Device for testing bearing capacity of pile end of prestressed pipe pile
By identifying and clustering stress loading rates in real time, and combining this with a random forest model to predict the lag deformation of bearing capacity, the problem of rate mismatch in the bearing capacity test of prestressed pipe pile ends was solved, achieving accurate testing and safety assurance.
Patent Information
- Application Number
- CN202511862103.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-11
- Publication Date
- 2026-01-09
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In existing technologies for testing the bearing capacity of prestressed concrete pipe piles, the testing rate is not matched with the deformation rate of the weak underlying layer, resulting in short-term test results that cannot accurately reflect the long-term bearing capacity of the pile end, affecting project safety and even causing safety accidents.
A piezoelectric pressure sensor and a GPS timing module are used to obtain the real-time stress loading rate. The overstress loading rate range is identified by clustering algorithm and random forest model. The bearing capacity lag deformation is predicted by combining historical misjudgment cases, and the lag deformation is verified and revised by correction coefficient.
Accurately identify risk zones for misjudgment, accurately predict and revise bearing capacity lag deformation, avoid safety accidents, improve test accuracy and reliability, and ensure the long-term safety of pile foundation projects.
Smart Images
Figure CN121301983A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of prestressed pipe pile end testing technology, specifically a prestressed pipe pile end bearing capacity testing device. Background Technology
[0002] In prestressed concrete pipe pile foundation engineering, the pile tip bearing capacity test is a key step in assessing the long-term safety of the pile foundation. The static load test method is usually used to observe the relationship between the pile top settlement and the pile tip reaction force by loading in stages, and to determine whether the pile tip resistance meets the design requirements. However, existing technologies have a significant problem in actual testing: the testing rate does not match the deformation rate of the weak underlying layer. This results in short-term test results failing to accurately reflect the long-term bearing capacity of the pile tip, seriously affecting engineering safety.
[0003] According to standards (such as JGJ106), the static load test loading rate should be 0.5~1.0 MPa / min, aiming to simulate the slow application of actual engineering loads, allowing the soil (especially the weak underlying layer) sufficient time to complete consolidation deformation. However, in engineering practice, due to factors such as construction period pressure, the accuracy of testing equipment control, and operator violations, the testing rate is often illegally increased to over 1.5 MPa / min.
[0004] At this time, the initial settlement of the pile top is significantly smaller than expected, which may be mistakenly judged as the pile end bearing capacity meeting the standard. However, in actual engineering, as the load continues to act for a long time, the weak underlying layer gradually completes consolidation and creep, resulting in excessive settlement of the pile end, and even causing shear failure at the interface between the bearing layer and the underlying layer, leading to safety accidents such as cracking of the bridge wall and uneven settlement of the road foundation.
[0005] Therefore, the present invention provides a device for testing the bearing capacity of prestressed pipe pile ends. Summary of the Invention
[0006] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.
[0007] The technical solution adopted by this invention to solve its technical problem is: One of the objectives of this invention is to provide a device for testing the end bearing capacity of prestressed concrete pipe piles, comprising: Stress misjudgment risk interval identification module: acquires real-time stress loading rate to identify the number of times the stress loading rate exceeds the limit, performs cluster analysis on the number of times the stress loading rate exceeds the limit through clustering algorithm to identify the stress loading rate interval, and combines historical misjudgment cases to identify the current stress misjudgment risk interval by constructing a random forest model; Load-bearing capacity lag deformation output module: For the current stress misjudgment risk range, combined with the load-bearing capacity lag deformation corresponding to the stress misjudgment risk range in historical misjudgment cases, predict the load-bearing capacity lag deformation of the current stress misjudgment risk range; The load-bearing capacity lag deformation revision module: performs a matching verification between the load-bearing capacity lag deformation in the current stress misjudgment risk range and the load-bearing capacity lag deformation corresponding to the stress misjudgment risk range in historical misjudgment cases. If they match, it determines the load-bearing capacity lag correction coefficient by combining the load-bearing capacity lag deformation corresponding to the stress misjudgment risk range in historical misjudgment cases, verifies the load-bearing capacity lag correction coefficient, outputs the comprehensive load-bearing capacity lag correction coefficient, and uses the comprehensive load-bearing capacity lag correction coefficient to revise the load-bearing capacity lag deformation in the current stress misjudgment risk range.
[0008] As a further improvement of the present invention, the specific process of obtaining the real-time stress loading rate is as follows: Pressure data is collected using a piezoelectric pressure sensor and a GPS timing module is used to add precise timestamps to the pressure data. An embedded processor performs differential calculations on two consecutive pressure values and their corresponding times to obtain the pressure change per unit time, i.e., the real-time stress loading rate.
[0009] As a further improvement of the present invention, the specific process of identifying the number of overstress loading rates is as follows: Using the timestamp of the real-time stress loading rate as the X-axis and the real-time stress loading rate as the Y-axis, all real-time stress loading rates are plotted on a two-dimensional plane to obtain the real-time stress loading rate curve. In the real-time stress loading rate curve, with 1.0 MPa / min as the first reference point, a straight line parallel to the X-axis is drawn on the first reference point to obtain the first reference line; In the real-time stress loading rate curve, the real-time stress loading rate that exceeds the first reference curve is recorded as the overstress loading rate; all real-time stress loading rates that exceed the first reference curve are counted to identify the number of overstress loading rates.
[0010] As a further improvement of the present invention, the specific process of identifying the overstress loading rate range is as follows: Obtain the timestamp and real-time stress loading rate corresponding to each overstress loading rate, and represent them as data points; Each data point is represented as , Let i be the timestamp corresponding to the i-th overstress loading rate. This is the real-time stress loading rate corresponding to the i-th overstress loading rate; In the real-time stress loading rate curve, the overstress loading rate is marked as event 1, event 2... event n in ascending order of timestamps, where n is the total number of overstress loading rate events; The DBSCAN algorithm is used to cluster data points and output the clusters corresponding to the overstress loading rate intervals; the corresponding clusters are the overstress loading rate intervals.
[0011] As a further improvement of the present invention, the specific process of identifying the current stress misjudgment risk range is as follows: Obtain the clusters output by the clustering algorithm corresponding to the overstress loading rate range, calculate the minimum and maximum range of loading rates within the cluster, denoted as the loading rate difference within the cluster, and calculate the number of overstress loading rate events. Obtain the clusters output by the clustering algorithm for each misjudgment in historical misjudgment cases, and calculate the number of loading rate differences and overstress loading rate events within the clusters; Set up positive and negative sample cases, extract the intra-cluster loading rate difference and the number of overstress loading rate events corresponding to each group of positive and negative samples to form a training set and divide it into a sub-training set and a validation set, and train the random forest model. Input the intra-cluster loading rate difference and the number of overstress loading rate events corresponding to the overstress loading rate interval into the trained random forest model to identify the stress misjudgment risk interval.
[0012] As a further improvement of the present invention, the specific process of predicting the bearing capacity hysteresis deformation in the current stress misjudgment risk range is as follows: Obtain the characteristic parameters of the stress misjudgment risk range from historical misjudgment cases, including: Independent variable characteristics include: intra-cluster loading rate difference, number of overstress loading rate events, duration of stress misjudgment risk interval, and soil parameters, including: compression modulus and moisture content; dependent variable characteristics include: measured value of bearing capacity hysteresis deformation. Historical misjudgment cases are divided into training and validation sets; the independent variable features are used as inputs to the multiple linear regression model, and the dependent variable features are used as outputs. When the determination coefficient of the multiple linear regression model is greater than 0.85, the multiple linear regression model is completed. For the current stress misjudgment risk range, the independent variable features of the current stress misjudgment risk range are extracted and used to construct a multiple linear regression model to predict the bearing capacity hysteresis deformation of the current stress misjudgment risk range.
[0013] As a further improvement of the present invention, the specific process for verifying the matching of the bearing capacity lag deformation in the current stress misjudgment risk range with the bearing capacity lag deformation in the stress misjudgment risk range in historical misjudgment cases is as follows: Historical misjudgment cases that are consistent with the soil type parameters in the current stress misjudgment risk range are selected to form a historical reference case set; For cases in the current stress misjudgment risk range and the historical reference case set, extract core feature parameters and process them uniformly; Key features include: intra-cluster loading rate difference, number of overstress loading rate events, and duration of stress misjudgment risk interval; The core feature parameters of the current stress misjudgment risk range form the current feature vector; the core feature parameters of the cases in the historical reference case set form the historical feature vector; Using the Euclidean distance formula, the spatial distance between the current feature vector and each historical feature vector is calculated and recorded as the feature similarity. If the feature similarity is less than or equal to the feature similarity threshold, it means that the bearing capacity lag deformation output by the multiple linear regression model matches the historical misjudgment cases.
[0014] As a further improvement of the present invention, the specific process of determining the bearing capacity lag correction coefficient by combining the bearing capacity lag deformation corresponding to the stress misjudgment risk interval in historical misjudgment cases is as follows: Historical misjudgment cases that match the current stress misjudgment risk range are selected to form a historical matching case set; Extract the two sets of core data for each case from the historical matching case set, including: measured bearing capacity lag deformation and bearing capacity lag deformation; Calculate the bearing capacity lag correction coefficient for each case in the historical matching set. : ; For the j-th bearing capacity lag correction coefficient in the historical matching case set, For the historical matching case set, the measured bearing capacity lags behind the deformation. For the historical matching case set, the bearing capacity lag deformation output by the j-th multiple linear regression model.
[0015] As a further improvement of the present invention, the specific process of outputting the comprehensive bearing capacity hysteresis correction coefficient is as follows: Calculate the mean and standard deviation of the correction factor; After removing the bearing capacity lag correction coefficients that are outside the normal range, the arithmetic mean of the remaining individual bearing capacity lag correction coefficients after removing outliers is taken as the final comprehensive bearing capacity lag correction coefficient.
[0016] As a further improvement of the present invention, the specific process of revising the bearing capacity lag deformation in the current stress misjudgment risk range using the comprehensive bearing capacity lag correction coefficient is as follows: The comprehensive correction factor K is applied to the calculated value of bearing capacity hysteresis deformation in the current interval to obtain the revised bearing capacity hysteresis deformation. : ; This refers to the revised bearing capacity hysteresis deformation. This represents the bearing capacity hysteresis deformation output by the multiple linear regression model.
[0017] The second objective of this invention is to provide a method for testing the end bearing capacity of prestressed concrete pipe piles, comprising: S1: Obtain the real-time stress loading rate to identify the number of overstress loading rates. Perform cluster analysis on the number of overstress loading rates using a clustering algorithm to identify the overstress loading rate range. Combined with historical misjudgment cases, construct a random forest model to identify the current stress misjudgment risk range. S2: For the current stress misjudgment risk range, combined with the bearing capacity lag deformation corresponding to the stress misjudgment risk range in historical misjudgment cases, predict the bearing capacity lag deformation of the current stress misjudgment risk range. S3: Perform a matching verification between the bearing capacity lag deformation in the current stress misjudgment risk zone and the bearing capacity lag deformation in the corresponding stress misjudgment risk zone in historical misjudgment cases. If they match, determine the bearing capacity lag correction coefficient by combining the bearing capacity lag deformation in the corresponding stress misjudgment risk zone in historical misjudgment cases. Verify the bearing capacity lag correction coefficient, output the comprehensive bearing capacity lag correction coefficient, and revise the bearing capacity lag deformation in the current stress misjudgment risk zone using the comprehensive bearing capacity lag correction coefficient.
[0018] The beneficial effects of this invention are as follows: It can accurately identify the risk range of misjudgment caused by overstress loading, accurately predict and revise the bearing capacity lag deformation, avoid safety accidents such as bridge wall cracking and uneven road surface settlement caused by short-term test misjudgment; improve the accuracy and reliability of pile end bearing capacity testing, and ensure the long-term safety of pile foundation projects. Attached Figure Description
[0019] The invention will now be further described with reference to the accompanying drawings.
[0020] Figure 1 This is a system module diagram of a prestressed pipe pile end bearing capacity testing device according to the present invention; Figure 2 This is a flowchart of the steps of a method for testing the bearing capacity of prestressed pipe pile ends according to the present invention. Detailed Implementation
[0021] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.
[0022] Example 1: This invention aims to solve the problem of misjudgment caused by improper loading rates in existing static load tests, which affects engineering safety. It provides a prestressed pipe pile end bearing capacity testing device, including: a stress misjudgment risk zone identification module, which acquires the real-time stress loading rate, identifies the number of times the rate exceeds the limit, performs cluster analysis on the rate exceeding the limit zone, and uses random forest to identify risk zones based on historical cases; a bearing capacity lag deformation output module, which uses multiple linear regression to predict the deformation amount in the current risk zone based on historical cases; and a bearing capacity lag deformation revision module, which verifies the deformation matching, determines and verifies the correction coefficient, and revises the current deformation amount, such as... Figure 1 As shown in the embodiment of the present invention, a prestressed pipe pile end bearing capacity testing device includes: Stress misjudgment risk interval identification module: acquires real-time stress loading rate to identify the number of times the stress loading rate exceeds the limit, performs cluster analysis on the number of times the stress loading rate exceeds the limit through clustering algorithm to identify the stress loading rate interval, and combines historical misjudgment cases to identify the current stress misjudgment risk interval by constructing a random forest model; In the stress misjudgment risk zone identification module, obtaining the real-time stress loading rate is the basis for identifying the over-stress loading rate and subsequent risk assessment. It is necessary to achieve dynamic monitoring of the loading process through a high-precision sensing and data transmission system to ensure the real-time nature and accuracy of the rate data. The specific process for obtaining the real-time stress loading rate is as follows: A rate monitoring module is integrated into the loading system for testing the end bearing capacity of prestressed concrete pipe piles. The rate monitoring module consists of the following unit components working together: The pressure sensing unit uses a piezoelectric pressure sensor to collect pressure data. The sensing end of the piezoelectric pressure sensor is rigidly connected to the output end of the load cell to ensure that the instantaneous pressure value applied by the load cell to the pile end can be captured in real time. The time synchronization unit uses a GPS time module to add precise timestamps to the pressure data; The rate calculation unit performs differential calculations on the pressure values and corresponding times collected in two consecutive tests using an embedded processor to obtain the pressure change per unit time, i.e., the real-time stress loading rate. The real-time stress loading rate is preprocessed, and the calculation results are filtered by moving average using an embedded processor. The arithmetic mean of five consecutive rate calculations is taken as the effective loading rate at the current moment, and the filtering window is dynamically adjusted according to the rate stability. When the rate fluctuation exceeds 0.2 MPa / min, the window is automatically expanded to the arithmetic mean of 10 rate calculations as the effective loading rate at the current moment to smooth out noise. In the stress misjudgment risk zone identification module, the specific process for identifying the number of overstress loading rates is as follows: Using the timestamp of the real-time stress loading rate as the X-axis and the real-time stress loading rate as the Y-axis, all real-time stress loading rates are plotted on a two-dimensional plane to obtain the real-time stress loading rate curve. In the real-time stress loading rate curve, with 1.0 MPa / min as the first reference point, a straight line parallel to the X-axis is drawn on the first reference point to obtain the first reference line; In the real-time stress loading rate curve, the real-time stress loading rate that exceeds the first reference curve is denoted as the overstress loading rate. Statistically analyze all real-time stress loading rates exceeding the first reference curve to identify the number of overstress loading rates; In the stress misjudgment risk interval identification module, the specific process of performing cluster analysis on the number of overstress loading rates to identify the overstress loading rate interval is as follows: Obtain the timestamp and real-time stress loading rate corresponding to each overstress loading rate, and represent them as data points; Each data point is represented as , Let i be the timestamp corresponding to the i-th overstress loading rate. This is the real-time stress loading rate corresponding to the i-th overstress loading rate; In the real-time stress loading rate curve, the overstress loading rate is marked as event 1, event 2... event n in ascending order of timestamps (n is the total number of overstress loading rate events). Since the timestamps and real-time stress loading rates in the data points have different dimensions, the Z-score normalization method is used to standardize the data points to eliminate the influence of dimensions and ensure the fairness of subsequent clustering algorithms. The DBSCAN algorithm is used to cluster the standardized data points to identify dense regions in the timestamp-real-time stress loading rate space; The neighborhood radius is set by comprehensively considering both the timestamp dimension and the real-time stress loading rate dimension. In terms of timestamps, if the time difference between two events is less than a time difference threshold... (like (equal to 30 seconds), considered to be time proximity; in the dimension of real-time stress loading rate, if the rate difference between two events is less than the rate difference threshold... (like (Equal to 0.1 MPa / min), considered as a rate proximity; The neighborhood radius is set by combining the timestamp dimension and the real-time stress loading rate dimension. ;in, The time difference between two events. The difference in rates between the two events; If the neighborhood radius of any two consecutive events is less than 1, it means that the relative difference between the two points in time and rate is relatively small, and they can be regarded as neighboring points. The DBSCAN algorithm is used to cluster data points and output the clusters corresponding to the overstress loading rate intervals. The corresponding cluster is the overstress loading rate range; In the stress misjudgment risk zone identification module, the specific process of identifying the current stress misjudgment risk zone by constructing a random forest model based on historical misjudgment cases is as follows: Obtain the clusters output by the clustering algorithm corresponding to the overstress loading rate range, calculate the minimum and maximum range of loading rates within the cluster, denoted as the loading rate difference within the cluster, and calculate the number of overstress loading rate events. Obtain the clusters output by the clustering algorithm for each misjudgment in historical misjudgment cases, and calculate the number of loading rate differences and overstress loading rate events within the clusters; Set up positive sample cases and negative sample cases. Positive sample cases are those that cause misjudgment of bearing capacity due to the overstress loading rate range, and are marked as normal with a label of 1. Negative sample cases are those that do not cause misjudgment due to the overstress loading rate range, and are marked as misjudged with a label of 0. Extract the intra-cluster loading rate difference and the number of overstress loading rate events corresponding to each group of positive and negative samples to form a training set. Divide the training set into a sub-training set and a validation set in an 8:2 ratio. Use the training set as input and the misclassified labels as output, and input them into the random forest model for training. Evaluate the performance of the random forest model through the validation set until the accuracy of the random forest model in correctly identifying misclassified labels is greater than 95%, then the random forest model is completed. Input the intra-cluster loading rate difference and the number of overstress loading rate events corresponding to the overstress loading rate interval into the trained random forest model to identify the stress misjudgment risk interval. The historical misjudgment cases refer to past cases where bearing capacity was misjudged due to overstress loading rate intervals. Specifically, historical misjudgment cases form the basis for constructing positive and negative samples, which are used to extract features such as intra-cluster loading rate differences and the number of overstress loading rate events. These features are then used to train a random forest model to identify stress misjudgment risk intervals and, combined with the corresponding bearing capacity lag deformation, predict the deformation amount in the current risk interval. Ultimately, this achieves effective identification and early warning of bearing capacity misjudgment risks.
[0023] Load-bearing capacity lag deformation output module: For the current stress misjudgment risk range, combined with the load-bearing capacity lag deformation corresponding to the stress misjudgment risk range in historical misjudgment cases, predict the load-bearing capacity lag deformation of the current stress misjudgment risk range; In the load-bearing capacity hysteresis deformation output module, the specific process for predicting the load-bearing capacity hysteresis deformation in the current stress misjudgment risk range is as follows: Obtain the characteristic parameters of the stress misjudgment risk range in historical misjudgment cases, including: independent variable characteristics and dependent variable characteristics; The independent variable characteristics include: intra-cluster loading rate difference, number of overstress loading rate events, duration of stress misjudgment risk interval, and soil parameters, including: compression modulus and moisture content. The dependent variable characteristics include: the measured value of bearing capacity lag deformation (the difference between the settlement that should occur under normal loading rate and the settlement that occurs under actual over-rate loading). The independent variable characteristics of the stress misjudgment risk range in historical misjudgment cases are normalized; Historical misjudgment cases were randomly divided into a training set and a validation set in a 7:3 ratio (ensuring that the distribution of independent variable features and the range of dependent variable features are consistent between the two sets of samples). The training set was used for learning the parameters of the multiple linear regression model, and the validation set was used for evaluating the fit of the multiple linear regression model.
[0024] The independent variable features are used as inputs to the multiple linear regression model, and the dependent variable features are used as outputs. The multiple linear regression model is trained using the training set. The linear relationship between the independent and dependent variables is fitted using the least squares method. The regression equation is solved. The normalized independent variables of the validation set are input into the model, and the coefficient of determination between the predicted and measured values is calculated. When the coefficient of determination of the multiple linear regression model is greater than 0.85, the model fit is deemed to meet the engineering requirements, and the multiple linear regression model is considered to have been successfully constructed. For the current stress misjudgment risk range, the independent variable features of the current stress misjudgment risk range are extracted, and after normalization, they are input into the constructed multiple linear regression model. The multiple linear regression model predicts the bearing capacity hysteresis deformation of the current stress misjudgment risk range. The bearing capacity lag deformation revision module: performs a matching verification between the bearing capacity lag deformation of the current stress misjudgment risk range and the bearing capacity lag deformation of the corresponding stress misjudgment risk range in historical misjudgment cases. If they match, it determines the bearing capacity lag correction coefficient by combining the bearing capacity lag deformation of the corresponding stress misjudgment risk range in historical misjudgment cases, verifies the bearing capacity lag correction coefficient, outputs the comprehensive bearing capacity lag correction coefficient, and uses the comprehensive bearing capacity lag correction coefficient to revise the bearing capacity lag deformation of the current stress misjudgment risk range. In the bearing capacity lag deformation revision module, the specific process for matching verification between the bearing capacity lag deformation in the current stress misjudgment risk range and the bearing capacity lag deformation corresponding to the stress misjudgment risk range in historical misjudgment cases is as follows: Historical misjudgment cases that are consistent with the soil type parameters in the current stress misjudgment risk range are selected to form a historical reference case set; For cases in the current stress misjudgment risk range and the historical reference case set, extract core feature parameters and process them uniformly; Among them, the core features include: Intra-cluster loading rate difference, number of overstress loading rate events, and duration of stress misjudgment risk interval; The specific steps for uniformly processing the core characteristic parameters of the current stress misjudgment risk range and cases in the historical reference case set are as follows: The min-max standardization method was used to normalize the intra-cluster loading rate difference, the number of overstress loading rate events, and the duration of the stress misjudgment risk interval to eliminate dimensional differences. After normalization: The core feature parameters of the current stress misjudgment risk range form the current feature vector; The core feature parameters of cases in the historical reference case set form a historical feature vector; Using the Euclidean distance formula, the spatial distance between the current feature vector and each historical feature vector is calculated and denoted as the feature similarity: Compare feature similarity with a feature similarity threshold; If the feature similarity is less than or equal to the feature similarity threshold, it indicates that the bearing capacity lag deformation output by the multiple linear regression model matches historical misjudgment cases; it can be used for subsequent deformation correction. If the feature similarity is greater than the feature similarity threshold, it means that the bearing capacity lag deformation output by the multiple linear regression model does not match the historical misjudgment cases; it is necessary to return to the multiple linear regression model to re-verify the bearing capacity lag deformation until a match is found. It should be noted that the feature similarity threshold is a judgment threshold determined by those skilled in the art based on historical experience; In the bearing capacity lag deformation revision module, the specific process for determining the bearing capacity lag correction coefficient by combining the bearing capacity lag deformation corresponding to the stress misjudgment risk interval in historical misjudgment cases is as follows: Historical misjudgment cases that match the current stress misjudgment risk range are selected to form a historical matching case set; Extract the two core sets of data for each case from the historical matching case set, including: Measured bearing capacity hysteresis deformation and bearing capacity hysteresis deformation; Calculate the bearing capacity lag correction coefficient for each case in the historical matching set. : ; For the j-th bearing capacity lag correction coefficient in the historical matching case set, For the historical matching case set, the measured bearing capacity lags behind the deformation. For the historical matching case set, the bearing capacity hysteresis deformation output by the j-th multiple linear regression model; In the bearing capacity lag deformation revision module, the specific process of verifying the bearing capacity lag correction coefficient and outputting the comprehensive bearing capacity lag correction coefficient is as follows: Calculate the mean and standard deviation of the correction factor; Using the 3σ principle, bearing capacity lag correction coefficients that exceed the normal range are eliminated. The arithmetic mean of the remaining individual bearing capacity lag correction coefficients after eliminating outliers is taken as the final comprehensive bearing capacity lag correction coefficient. In the bearing capacity lag deformation revision module, the specific process of revising the bearing capacity lag deformation in the current stress misjudgment risk range using the comprehensive bearing capacity lag correction coefficient is as follows: The comprehensive correction factor K is applied to the calculated value of bearing capacity hysteresis deformation in the current interval to obtain the revised bearing capacity hysteresis deformation. : ; in, This refers to the revised bearing capacity hysteresis deformation. This represents the bearing capacity hysteresis deformation output by the multiple linear regression model.
[0025] Example 2 like Figure 2 As shown in Example 1, this invention provides a method for testing the end bearing capacity of prestressed pipe piles, specifically including: S1: Obtain the real-time stress loading rate to identify the number of overstress loading rates. Perform cluster analysis on the number of overstress loading rates using a clustering algorithm to identify the overstress loading rate range. Combined with historical misjudgment cases, construct a random forest model to identify the current stress misjudgment risk range. S2: For the current stress misjudgment risk range, combined with the bearing capacity lag deformation corresponding to the stress misjudgment risk range in historical misjudgment cases, predict the bearing capacity lag deformation of the current stress misjudgment risk range. S3: Perform a matching verification between the bearing capacity lag deformation in the current stress misjudgment risk zone and the bearing capacity lag deformation in the corresponding stress misjudgment risk zone in historical misjudgment cases. If they match, determine the bearing capacity lag correction coefficient by combining the bearing capacity lag deformation in the corresponding stress misjudgment risk zone in historical misjudgment cases. Verify the bearing capacity lag correction coefficient, output the comprehensive bearing capacity lag correction coefficient, and revise the bearing capacity lag deformation in the current stress misjudgment risk zone using the comprehensive bearing capacity lag correction coefficient.
[0026] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.
Claims
1. A device for testing the bearing capacity of prestressed concrete pipe pile ends, characterized in that: include: Stress misjudgment risk interval identification module: acquires real-time stress loading rate to identify the number of times the stress loading rate exceeds the limit, performs cluster analysis on the number of times the stress loading rate exceeds the limit through clustering algorithm to identify the stress loading rate interval, and combines historical misjudgment cases to identify the current stress misjudgment risk interval by constructing a random forest model; Load-bearing capacity lag deformation output module: For the current stress misjudgment risk range, combined with the load-bearing capacity lag deformation corresponding to the stress misjudgment risk range in historical misjudgment cases, predict the load-bearing capacity lag deformation of the current stress misjudgment risk range; The load-bearing capacity lag deformation revision module: performs a matching verification between the load-bearing capacity lag deformation in the current stress misjudgment risk range and the load-bearing capacity lag deformation corresponding to the stress misjudgment risk range in historical misjudgment cases. If they match, it determines the load-bearing capacity lag correction coefficient by combining the load-bearing capacity lag deformation corresponding to the stress misjudgment risk range in historical misjudgment cases, verifies the load-bearing capacity lag correction coefficient, outputs the comprehensive load-bearing capacity lag correction coefficient, and uses the comprehensive load-bearing capacity lag correction coefficient to revise the load-bearing capacity lag deformation in the current stress misjudgment risk range.
2. The prestressed pipe pile end bearing capacity testing device according to claim 1, characterized in that: The specific process for obtaining the real-time stress loading rate is as follows: Pressure data is collected using a piezoelectric pressure sensor and a GPS timing module is used to add precise timestamps to the pressure data. An embedded processor performs differential calculations on two consecutive pressure values and their corresponding times to obtain the pressure change per unit time, i.e., the real-time stress loading rate.
3. The prestressed pipe pile end bearing capacity testing device according to claim 1, characterized in that: The specific process for identifying the number of overstress loading rates is as follows: Using the timestamp of the real-time stress loading rate as the X-axis and the real-time stress loading rate as the Y-axis, all real-time stress loading rates are plotted on a two-dimensional plane to obtain the real-time stress loading rate curve. In the real-time stress loading rate curve, with 1.0 MPa / min as the first reference point, a straight line parallel to the X-axis is drawn on the first reference point to obtain the first reference line; In the real-time stress loading rate curve, the real-time stress loading rate that exceeds the first reference curve is denoted as the overstress loading rate. Statistically analyze all real-time stress loading rates that exceed the first baseline curve to identify the number of overstress loading rates.
4. The prestressed pipe pile end bearing capacity testing device according to claim 1, characterized in that: The specific process for identifying the overstress loading rate range is as follows: Obtain the timestamp and real-time stress loading rate corresponding to each overstress loading rate, and represent them as data points; Each data point is represented as , Let i be the timestamp corresponding to the i-th overstress loading rate. This is the real-time stress loading rate corresponding to the i-th overstress loading rate; In the real-time stress loading rate curve, the overstress loading rate is marked as event 1, event 2... event n in ascending order of timestamps, where n is the total number of overstress loading rate events; The DBSCAN algorithm is used to cluster data points and output the clusters corresponding to the overstress loading rate intervals; the corresponding clusters are the overstress loading rate intervals.
5. The prestressed pipe pile end bearing capacity testing device according to claim 1, characterized in that: The specific process for identifying the current stress misjudgment risk range is as follows: Obtain the clusters output by the clustering algorithm corresponding to the overstress loading rate range, calculate the minimum and maximum range of loading rates within the cluster, denoted as the loading rate difference within the cluster, and calculate the number of overstress loading rate events. Obtain the clusters output by the clustering algorithm for each misjudgment in historical misjudgment cases, and calculate the number of loading rate differences and overstress loading rate events within the clusters; Set up positive and negative sample cases, extract the intra-cluster loading rate difference and the number of overstress loading rate events corresponding to each group of positive and negative samples to form a training set and divide it into a sub-training set and a validation set, and train the random forest model. Input the intra-cluster loading rate difference and the number of overstress loading rate events corresponding to the overstress loading rate interval into the trained random forest model to identify the stress misjudgment risk interval.
6. The prestressed pipe pile end bearing capacity testing device according to claim 1, characterized in that: The specific process for predicting the bearing capacity hysteresis deformation in the current stress misjudgment risk range is as follows: Obtain the characteristic parameters of the stress misjudgment risk range from historical misjudgment cases, including: Independent variable characteristics include: intra-cluster loading rate difference, number of overstress loading rate events, duration of stress misjudgment risk interval, and soil parameters, including: compression modulus and moisture content; dependent variable characteristics include: measured value of bearing capacity hysteresis deformation. Historical misjudgment cases are divided into training and validation sets; the independent variable features are used as inputs to the multiple linear regression model, and the dependent variable features are used as outputs. When the determination coefficient of the multiple linear regression model is greater than 0.85, the multiple linear regression model is completed. For the current stress misjudgment risk range, the independent variable features of the current stress misjudgment risk range are extracted and used to construct a multiple linear regression model to predict the bearing capacity hysteresis deformation of the current stress misjudgment risk range.
7. The prestressed pipe pile end bearing capacity testing device according to claim 1, characterized in that: The specific process for verifying the matching between the bearing capacity hysteresis deformation in the current stress misjudgment risk zone and the bearing capacity hysteresis deformation in the corresponding stress misjudgment risk zone in historical misjudgment cases is as follows: Historical misjudgment cases that are consistent with the soil type parameters in the current stress misjudgment risk range are selected to form a historical reference case set; For cases in the current stress misjudgment risk range and the historical reference case set, extract core feature parameters and process them uniformly; Key features include: intra-cluster loading rate difference, number of overstress loading rate events, and duration of stress misjudgment risk interval; The core feature parameters of the current stress misjudgment risk range form the current feature vector; the core feature parameters of the cases in the historical reference case set form the historical feature vector; Using the Euclidean distance formula, the spatial distance between the current feature vector and each historical feature vector is calculated and recorded as the feature similarity. If the feature similarity is less than or equal to the feature similarity threshold, it means that the bearing capacity lag deformation output by the multiple linear regression model matches the historical misjudgment cases.
8. The prestressed pipe pile end bearing capacity testing device according to claim 1, characterized in that: The specific process for determining the bearing capacity lag correction coefficient by combining the bearing capacity lag deformation corresponding to the stress misjudgment risk interval in historical misjudgment cases is as follows: Historical misjudgment cases that match the current stress misjudgment risk range are selected to form a historical matching case set; Extract the two sets of core data for each case from the historical matching case set, including: measured bearing capacity lag deformation and bearing capacity lag deformation; Calculate the bearing capacity lag correction coefficient for each case in the historical matching set. : ; For the j-th bearing capacity lag correction coefficient in the historical matching case set, For the historical matching case set, the measured bearing capacity lags behind the deformation. For the historical matching case set, the bearing capacity lag deformation output by the j-th multiple linear regression model.
9. The prestressed pipe pile end bearing capacity testing device according to claim 1, characterized in that: The specific process for outputting the comprehensive bearing capacity hysteresis correction coefficient is as follows: Calculate the mean and standard deviation of the correction factor; After removing the bearing capacity lag correction coefficients that are outside the normal range, the arithmetic mean of the remaining individual bearing capacity lag correction coefficients after removing outliers is taken as the final comprehensive bearing capacity lag correction coefficient.
10. The prestressed pipe pile end bearing capacity testing device according to claim 1, characterized in that: The specific process for revising the bearing capacity lag deformation in the current stress misjudgment risk range using the comprehensive bearing capacity lag correction coefficient is as follows: The comprehensive correction factor K is applied to the calculated value of bearing capacity hysteresis deformation in the current interval to obtain the revised bearing capacity hysteresis deformation. : ; This refers to the revised bearing capacity hysteresis deformation. This represents the bearing capacity hysteresis deformation output by the multiple linear regression model.