Data processing method applied to strength detection of cast-in-place concrete foundation pile
Through data processing methods, the internal structure data and environmental factors of concrete poured foundation piles are obtained and analyzed, and the feature map is drawn and real-time monitoring is carried out, which solves the problems of low efficiency and insufficient accuracy of traditional detection methods, and realizes high-precision, real-time monitoring and intelligent prediction detection effect.
Patent Information
- Application Number
- CN202411817494.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-11
- Publication Date
- 2025-02-14
AI Technical Summary
Traditional concrete cast-injected pile inspection methods have problems such as inefficiency, high cost or insufficient results, which are difficult to meet the requirements of modern construction projects for inspection accuracy and efficiency.
Data processing methods are adopted, including obtaining the internal structure data of the pile body of concrete poured foundation piles, the concrete strength and environmental factors of the pile body, drawing a feature map, and real-time monitoring of the foundation pile status through real-time threshold monitoring and intelligent prediction model to achieve real-time monitoring of the foundation pile status and Predictions of potential future problems.
It improves detection accuracy, enhances real-time monitoring capabilities, realizes intelligent prediction and early warning, optimizes resource allocation, and improves project quality and safety.
Smart Images
Figure CN119434350A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of concrete detection, in particular to a data processing method applied to strength detection of concrete cast-in-place foundation piles. Background Art
[0002] CN112285212A "A system and method for detecting concrete strength based on ultrasonic rebound method" includes a calibration instrument and a measuring instrument. The calibration instrument is used to calibrate the detection points in the target measurement area of the concrete; the measuring instrument is used to be set at the detection points to collect corresponding ultrasonic detection data and rebound detection data, and calculate the concrete strength based on the acquired detection data.
[0003] CN112147228A "Method for establishing a strength curve for detecting concrete strength by a comprehensive method of rebound ultrasonic angle measurement" includes arranging ultrasonic angle measurement areas on two adjacent sides of a concrete column component, measuring the rebound value of a single-side measurement area in one of the measurement areas, and calculating the single-side measurement area rebound representative value; measuring the ultrasonic angle sound time and the distance between the geometric center of the ultrasonic receiving and transmitting transducers and the corner in the corresponding rebound measurement area, calculating the distance between the geometric centers of the transducers, and then calculating the representative value of the angle sound velocity of the measurement area; drilling core samples in the corresponding rebound and ultrasonic angle measurement areas, and testing to obtain the compressive strength of the core sample concrete;
[0004] In modern construction projects, cast-in-place concrete piles are widely used as an important form of deep foundation. In order to ensure its bearing capacity and long-term stability, it must be strictly quality controlled and tested. Although traditional testing methods such as core sampling, low-strain reflection wave method, and acoustic transmission method can provide certain information, they have problems such as low efficiency, high cost, or inaccurate results. With the development of information technology, it has become possible to use advanced data processing technology and algorithms to optimize these traditional methods. Improving the accuracy and efficiency of detection is an issue that we urgently need to solve. Summary of the invention
[0005] In order to solve the above technical problems, the object of the present invention is to provide a data processing method for concrete cast-in-place pile strength detection, comprising the following steps:
[0006] Step s1: Obtain the internal structure data of each concrete cast-in-place pile, the concrete strength of the pile body, and environmental factors, mark the collection time, and set the collection cycle;
[0007] Step s2: obtaining key features of each concrete cast-in-place foundation pile and the pile body depth corresponding to the key features, and drawing a feature map according to the key features and the pile body depth corresponding to the key features;
[0008] Step s3: obtaining threshold intervals corresponding to key features according to pile body concrete strength and environmental factors, performing real-time threshold monitoring on feature maps of each cast-in-place concrete pile, and classifying each cast-in-place concrete pile as a defective cast-in-place concrete pile or a qualified cast-in-place concrete pile;
[0009] Step s4: Perform data processing on the key features, pile body concrete strength and environmental factors within the historical collection period of each concrete cast-in-place pile foundation, build a foundation pile intelligent estimation model based on the data processing results, and estimate the qualified concrete cast-in-place pile foundation.
[0010] Furthermore, the process of obtaining the key features of each concrete cast-in-place pile and the pile body depth corresponding to the key features includes:
[0011] The internal structure data of the pile body are preprocessed, and key features are extracted from the preprocessed internal structure data of the pile body to obtain the key features and the pile body depth corresponding to the key features.
[0012] Furthermore, the process of drawing the feature map includes:
[0013] A rectangular coordinate system is established with the pile body depth as the horizontal axis and the key feature as the vertical axis. A feature map is drawn in the rectangular coordinate system according to the key features and the pile body depths corresponding to the key features. Each key feature corresponding to the pile body depth constitutes a unique coordinate point in the rectangular coordinate system.
[0014] Furthermore, the process of obtaining the threshold interval corresponding to the key feature according to the pile concrete strength and environmental factors includes:
[0015] Based on the simulation learning concept, simulation tests are performed on qualified cast-in-place concrete piles with different pile body concrete strengths under different environmental factors, and simulated pile body internal structure data of qualified cast-in-place concrete piles with different pile body concrete strengths under different environmental factors are obtained. Data preprocessing and key feature extraction are performed on the simulated pile body internal structure data to obtain key features corresponding to the simulated pile body internal structure data. According to the threshold interval of the key features, the threshold interval corresponding to the key features under different pile body concrete strengths and environmental factors is obtained, and a threshold interval lookup table is constructed according to the threshold interval corresponding to the key features under different pile body concrete strengths and environmental factors.
[0016] According to the pile body concrete strength, environmental factors and threshold interval lookup table in the current collection period, the threshold interval corresponding to the key feature under the pile body concrete strength and environmental factors in the current collection period is obtained.
[0017] Furthermore, the process of real-time monitoring the threshold of the characteristic spectrum of each concrete cast-in-place pile includes:
[0018] Compare the key features of each coordinate point in the feature map with the corresponding threshold interval. If the key features of the coordinate points in the feature map are not in the corresponding threshold interval, the concrete cast-in-place piles are marked as defective concrete cast-in-place piles. The coordinate points whose key features in the feature map are not in the corresponding threshold interval are marked in red and recorded as abnormal coordinate points. Generate abnormal repair measures based on the key features of the abnormal coordinate points.
[0019] If the key features of each coordinate point in the feature map are all within the corresponding threshold range, the cast-in-place concrete piles are marked as qualified cast-in-place concrete piles.
[0020] Furthermore, the process of generating anomaly repair measures according to the key features of the abnormal coordinate points includes:
[0021] An abnormality repair measure database is constructed, wherein the abnormality repair measure database includes abnormality repair measures corresponding to different key features, the key features of the abnormal coordinate points are input into the abnormality repair measure database for similarity retrieval, and the feature similarities between different key features in the abnormality repair measure database and the key features of the abnormal coordinate points are obtained. Key features whose feature similarity is greater than a preset feature similarity threshold are screened out in the abnormality repair measure database, and the abnormality repair measures corresponding to the key features are extracted.
[0022] Furthermore, the process of data processing of key features, pile concrete strength and environmental factors during the historical collection period of each concrete cast-in-place foundation pile includes:
[0023] Constructing a feature database, wherein the feature database is used to store the key features of the concrete cast-in-place piles during the collection period, the time series corresponding to the pile body concrete strength and environmental factors;
[0024] Extract the key features of each concrete pile in several historical collection periods, the time series corresponding to the pile concrete strength and environmental factors from the feature database, and mark the historical collection period where the concrete pile is located;
[0025] If the concrete cast-in-place piles are marked as qualified concrete cast-in-place piles in the historical collection period, the historical collection period in which the concrete cast-in-place piles are located is marked as a qualified historical collection period; if the concrete cast-in-place piles are marked as defective concrete cast-in-place piles in the historical collection period, the historical collection period in which the concrete cast-in-place piles are located is marked as a defective historical collection period;
[0026] The moment when the concrete cast-in-place piles are marked as defective concrete cast-in-place piles within the defect history collection period is marked as the critical moment, the time series corresponding to the key features, pile body concrete strength and environmental factors from the start time of the defect history collection period to the critical moment is marked as the critical feature set sequence, and the time series corresponding to the key features, pile body concrete strength and environmental factors from the start time of the qualified history collection period to the end time of the qualified history collection period is marked as the qualified feature set sequence;
[0027] Data processing is performed on the critical feature set sequence and the qualified feature set sequence to obtain a critical evaluation index corresponding to the critical feature set sequence and a qualified evaluation index corresponding to the qualified feature set sequence.
[0028] Furthermore, the intelligent prediction model for foundation piles is constructed, and the process of estimating qualified concrete cast-in-place foundation piles includes:
[0029] Constructing a foundation pile intelligent prediction model based on deep learning, taking the critical evaluation indicators corresponding to the critical feature set sequences in several defect history acquisition cycles and the qualified evaluation indicators corresponding to the qualified feature set sequences in several qualified history acquisition cycles as training sets and test sets, inputting the training sets into the foundation pile intelligent prediction model for training until the loss function training is stable, and saving the model parameters, testing the foundation pile intelligent prediction model through the test set until it meets the preset requirements, and outputting the foundation pile intelligent prediction model;
[0030] Data processing is performed on the time series corresponding to the key features of qualified cast-in-place concrete piles, the concrete strength of the pile body and the environmental factors in the current collection period, and qualified evaluation indicators corresponding to the qualified feature set sequence are obtained. The qualified evaluation indicators are input into the intelligent prediction model for piles, and the prediction results of qualified cast-in-place concrete piles are output according to the intelligent prediction model for piles. When the prediction result is an estimated defective cast-in-place concrete pile, an early warning signal for qualified cast-in-place concrete piles is generated.
[0031] Compared with the prior art, the present invention has the following beneficial effects:
[0032] 1. Improve detection accuracy: By obtaining the internal structure data of each concrete pile, the concrete strength of the pile body and environmental factors, and combining time marking and periodic collection, the actual condition of the pile can be more comprehensively reflected, avoiding the errors that may be caused by measurement at a single time point, thereby improving the accuracy of the detection results.
[0033] 2. Enhanced real-time monitoring capability: The present invention can draw characteristic maps based on key features and their corresponding pile depths, and monitor these maps in real time based on thresholds. This ensures that potential problems, such as deformation of the reinforcement cage or abnormal spacing between main bars, are discovered immediately during the construction process, and corrective measures are taken in a timely manner to prevent the problem from deteriorating, thereby ensuring the quality and safety of the project.
[0034] 3. Realize intelligent prediction and early warning: The intelligent prediction model of pile foundation built based on historical data can not only predict possible problems in the future, but also send out early warning signals when it is detected that qualified pile foundations may turn into defective pile foundations. This forward-looking management method helps to plan maintenance work in advance, reduce the probability of accidents, and ensure the safety and reliability of long-term operations.
[0035] 4. Optimize resource allocation: By processing the data of critical feature set sequences and qualified feature set sequences, the present invention can identify which factors are most likely to cause the quality of pile foundations to deteriorate, and then guide the construction unit to adjust the construction parameters or improve the process flow in a targeted manner, reasonably allocate resources, reduce costs, and improve the overall project quality. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] Figure 1 This is a schematic diagram of a data processing method for strength testing of cast-in-place concrete piles according to an embodiment of the present application. DETAILED DESCRIPTION
[0037] The following is a clear and complete description of the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.
[0038] like Figure 1 As shown, the data processing method applied to the strength detection of cast-in-place concrete piles comprises the following steps:
[0039] Step s1: Obtain the internal structure data of each concrete cast-in-place foundation pile, the concrete strength of the pile body (the core sampling method is used to obtain the compressive strength value of the actual core sample; for piles that have not yet reached the specified age, the ultrasonic rebound comprehensive method or other non-destructive testing methods can also be used to estimate their early strength), environmental factors, mark the collection time, and set the collection cycle;
[0040] The internal structure data of the pile body include stress wave propagation time (the time difference between the transmitting point and the receiving point is measured by the low strain reflection wave method or the acoustic wave transmission method to infer the wave velocity of the pile body), amplitude attenuation (the intensity change of the signal after passing through different media is recorded by the low strain reflection wave method or the acoustic wave transmission method to evaluate the uniformity of the material and whether there are defects) and frequency response (the waveform characteristics received are analyzed by the low strain reflection wave method or the acoustic wave transmission method, such as the main frequency, the resonance peak spacing, etc., to determine the integrity of the pile body and its internal structure status); the environmental factors include temperature, humidity, rainfall and other factors; integrating data from different detection methods (low strain reflection wave, acoustic wave transmission, core sampling), and combining the advantages of multiple detection methods, can more comprehensively reflect the state of the pile body and improve the accuracy of defect identification;
[0041] Step s2: obtaining key features of each concrete cast-in-place foundation pile and the pile body depth corresponding to the key features, and drawing a feature map according to the key features and the pile body depth corresponding to the key features;
[0042] Step s3: obtaining threshold intervals corresponding to key features according to pile body concrete strength and environmental factors, performing real-time threshold monitoring on feature maps of each cast-in-place concrete pile, and classifying each cast-in-place concrete pile as a defective cast-in-place concrete pile or a qualified cast-in-place concrete pile;
[0043] Step s4: Perform data processing on the key features, pile body concrete strength and environmental factors within the historical collection period of each concrete cast-in-place pile foundation, build a foundation pile intelligent estimation model based on the data processing results, and estimate the qualified concrete cast-in-place pile foundation.
[0044] It should be further explained that, in the specific implementation process, the process of obtaining the key features of each concrete cast-in-place foundation pile and the pile body depth corresponding to the key features includes:
[0045] The internal structure data of the pile body is preprocessed. The high-frequency noise and other non-correlated signals introduced in the measurement process are eliminated by Gaussian filtering method to remove noise interference and outliers. The data points far away from the group distribution are identified and marked as outliers by statistical tools such as box plots, Z-score or IQR (interquartile range) rules, and the outliers are removed. The wavelet transform technology is used to extract the key features of the internal structure data of the pile body after data preprocessing to obtain the key features reflecting the internal structure state of the pile body and the pile body depth corresponding to the key features. The key features include the first arrival time, the centroid frequency and its variance. The pile body depth represents the distance from the natural ground (or a reference plane) to the bottom of the pile. The pile body depth is crucial to understanding the internal structure of the pile body because it directly reflects the changes in material properties at different depths.
[0046] It should be further explained that, in the specific implementation process, the process of drawing the feature map includes:
[0047] A rectangular coordinate system is established with the pile body depth as the horizontal axis and the key feature as the vertical axis. A feature map is drawn in the rectangular coordinate system according to the key features and the pile body depths corresponding to the key features. Each key feature corresponding to the pile body depth constitutes a unique coordinate point in the rectangular coordinate system.
[0048] It should be further explained that, in the specific implementation process, the process of obtaining the threshold range corresponding to the key characteristics according to the pile concrete strength and environmental factors includes:
[0049] Based on the simulation learning concept, simulation tests are performed on qualified cast-in-place concrete piles with different pile body concrete strengths under different environmental factors, and simulated pile body internal structure data of qualified cast-in-place concrete piles with different pile body concrete strengths under different environmental factors are obtained. Data preprocessing and key feature extraction are performed on the simulated pile body internal structure data to obtain key features corresponding to the simulated pile body internal structure data. According to the threshold interval of the key features, the threshold interval corresponding to the key features under different pile body concrete strengths and environmental factors is obtained, and a threshold interval lookup table is constructed according to the threshold interval corresponding to the key features under different pile body concrete strengths and environmental factors.
[0050] According to the pile body concrete strength, environmental factors and threshold interval lookup table in the current collection period, the threshold interval corresponding to the key feature under the pile body concrete strength and environmental factors in the current collection period is obtained.
[0051] It should be further explained that, in the specific implementation process, the process of real-time monitoring of the characteristic spectrum of each concrete cast-in-place pile includes:
[0052] Compare the key features of each coordinate point in the feature map with the corresponding threshold interval. If the key features of the coordinate points in the feature map are not in the corresponding threshold interval, the concrete cast-in-place piles are marked as defective concrete cast-in-place piles. The coordinate points whose key features in the feature map are not in the corresponding threshold interval are marked in red and recorded as abnormal coordinate points. Generate abnormal repair measures based on the key features of the abnormal coordinate points.
[0053] If the key features of each coordinate point in the feature map are all within the corresponding threshold range, the cast-in-place concrete piles are marked as qualified cast-in-place concrete piles.
[0054] It should be further explained that, in the specific implementation process, the process of generating anomaly repair measures according to the key features of the abnormal coordinate points includes:
[0055] An abnormal repair measure database is constructed, wherein the abnormal repair measure database includes abnormal repair measures corresponding to different key features (such as broken pile repair, pile body crack repair, necking and mud layer treatment, etc.), and the key features of the abnormal coordinate points are input into the abnormal repair measure database for similarity retrieval, and the feature similarities between different key features in the abnormal repair measure database and the key features of the abnormal coordinate points are obtained. The key features whose feature similarity is greater than a preset feature similarity threshold are screened out in the abnormal repair measure database, and the abnormal repair measures corresponding to the key features are extracted.
[0056] It should be further explained that, in the specific implementation process, the process of data processing of key features, pile concrete strength and environmental factors in the historical collection period of each concrete cast-in-place foundation pile includes:
[0057] Constructing a feature database, wherein the feature database is used to store the key features of the concrete cast-in-place piles during the collection period, the time series corresponding to the pile body concrete strength and environmental factors;
[0058] Extract the key features of each concrete pile in several historical collection periods, the time series corresponding to the pile concrete strength and environmental factors from the feature database, and mark the historical collection period where the concrete pile is located;
[0059] If the concrete cast-in-place piles are marked as qualified concrete cast-in-place piles in the historical collection period, the historical collection period in which the concrete cast-in-place piles are located is marked as a qualified historical collection period; if the concrete cast-in-place piles are marked as defective concrete cast-in-place piles in the historical collection period, the historical collection period in which the concrete cast-in-place piles are located is marked as a defective historical collection period;
[0060] The moment when the concrete cast-in-place piles are marked as defective concrete cast-in-place piles within the defect history collection period is marked as the critical moment, the time series corresponding to the key features, pile body concrete strength and environmental factors from the start time of the defect history collection period to the critical moment is marked as the critical feature set sequence, and the time series corresponding to the key features, pile body concrete strength and environmental factors from the start time of the qualified history collection period to the end time of the qualified history collection period is marked as the qualified feature set sequence;
[0061] Data processing is performed on the critical feature set sequence and the qualified feature set sequence to obtain critical evaluation indicators corresponding to the critical feature set sequence and qualified evaluation indicators corresponding to the qualified feature set sequence, wherein the critical evaluation indicators and qualified evaluation indicators include the average standard deviation of the key features, the average standard deviation of the pile body concrete strength, the average standard deviation of the environmental factors, the Pearson correlation coefficient between the key features and the pile body concrete strength, the Pearson correlation coefficient between the key features and the environmental factors, and the Pearson correlation coefficient between the pile body concrete strength and the environmental factors.
[0062] It should be further explained that, in the specific implementation process, the process of constructing the intelligent prediction model for foundation piles and estimating the qualified concrete cast-in-place foundation piles includes:
[0063] Constructing a foundation pile intelligent prediction model based on deep learning, taking the critical evaluation indicators corresponding to the critical feature set sequences in several defect history acquisition cycles and the qualified evaluation indicators corresponding to the qualified feature set sequences in several qualified history acquisition cycles as training sets and test sets, inputting the training sets into the foundation pile intelligent prediction model for training until the loss function training is stable, and saving the model parameters, testing the foundation pile intelligent prediction model through the test set until it meets the preset requirements, and outputting the foundation pile intelligent prediction model;
[0064] Building a deep learning-based intelligent prediction model for foundation piles is a complex process that involves multiple steps such as model selection, training, verification, and testing. The following is a detailed supplementary explanation of this process:
[0065] Taking into account the characteristics of the pile detection task, the present invention selects a convolutional neural network (CNN) suitable for time series analysis as a deep learning architecture. After determining the model architecture, the loss function is defined next. In the present invention, since the goal is to distinguish between qualified and defective piles, the binary cross-entropy loss is selected as the optimization target, and then the prepared training set is input into the selected deep learning model to start training. During the training process, the weights are continuously updated by the back propagation algorithm, so that the loss function gradually decreases until a stable state is reached. During this period, the early stopping method and other techniques are used to avoid overfitting. In addition to the basic training process, various parameters of the model are tuned by grid search, and the parameters include learning rate, batch size, regularization coefficient, etc.
[0066] After the model training is completed and the parameters are adjusted, the final evaluation is performed through the test set to obtain the evaluation results of the model, which include classification indicators such as accuracy, recall, F1 score, etc. Based on the evaluation results on the test set, it is judged whether the model has reached the expected standards. If the requirements are met, the model parameters are saved and ready for deployment; if not ideal, it is necessary to return to a previous stage to re-examine issues such as data quality, model structure or training strategy.
[0067] Data processing is performed on the time series corresponding to the key features of qualified cast-in-place concrete piles, the concrete strength of the pile body and the environmental factors in the current collection period, and qualified evaluation indicators corresponding to the qualified feature set sequence are obtained. The qualified evaluation indicators are input into the intelligent prediction model for piles, and the prediction results of qualified cast-in-place concrete piles are output according to the intelligent prediction model for piles. The prediction results include estimated qualified cast-in-place concrete piles and estimated defective cast-in-place concrete piles. When the prediction result is an estimated defective cast-in-place concrete pile, an early warning signal for qualified cast-in-place concrete piles is generated, and professional engineers and technicians are organized to go to the construction site to conduct a detailed physical inspection of the piles marked as "estimated defects", which includes but is not limited to the use of non-destructive testing technology (such as ultrasonic testing, radar scanning, etc.) to confirm whether there are actual defects, and to evaluate the specific nature and severity of these defects.
[0068] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.
Claims
1. A data processing method for testing the strength of concrete piles, characterized in that: The following steps are involved: Step s1: Obtain the internal structure data of each concrete cast-in-place pile, the concrete strength of the pile body, and environmental factors, mark the collection time, and set the collection cycle; Step s2: Obtain key features of each concrete cast-in-place foundation pile and the pile body depth corresponding to the key features, and draw a feature map according to the key features and the pile body depth corresponding to the key features; Step s3: obtaining threshold intervals corresponding to key features according to pile body concrete strength and environmental factors, performing real-time threshold monitoring on feature maps of each cast-in-place concrete pile, and classifying each cast-in-place concrete pile as a defective cast-in-place concrete pile or a qualified cast-in-place concrete pile; Step s4: Perform data processing on the key features, pile body concrete strength and environmental factors within the historical collection period of each concrete cast-in-place pile foundation, build a foundation pile intelligent estimation model based on the data processing results, and estimate the qualified concrete cast-in-place pile foundation.
2. The data processing method for concrete cast-in-place pile strength detection according to claim 1 is characterized in that: The process of obtaining the key features of each concrete cast-in-place pile and the pile depth corresponding to the key features includes: The internal structure data of the pile body is preprocessed, and key features are extracted from the preprocessed internal structure data of the pile body to obtain the key features and the pile body depth corresponding to the key features.
3. The data processing method for concrete cast-in-place pile strength detection according to claim 2 is characterized in that: The process of drawing feature maps includes: A rectangular coordinate system is established with the pile body depth as the horizontal axis and the key feature as the vertical axis. A feature map is drawn in the rectangular coordinate system according to the key features and the pile body depths corresponding to the key features. Each key feature corresponding to the pile body depth constitutes a unique coordinate point in the rectangular coordinate system.
4. The data processing method for concrete cast-in-place pile strength detection according to claim 3 is characterized in that: The process of obtaining the threshold interval corresponding to the key characteristics according to the pile concrete strength and environmental factors includes: Based on the simulation learning concept, simulation tests are performed on qualified cast-in-place concrete piles with different pile body concrete strengths under different environmental factors, and simulated pile body internal structure data of qualified cast-in-place concrete piles with different pile body concrete strengths under different environmental factors are obtained. Data preprocessing and key feature extraction are performed on the simulated pile body internal structure data to obtain key features corresponding to the simulated pile body internal structure data. According to the threshold interval of the key features, the threshold interval corresponding to the key features under different pile body concrete strengths and environmental factors is obtained, and a threshold interval lookup table is constructed according to the threshold interval corresponding to the key features under different pile body concrete strengths and environmental factors. According to the pile body concrete strength, environmental factors and threshold interval lookup table in the current collection period, the threshold interval corresponding to the key feature under the pile body concrete strength and environmental factors in the current collection period is obtained.
5. The data processing method for concrete cast-in-place pile strength detection according to claim 4 is characterized in that: The process of real-time monitoring the threshold of the characteristic spectrum of each concrete pile includes: Compare the key features of each coordinate point in the feature map with the corresponding threshold interval. If the key features of the coordinate points in the feature map are not in the corresponding threshold interval, the concrete cast-in-place piles are marked as defective concrete cast-in-place piles. The coordinate points whose key features in the feature map are not in the corresponding threshold interval are marked in red and recorded as abnormal coordinate points. Generate abnormal repair measures based on the key features of the abnormal coordinate points. If the key features of each coordinate point in the feature map are all within the corresponding threshold range, the cast-in-place concrete piles are marked as qualified cast-in-place concrete piles.
6. The data processing method for concrete cast-in-place pile strength detection according to claim 5 is characterized in that: The process of generating anomaly repair measures based on the key features of the abnormal coordinate points includes: An abnormality repair measure database is constructed, wherein the abnormality repair measure database includes abnormality repair measures corresponding to different key features, the key features of the abnormal coordinate points are input into the abnormality repair measure database for similarity retrieval, and the feature similarities between different key features in the abnormality repair measure database and the key features of the abnormal coordinate points are obtained. Key features whose feature similarity is greater than a preset feature similarity threshold are screened out in the abnormality repair measure database, and the abnormality repair measures corresponding to the key features are extracted.
7. The data processing method for concrete cast-in-place pile strength detection according to claim 6 is characterized in that: The process of data processing for key features, pile concrete strength and environmental factors during the historical collection period of each cast-in-place concrete foundation pile includes: Constructing a feature database, wherein the feature database is used to store the key features of the concrete cast-in-place piles during the collection period, the time series corresponding to the pile body concrete strength and environmental factors; Extract the key features of each concrete pile in several historical collection periods, the time series corresponding to the pile concrete strength and environmental factors from the feature database, and mark the historical collection period where the concrete pile is located; If the concrete cast-in-place piles are marked as qualified concrete cast-in-place piles in the historical collection period, the historical collection period in which the concrete cast-in-place piles are located is marked as a qualified historical collection period; if the concrete cast-in-place piles are marked as defective concrete cast-in-place piles in the historical collection period, the historical collection period in which the concrete cast-in-place piles are located is marked as a defective historical collection period; The moment when the concrete cast-in-place piles are marked as defective concrete cast-in-place piles within the defect history collection period is marked as the critical moment, the time series corresponding to the key features, pile body concrete strength and environmental factors from the start time of the defect history collection period to the critical moment is marked as the critical feature set sequence, and the time series corresponding to the key features, pile body concrete strength and environmental factors from the start time of the qualified history collection period to the end time of the qualified history collection period is marked as the qualified feature set sequence; Data processing is performed on the critical feature set sequence and the qualified feature set sequence to obtain a critical evaluation index corresponding to the critical feature set sequence and a qualified evaluation index corresponding to the qualified feature set sequence.
8. The data processing method for concrete cast-in-place pile strength detection according to claim 7 is characterized in that: The process of constructing an intelligent prediction model for foundation piles and estimating qualified concrete cast-in-place foundation piles includes: Constructing a foundation pile intelligent prediction model based on deep learning, taking the critical evaluation indicators corresponding to the critical feature set sequences in several defect history acquisition cycles and the qualified evaluation indicators corresponding to the qualified feature set sequences in several qualified history acquisition cycles as training sets and test sets, inputting the training sets into the foundation pile intelligent prediction model for training until the loss function training is stable, and saving the model parameters, testing the foundation pile intelligent prediction model through the test set until it meets the preset requirements, and outputting the foundation pile intelligent prediction model; Data processing is performed on the time series corresponding to the key features of qualified cast-in-place concrete piles, the concrete strength of the pile body and the environmental factors in the current collection period, and qualified evaluation indicators corresponding to the qualified feature set sequence are obtained. The qualified evaluation indicators are input into the intelligent prediction model for piles, and the prediction results of qualified cast-in-place concrete piles are output according to the intelligent prediction model for piles. When the prediction result is an estimated defective cast-in-place concrete pile, an early warning signal for qualified cast-in-place concrete piles is generated.
Citation Information
Patent Citations
Method for establishing concrete strength measurement curve for detecting concrete strength by rebound-ultrasonic angle measurement comprehensive method
CN112147228A
System and method for detecting concrete strength based on ultrasonic springback method
CN112285212A