are found or the bearing capacity is severely insufficient, the grade is Du. The safety level assessment conclusion is generated based on the pile foundation integrity classification standards and safety assessment grading principles clearly defined in current national standards. The safety level assessment conclusion is the final qualitative judgment of the overall safety status of the existing building's foundation, transforming complex and professional testing data into a legally binding assessment conclusion that can be clearly understood by construction administrative departments, owners, and engineers, and directly guiding subsequent decision-making.

[0036] Step S402: Based on the current stage dataset after statistical redundancy assessment, calculate data sufficiency factors, detection consistency factors, and model determinism factors; The data sufficiency factor quantifies the contribution of the statistical completeness of the current stage dataset to the final confidence level. It includes the achievement of three indicators: the coefficient of variation, the sample size sufficiency index, and the confidence interval width. These reflect the impact of the data quality itself on the credibility of the assessment conclusions; the more complete the data, the higher the score for this factor. When calculating the data sufficiency factor, the coefficient of variation, sample size sufficiency index, and confidence interval width calculated for the current stage dataset are compared with the preset data sufficiency thresholds for that stage. A score is calculated for each indicator based on the range by which it exceeds the corresponding data sufficiency threshold. For example, if the coefficient of variation is less than the threshold by 20%, the score for this item is 1.2, exceeding the maximum score of 1; if it is exactly equal to the threshold, the score is 1.0; if it exceeds the threshold by 10%, the score is 0.9; and so on, calculating the scores for the three indicators after comparison. The scores of the three indicators are then weighted and summed, with each indicator accounting for one-third of the total score. The weighted sum is then normalized to the 0-1 range to obtain the final data sufficiency factor score. The higher the statistical redundancy of the current stage dataset—that is, the smaller the coefficient of variation, the more sufficient the sample size, and the narrower the confidence interval—the more reliable the statistical inferences made based on the current stage dataset, and the higher the confidence level of the conclusions should be.

[0037] The consistency factor is an indicator used to measure the contribution of the degree of mutual corroboration between data from different testing sources to the final confidence level. It includes the degree of difference between results obtained from different testing methods for the same pile foundation, and the degree of spatial consistency between test results of pile foundations in different spatial locations. It is used to assess whether there are contradictions within the data; the better the mutual corroboration between the data, the higher the score of this factor, indicating a more reliable conclusion. When calculating the consistency factor, firstly, for pile foundations with data from at least two testing methods, calculate the difference in integrity level or bearing capacity estimates obtained from different methods; the smaller the difference, the higher the score. Then, sort all tested pile foundations according to their coordinates and corresponding geological parameters, and observe whether the changing trend of the testing indicators (such as bearing capacity) is consistent with the changing trend of the soil layers. For example, the bearing capacity should be lower in areas with a large bearing layer depth. Calculate the correlation coefficient between the actual data and the theoretical trend. Finally, weightedly combine the above two scores to obtain the consistency factor.

[0038] Model determinism is an index used to quantify the inherent uncertainty of the calculation model or empirical formula used in the process of extrapolating from test data to the final safety level conclusion. It is a preset, fixed value related to the assessment stage and the method used, including the error range of the bearing capacity calculation formula, the fuzzy boundary of the integrity classification, etc. It reflects the impact of the theoretical limitations of the assessment method itself on the confidence of the conclusion. The more mature and accurate the model, the higher the value of the model determinism, that is, the smaller the uncertainty. When calculating the model determinism, first determine the core assessment model used in the current assessment stage; for example, in the rapid screening stage, the error of using rebound values ​​to extrapolate concrete strength is about 15-20%, so the model determinism is preset to 0.8; in the routine detailed investigation stage, the error of using the low strain method to estimate bearing capacity is about 10-15%, so the factor is preset to 0.85; in the high-precision verification stage, the core drilling method directly measures strength, and the static load method directly measures bearing capacity. The errors of these methods are extremely small, generally less than 5%, so the factor is preset to 0.95. Then, based on the specific model and specifications used, directly look up the table or use the default value. The calculation of the deterministic factors of the model is based on the accuracy range of various testing methods and calculation models obtained from a large number of experimental studies and engineering statistics, which are recognized in national standards and industry guidelines.

[0039] The confidence index is obtained by weighting the data sufficiency factor, the detection consistency factor, and the model determinism factor. First, the calculated data sufficiency factor, detection consistency factor, and model determinism factor are mapped to values ​​between 0 and 1. Then, these three factors are weighted and summed to obtain the original confidence score. The weights in the weighted summation are set according to the degree of influence of each factor on the credibility of the conclusion. Generally, the data sufficiency factor is set as the foundation, the model determinism factor as the method, and the detection consistency factor as the verification. For example, the weight of the data sufficiency factor is set to 0.5, the weight of the detection consistency factor to 0.3, and the weight of the model determinism factor to 0.2. Finally, the original weighted summation score is multiplied by 100% to convert it into a percentage confidence index. The confidence index comprehensively reflects the overall credibility of the current security level assessment conclusion, with a value range of 0% to 100%. The higher the confidence index, the higher the credibility of the current assessment conclusion.

[0040] In this embodiment, the process of comparing the confidence index with the preset confidence target threshold for the current stage includes: Step S403: When the confidence index is determined to be greater than or equal to the confidence target threshold, the security level assessment conclusion of the current stage is output as the final assessment conclusion, and the following information is recorded: the stage index k at which the assessment is terminated, the confidence index corresponding to the final assessment conclusion, the cumulative detection cost of the current stage, and the statistical redundancy assessment result of the dataset of the current stage. The assessment process terminates, and the process jumps to step S500.

[0041] When the confidence index is determined to be lower than the confidence target threshold, it is further determined whether the cumulative detection cost is greater than or equal to the current stage cost limit; When the confidence index is lower than the target confidence threshold and the cumulative detection cost is less than the current stage's cost ceiling, the main reasons for the insufficient confidence index are further determined. In the gap between the confidence index and the target confidence threshold, the prerequisite for determining the main reasons for the insufficient confidence index is that the difference between the target confidence threshold and the confidence index C0 is a positive number greater than zero, let's call this difference D0. Next, an ideal confidence index C1 is calculated under hypothetical ideal data. That is, while keeping the scores of the current model's deterministic factors and detection consistency factors unchanged, the data sufficiency factor is forcibly set to the maximum value of 1.0, representing perfect data. The confidence index is then recalculated to obtain C1. The gap D1 attributable to insufficient data redundancy is calculated, D1 = C1 - C0. The gap D2 attributable to inherent model uncertainty is calculated, D2 = D0 - D1. If the portion attributable to insufficient statistical redundancy is greater than the portion attributable to the inherent uncertainty of the evaluation model, i.e., D1 > D0, then the main reason is determined to be insufficient data redundancy, and the process returns to the step of evaluating the statistical redundancy of the dataset at the current stage. If the portion attributable to the inherent uncertainty of the assessment model is greater than the portion attributable to insufficient statistical redundancy (i.e., D1≤D0), then the primary reason is determined to be that the inherent uncertainty of the assessment model itself has reached the limit of this stage. The current stage's security level assessment conclusion is output as an intermediate conclusion with a confidence level indicator, and a prompt message suggesting the next assessment stage is generated.

[0042] When the confidence index falls below the confidence target threshold and the cumulative detection cost is greater than or equal to the cost limit of the current stage, a stage progression is triggered. The current stage index k is incremented by 1 (k = k + 1), and the process returns to step S200. The higher-precision detection methods configured for the next evaluation stage are then used for a re-evaluation. Upon entering the next evaluation stage, this stage is taken as the new current stage. The dataset for this stage is reset to empty. The higher-precision detection methods configured for this stage are used to re-collect detection data to form the current stage dataset. The security level evaluation conclusion, confidence index, statistical redundancy evaluation results, and progression reasons from the previous evaluation stage are inherited as prior information to guide the initial selection of detection points for this stage. For example, if the current stage is stage 0, the stage progression becomes stage 1. The current stage dataset is initialized, and the process proceeds to step S200. The detection methods of stage 1 are used to collect foundation pile data, forming the current stage dataset for stage 1. The current stage dataset of stage 0 is used as prior information to guide the selection of detection points for stage 1.

[0043] In this process, as we progress from stage k to stage k+1, the safety level assessment conclusion of the previous assessment stage serves as prior information for the next assessment stage, indicating the key areas of focus for that stage. For example, if the previous assessment stage was at level Cu, the next stage should focus on verifying the pile foundations that pose safety hazards. The confidence index of the previous assessment stage quantifies the credibility of its conclusions, helping to prioritize initial detection points for the next stage. The lower the confidence index, the less reliable the data from the previous assessment stage, and the more necessary it is to conduct basic data collection again in the next stage. The statistical redundancy assessment results from the previous evaluation phase include specific coefficients of variation, sample adequacy index, confidence interval width, and the gap with the corresponding data adequacy threshold. These are used to accurately diagnose whether the deficiencies in the data from the previous evaluation phase are due to excessive data dispersion, insufficient sample size, or inaccurate estimation, thus guiding the development of a more targeted initial detection plan for the next evaluation phase. For example, if the coefficient of variation was too large in the previous evaluation phase, the next evaluation phase should prioritize high-precision retesting of the anomaly stakes causing dispersion; if the sample adequacy index was low, the next evaluation phase should prioritize deploying points in spatially blank areas. Progressive reasons, such as "confidence level not met and cost reached its limit" or "confidence level not met and attributed to model uncertainty," record the direct decision basis for triggering the phase switch, facilitating explanation in subsequent evaluation report generation why increased evaluation costs are necessary to proceed to the next evaluation phase.

[0044] In this embodiment, the method further includes: Step S500: Repeat steps S200 to S400 until any of the following conditions are met, at which point the evaluation process terminates: The confidence index of the security level assessment conclusion at a certain stage reaches the confidence target threshold for that stage; If the highest stage N has been completed, and the confidence index of the evaluation conclusion of this stage has not yet reached the confidence target threshold, but the cumulative detection cost has reached the cost limit of the highest stage, then the evaluation conclusion of the highest stage shall be taken as the final conclusion, and a special mark of "confidence not met" shall be added to the evaluation report; N is the index of the highest stage in the evaluation stage sequence, and N≥2 is taken as an integer; The user manually terminates the evaluation process.

[0045] Once the assessment process is completed, a comprehensive assessment report is generated. This report includes the safety level assessment conclusion, confidence level indicators, confidence interval width, coefficient of variation, sample size adequacy index, and a detailed breakdown of testing costs at each stage. Based on the safety level assessment conclusion, corresponding treatment strategies are matched: Au-level piles require no treatment and are included in routine monitoring; Bu-level piles can continue to be used and periodic re-inspections are recommended; Cu-level piles should be reinforced and re-inspected after reinforcement; and Du-level piles should be immediately addressed. For piles requiring treatment, priority is given based on the severity of damage reflected in the testing data, and the results are then included in the report.

[0046] After the assessment process is terminated, all recorded information is first collected and summarized from various modules of the system, including the initial parameters of the information acquisition and phase configuration module, the expansion records of each round of the inner loop control module, the phase switching records of the outer loop control module, and the final conclusions of the assessment output module. Then, the pile foundation test data is matched with the integrity classification results to generate test results for each tested pile. Following a preset report template, basic information is filled into the corresponding sections, and the assessment process records are organized into clear flowcharts and text descriptions according to chronological or phase order. Finally, the final assessment conclusions and confidence levels are prominently presented. Based on the level results given in the safety level assessment conclusion, corresponding treatment suggestions are retrieved from the preset strategy library. Based on the severity of damage reflected in the test data (e.g., Class III piles are more urgent than Class II piles) and the importance of the pile foundation (e.g., corner piles are more urgent than internal piles), all pile foundations requiring treatment are prioritized (e.g., emergency treatment, planned treatment, routine monitoring, etc.), and written into the report, along with a limitation statement of the assessment conclusions, such as "This conclusion is based on the low strain method with a confidence level of 85%." The final PDF or paper report is then generated.

[0047] like Figure 2 A non-destructive rapid diagnostic assessment system for the structural safety of existing buildings, comprising: The information collection and stage configuration module is used to acquire design documents, construction records, engineering geological survey reports, and historical testing data of existing buildings. It supports data collection interfaces for various testing methods. Based on the building's importance level and foundation design level, it configures an assessment stage sequence for the existing building to be evaluated, and sets a confidence target threshold, cost ceiling, and testing method combination for each assessment stage. Upon entering the current assessment stage, it collects the testing data of the existing building's foundation piles according to the testing method combination configured for the assessment stage and stores it in the current stage's dataset. The inner loop control module is used to evaluate the statistical redundancy of the dataset in the current stage, calculate the statistical redundancy index, and determine whether the statistical redundancy index meets the requirements based on the preset data sufficiency threshold in the current evaluation stage. If not, the inner loop adaptive expansion strategy is executed and the statistical redundancy evaluation is performed again until the requirements are met or the cumulative detection cost reaches the preset cost limit in the current evaluation stage. The outer loop control module calculates the security level assessment conclusion and corresponding confidence index for the current stage dataset after statistical redundancy evaluation. It compares the confidence index with the preset confidence target threshold for the current stage. If the confidence index is greater than or equal to the target threshold, the security level assessment conclusion is output and the assessment process terminates. Otherwise, it checks whether the detection cost has reached the cost limit for the current assessment stage. If not, it returns to the step of performing statistical redundancy evaluation on the current stage dataset; if so, it proceeds to the next assessment stage. The outer loop control module manages the progressive switching of assessment stages, receives the confidence index of the current stage security level assessment conclusion, compares it with the preset confidence target threshold for the current stage, and decides whether to terminate the assessment, continue detection in the current stage, or proceed to the next assessment stage based on the comparison result.

[0048] The assessment output module is used for the final assessment conclusion, generating a comprehensive assessment report that includes the security level assessment conclusion, confidence index, statistical redundancy assessment results, and handling recommendations.

[0049] In this embodiment, an adaptive cost optimization of the detection and evaluation process is achieved by constructing a two-layer collaborative architecture of outer-loop control and inner-loop control. The outer-loop control uses the comparison results between the confidence index of the evaluation conclusion and the target confidence threshold to decide whether to terminate the evaluation, continue detection at the current stage, or enter a higher-precision evaluation stage, forming a macro-level stage-progressive control. The inner-loop control performs a statistical redundancy assessment on the collected detection dataset within the current evaluation stage. When indicators such as the coefficient of variation and sample sufficiency index do not reach the data sufficiency threshold, a spatial expansion strategy of increasing detection points is prioritized. Only when spatial expansion cannot effectively reduce the coefficient of variation is a strategy of upgrading the accuracy of detection methods implemented, forming a micro-level adaptive adjustment of the detection strategy. Through the collaborative operation of the inner and outer loops, while ensuring the data sufficiency of the evaluation conclusion, a dataset meeting the statistical redundancy requirements is obtained with the fewest number of detections and the lowest detection cost. The evaluation process is automatically terminated when the confidence level is met, achieving a dynamic balance between detection cost and data sufficiency, avoiding cost waste or insufficient data caused by preset fixed processes.

[0050] A direct mapping channel from testing data to national standard safety levels has been established, and assessment conclusions with quantified confidence indices are output. By calculating on a dataset that has undergone statistical redundancy assessment and adaptive expansion, the assessment conclusions are divided into four levels—Au, Bu, Cu, and Du—according to the grading principles of foundation safety assessment in national standards. This allows the testing and assessment results to directly align with engineering acceptance and safety assessment standards. Simultaneously, the confidence index of the assessment conclusions comprehensively considers data sufficiency, testing consistency, and model determinism, quantifying the credibility of the assessment conclusions as a percentage. Engineering decision-makers can determine whether supplementary testing or improved testing accuracy is needed based on the confidence level. Differentiated treatment strategies are automatically matched according to the final safety level, prioritizing pile foundations requiring treatment according to their urgency. This makes the assessment report readily usable, establishing a step-by-step direct connection between the testing and assessment results and subsequent engineering decisions such as reinforcement, monitoring, and re-inspection. This solves the problem of the disconnect between testing conclusions and engineering applications in existing technologies.

[0051] This application also provides an electronic device. The electronic device may include one or more processors and one or more memories. The memories store computer-readable code, which, when executed by the one or more processors, can perform the non-destructive rapid diagnostic assessment system and method for the safety of existing building structures as described above.

[0052] The systems and methods according to the embodiments of this application can also be implemented using the architecture of the electronic device shown in this application. The electronic device may include a bus, one or more CPUs, ROM, RAM, a communication port connected to a network, input / output, a hard disk, etc. The storage device in the electronic device, such as a ROM or hard disk, may store the non-destructive rapid diagnostic assessment system and method for the safety of existing building structures provided in this application. Furthermore, the electronic device may also include a user interface. Of course, the architecture shown in this application is merely exemplary; when implementing different devices, one or more components of the electronic device shown in this application may be omitted according to actual needs.

[0053] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising a reference structure" does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0054] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A non-destructive rapid diagnostic assessment method for the structural safety of existing buildings, characterized in that, The method includes: For the evaluation of existing building configurations, a confidence target threshold, cost ceiling, and combination of detection methods are set for each evaluation stage. Upon entering the current assessment phase, the detection data of existing building foundation piles are collected and stored in the current phase dataset according to the combination of detection methods configured for the assessment phase. The current stage dataset is evaluated for statistical redundancy. The statistical redundancy index is calculated. Based on the data sufficiency threshold preset for the current evaluation stage, it is determined whether the statistical redundancy index meets the requirements. If not, the inner loop adaptive expansion strategy is executed and the statistical redundancy evaluation is performed again until the requirements are met or the cumulative detection cost reaches the cost limit preset for the current evaluation stage. The inner loop adaptive expansion strategy includes a first priority sub-step and a second priority sub-step. The execution process of the first priority sub-step is as follows: based on the spatial coordinates of the detected points in the current stage dataset in the overall distribution map of building foundation piles, identify the current spatial sampling blank area, and the current spatial sampling blank area is the area where the spatial distance between adjacent detected points exceeds a preset spatial threshold; select expanded detection points from the spatial sampling blank area based on preset point selection rules; according to the selected expanded detection points, use the same combination of detection methods as the current stage to perform supplementary detection; after each round of expanded detection is completed, the newly collected detection data is merged into the current stage dataset and the statistical redundancy index is recalculated. If, when the first priority sub-step is executed, the cumulative detection cost has reached more than 90% of the preset cost limit for the current stage, and the statistical redundancy index of the dataset in the current stage still does not meet the data sufficiency threshold requirement, or the reduction in the coefficient of variation after three consecutive rounds of executing the first priority sub-step is less than the preset improvement threshold, the second priority sub-step is triggered. The execution process of the second priority sub-step is as follows: identify the piles with the largest contribution of the coefficient of variation, the piles at the edge of the spatial distribution, and the piles with special stress conditions from the current stage dataset as key detection points; use a combination of detection methods with higher accuracy than the preset combination of detection methods in the current stage for the key detection points, re-collect detection data and supplement the corresponding pile records in the current stage dataset; and perform a statistical redundancy evaluation on the supplemented current stage dataset. For the current stage dataset after statistical redundancy assessment, calculate the security level assessment conclusion and the corresponding confidence index, and compare the confidence index with the preset confidence target threshold for the current stage: If the confidence index is greater than or equal to the confidence target threshold, the security level assessment conclusion is output and the assessment process is terminated; otherwise, it is determined whether the detection cost has reached the cost limit of the current assessment stage. If not, the step of performing statistical redundancy assessment on the dataset of the current stage is returned; if so, the next assessment stage is entered.

2. The non-destructive rapid diagnostic assessment method for the structural safety of existing buildings according to claim 1, characterized in that, The configured evaluation phase sequence includes a rapid screening phase, a routine detailed examination phase, and a high-precision verification phase; The rapid screening phase uses a combination of tapping method, rebound hammer and crack observation as detection methods, with a confidence target threshold of 70% to 80%. During the routine detailed investigation phase, the low-strain reflected wave method or the side-hole transmitted wave method are used as a combination of detection methods, with a confidence target threshold of 80% to 90%. In the high-precision verification stage, acoustic tomography, core drilling, or static load testing are used as a combination of detection methods, with a confidence target threshold of 90% to 99%.

3. The non-destructive rapid diagnostic assessment method for the structural safety of existing buildings according to claim 1, characterized in that, The step of evaluating the statistical redundancy of the current stage dataset and calculating the statistical redundancy index includes: Based on the current dataset, the coefficient of variation, sample adequacy index, and confidence interval width are calculated to form a statistical redundancy index. The data sufficiency thresholds include the coefficient of variation threshold, the sample size sufficiency index threshold, and the confidence interval width threshold preset for the current stage. The calculated statistical redundancy index is compared with the corresponding data sufficiency threshold. When the coefficient of variation is less than the coefficient of variation threshold, the sample size sufficiency index is greater than or equal to the sample size sufficiency index threshold, and the confidence interval width is less than the confidence interval width threshold, the dataset at the current stage is determined to meet the data sufficiency threshold requirements.

4. The non-destructive rapid diagnostic assessment method for the structural safety of existing buildings according to claim 1, characterized in that, The calculation of the security level assessment conclusion and corresponding confidence index for the current stage dataset after statistical redundancy assessment includes: Obtain the current stage dataset after statistical redundancy assessment, calculate the security level assessment conclusion for the current assessment stage, and classify the security level assessment conclusion into Au level, Bu level, Cu level and Du level; Based on the current stage dataset after statistical redundancy assessment, calculate data sufficiency factors, detection consistency factors, and model deterministic factors; The confidence index is obtained by weighting the factors of data sufficiency, detection consistency, and model determinism.

5. The non-destructive rapid diagnostic assessment method for the structural safety of existing buildings according to claim 1, characterized in that, The step of comparing the confidence index with the preset confidence target threshold for the current stage includes: When the confidence index is determined to be lower than the confidence target threshold and the cumulative detection cost has not reached the current stage cost limit, the main reasons for the insufficient confidence index should be further determined. If the portion of the gap between the confidence index and the confidence target threshold attributable to insufficient statistical redundancy is greater than the portion attributable to the inherent uncertainty of the evaluation model, then the main reason is determined to be insufficient data redundancy, and the process returns to the step of evaluating the statistical redundancy of the current stage dataset. If the portion attributable to the inherent uncertainty of the evaluation model is greater than the portion attributable to insufficient statistical redundancy, then the current stage evaluation conclusion is output as an intermediate conclusion with a confidence level, and a prompt message suggesting the next evaluation stage is generated.

6. The non-destructive rapid diagnostic assessment method for the structural safety of existing buildings according to claim 5, characterized in that, When entering the next evaluation stage, the next evaluation stage is taken as the new current stage. The new current stage dataset is reset to empty. The higher-precision detection methods configured in this stage are used to re-collect detection data to form the current stage dataset. The security level assessment conclusion, confidence index and statistical redundancy assessment results of the previous evaluation stage are inherited as prior information to guide the initial selection of detection points in this stage.

7. The non-destructive rapid diagnostic assessment method for the structural safety of existing buildings according to claim 1, characterized in that, The method further includes: Once the evaluation process is completed, a comprehensive evaluation report will be generated. The comprehensive assessment report includes safety level assessment conclusions, confidence index, confidence interval width, coefficient of variation, sample size adequacy index, and detailed cost breakdowns for each stage of testing. Based on the safety level assessment conclusions, corresponding treatment strategies are matched, and the piles requiring treatment are prioritized according to the severity of damage reflected by the testing data and then written into the comprehensive assessment report.

8. A non-destructive rapid diagnostic assessment system for the structural safety of existing buildings, characterized in that, The system includes: The information collection and stage configuration module is used to configure the evaluation stage sequence for the existing building to be evaluated. For each evaluation stage, the confidence target threshold, cost limit and detection method combination are set. When entering the current evaluation stage, the detection data of the existing building foundation piles are collected according to the detection method combination configured for the evaluation stage and then stored in the current stage dataset. The inner loop control module is used to evaluate the statistical redundancy of the dataset in the current stage, calculate the statistical redundancy index, and determine whether the statistical redundancy index meets the requirements based on the preset data sufficiency threshold in the current evaluation stage. If not, the inner loop adaptive expansion strategy is executed and the statistical redundancy evaluation is performed again until the requirements are met or the cumulative detection cost reaches the preset cost limit in the current evaluation stage. The inner loop adaptive expansion strategy includes a first priority sub-step and a second priority sub-step. The execution process of the first priority sub-step is as follows: based on the spatial coordinates of the detected points in the current stage dataset in the overall distribution map of building foundation piles, identify the current spatial sampling blank area, and the current spatial sampling blank area is the area where the spatial distance between adjacent detected points exceeds a preset spatial threshold; select expanded detection points from the spatial sampling blank area based on preset point selection rules; according to the selected expanded detection points, use the same combination of detection methods as the current stage to perform supplementary detection; after each round of expanded detection is completed, the newly collected detection data is merged into the current stage dataset and the statistical redundancy index is recalculated. If, when the first priority sub-step is executed, the cumulative detection cost has reached more than 90% of the preset cost limit for the current stage, and the statistical redundancy index of the dataset in the current stage still does not meet the data sufficiency threshold requirement, or the reduction in the coefficient of variation after three consecutive rounds of executing the first priority sub-step is less than the preset improvement threshold, the second priority sub-step is triggered. The execution process of the second priority sub-step is as follows: identify the piles with the largest contribution of the coefficient of variation, the piles at the edge of the spatial distribution, and the piles with special stress conditions from the current stage dataset as key detection points; use a combination of detection methods with higher accuracy than the preset combination of detection methods in the current stage for the key detection points, re-collect detection data and supplement the corresponding pile records in the current stage dataset; and perform a statistical redundancy evaluation on the supplemented current stage dataset. The outer loop control module is used to calculate the safety level assessment conclusion and the corresponding confidence index, and compare the confidence index with the preset confidence target threshold for the current stage. If the confidence index is greater than or equal to the confidence target threshold, the security level assessment conclusion is output and the assessment process is terminated; otherwise, it is determined whether the detection cost has reached the cost limit of the current assessment stage. If not, it returns to the inner loop control module; if so, it proceeds to the next assessment stage.

Images