Engineering site drilling vs30 query system based on shear wave prediction
By constructing and optimizing the shear wave prediction engineering model, and utilizing historical data and model validation indices, the problems of low efficiency and insufficient accuracy in existing shear wave prediction queries have been solved, achieving more efficient and accurate shear wave prediction queries.
Patent Information
- Application Number
- CN202411131790.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-19
- Publication Date
- 2024-11-01
- Estimated Expiration
- 2044-08-19
AI Technical Summary
The existing engineering site borehole VS30 query system relies on cumbersome on-site testing and manual analysis. The data processing cycle is long and easily affected by human factors. The complex and variable geological conditions of engineering sites make shear wave prediction query difficult.
By acquiring historical engineering data from the engineering site, a shear wave prediction engineering model is constructed. The model is then validated and optimized. The optimal hyperparameters are selected using the Bayesian optimization method. The model is further optimized by combining the shear wave prediction validation index and the application test performance index, and then actual queries are performed.
It improves the query efficiency and accuracy of shear wave prediction, enables numerical evaluation of the accuracy and performance of shear wave prediction models, and solves the problem of high query difficulty.
Smart Images

Figure CN118656306B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of electrical digital data processing, and in particular to an engineering site drilling VS30 query system established based on shear wave prediction. Background Art
[0002] With the rapid advancement of science and technology and the in-depth development of engineering practice, the fields of geological exploration and earthquake engineering have ushered in unprecedented changes. The emergence of high-precision and high-efficiency exploration equipment and data analysis technology has made it possible to accurately evaluate the characteristics of engineering sites. In particular, with the support of big data and artificial intelligence technology, the prediction of shear wave velocity is no longer limited to traditional empirical models, but can combine multi-source data to achieve more intelligent and dynamic prediction and analysis. This not only improves the accuracy of the prediction results, but also greatly shortens the evaluation cycle, providing more timely and reliable data support for engineering design and construction. Therefore, the engineering site drilling vs30 query system established based on shear wave prediction is an innovative achievement that has emerged in this era. It has broad application prospects and will make important contributions to engineering seismic fortification.
[0003] The existing engineering site borehole vs30 query system based on shear wave prediction has achieved efficient prediction and convenient query of the shear wave velocity of the engineering site borehole by integrating advanced data acquisition technology, intelligent data processing algorithms and user-friendly interactive interface. The system can automatically collect and analyze geological exploration data, use machine learning or deep learning models to accurately predict the shear wave velocity at different depths, and calculate the vs30 value accordingly. The user only needs to enter the basic information of the engineering site, and the system can quickly return a detailed prediction report, including the borehole distribution map, shear wave velocity curve and vs30 evaluation results, which greatly improves the efficiency and accuracy of the engineering evaluation. In addition, the system also has data visualization function, which makes the prediction results more intuitive and easy to understand, providing strong decision-making support for engineers.
[0004] For example, the method and device for calculating the shear wave velocity of a site disclosed in the invention patent announcement with the announcement number CN113267814B include: obtaining the lithology description information of the target soil layer, and confirming the soil type corresponding to the target soil layer according to the lithology description information; correcting the empirical shear wave velocity data corresponding to the soil type according to the additional stress on the soil of the target soil layer at its depth, and using the corrected data as the shear wave velocity data of the target soil layer; and calculating the average shear wave velocity of the covering soil layer within a preset depth range based on the shear wave velocity data of the target soil layer, as the shear wave velocity of the site to be measured.
[0005] For example, a method and system for accurately distinguishing seismic shear waves disclosed in a patent application with publication number: CN104656135A includes the following steps: converting waveform data of different formats recorded by instruments of different models into a predetermined proprietary data format; rearranging the component data and editing the file header of the data obtained after the format conversion; establishing an interpretation parameter file required for analysis and calculation; comparing and analyzing the characteristics of the shear waves and correctly picking up the first arrival travel time of the seismic shear waves, and automatically calculating the test results of each component; taking values of the obtained test results and outputting the results.
[0006] However, in the process of implementing the technical solution of the invention in the embodiments of the present application, the present application found that the above technology has at least the following technical problems:
[0007] In the prior art, the traditional engineering site drilling VS30 query system relies on cumbersome on-site testing and manual analysis, the data processing cycle is long and is easily affected by human factors, and the geological conditions of the engineering site are complex and changeable, and there is a problem of difficulty in querying the shear wave prediction of the engineering site. Summary of the invention
[0008] The embodiment of the present application solves the problem of difficulty in querying shear wave predictions of engineering sites in the prior art by providing a vs30 query system for drilling holes at engineering sites based on shear wave predictions, thereby improving the query efficiency of shear wave predictions at engineering sites.
[0009] The embodiment of the present application provides a drilling vs30 query system for an engineering site established based on shear wave prediction, comprising: a data acquisition module, a model establishment and optimization module and an application test module; wherein the data acquisition module is used to obtain historical engineering data of the engineering site, the historical engineering data including historical engineering site parameters and historical drilling vs30 data; the model establishment and optimization module is used to use the acquired historical engineering data to perform model training to construct a shear wave prediction engineering model, and to perform model verification on the constructed shear wave prediction engineering model to obtain a first shear wave prediction engineering model; the specific process of the model verification is: obtaining prediction verification indicators of the shear wave prediction engineering model from the model verification process, the prediction verification indicators including shear wave velocity and shear wave attenuation velocity; verifying the shear wave prediction according to the obtained prediction verification indicators The shear wave prediction engineering model is iteratively calculated to obtain a shear wave prediction verification index and an average shear wave prediction verification index, wherein the shear wave prediction verification index is used to describe the accuracy of the prediction of the shear wave prediction engineering model; the average shear wave prediction verification index is compared with a reference average verification value, if it is not less than the reference average verification value, the shear wave prediction engineering model is integrated with the shear wave query system for model system integration, otherwise the shear wave prediction engineering model is optimized according to the shear wave prediction verification index; the application test module is used to perform application testing on the first shear wave prediction engineering model to obtain an application test performance index, and the first shear wave prediction engineering model whose application test performance index meets the performance threshold range is applied to actual query, wherein the application test performance index is used to describe the performance of the first shear wave prediction engineering model.
[0010] Furthermore, the method for obtaining the shear wave prediction verification index is as follows: obtaining a prediction verification coefficient according to the prediction verification index, the prediction verification coefficient including a shear wave velocity coefficient and a shear wave attenuation velocity coefficient; obtaining a shear wave prediction verification index of the shear wave prediction engineering model according to the obtained prediction verification coefficient, the shear wave prediction verification index is calculated using the following formula:
[0011] ;
[0012] In the formula, f is the number of times the model is verified, , F is the total number of model validations, is the shear wave prediction validation index of the f-th model validation, is the shear wave velocity coefficient of the f-th model verification, is the reference shear wave velocity coefficient for the f-th model verification, is the shear wave attenuation velocity coefficient of the f-th model verification, is the reference shear wave attenuation velocity coefficient for the f-th model verification.
[0013] Furthermore, the specific process of the model optimization is: selecting the optimal hyperparameters of the shear wave prediction engineering model through the Bayesian optimization method in combination with the shear wave prediction verification index; obtaining historical engineering feature data and combining and cross-linking them to form first historical engineering feature data, normalizing the first historical engineering feature data, and optimizing the first historical engineering feature data whose skewness distribution exceeds the skewness threshold range to obtain second historical engineering feature data; using the obtained second historical engineering feature data to train the shear wave prediction engineering model until the obtained shear wave model optimization index is not less than the reference shear wave model optimization index, and the shear wave prediction engineering model is recorded as the first shear wave prediction engineering model, and the shear wave model optimization index is used to describe the optimization degree of the shear wave prediction engineering model.
[0014] Furthermore, the method for obtaining the shear wave model optimization index is as follows: extracting the model optimization index of the shear wave prediction engineering model from the model optimization process, the model optimization index including complexity and data distribution; obtaining the model optimization score according to the extracted model optimization index, the model optimization score including complexity score and data distribution score; obtaining the shear wave model optimization index of the shear wave prediction engineering model according to the obtained model optimization score.
[0015] Furthermore, the first shear wave prediction engineering model is used to obtain predicted drilling vs30 data. The specific method for obtaining the predicted drilling vs30 data is as follows: the first shear wave prediction engineering model includes a constant velocity prediction model and a velocity prediction model; the predicted drilling vs30 data is calculated by the constant velocity prediction model as follows:
[0016] ;
[0017] In the formula, e is a natural constant, p is the number of the site to be predicted, , P is the total number of sites to be predicted, is the predicted drilling vs30 data of the pth site to be predicted, is the soil layer depth of the pth site to be predicted, is the pth site to be predicted Shear wave travel time at soil depth, is the pth site to be predicted The instantaneous shear wave velocity at the depth of the soil layer; the predicted drilling vs30 data is calculated by the velocity prediction model as follows:
[0018] ;
[0019] In the formula, p is the number of the site to be predicted, , P is the total number of sites to be predicted, is the predicted drilling vs30 data of the pth site to be predicted, is the first regression coefficient, is the second regression coefficient, is the soil layer depth of the pth site to be predicted, is the pth site to be predicted The average shear wave velocity at the depth of the soil layer.
[0020] Furthermore, the specific steps of the application test are:
[0021] The test borehole vs30 data corresponding to the input test engineering site parameters obtained by the first shear wave prediction engineering model within a preset time interval is obtained, and the first shear wave prediction engineering model is performance tested to obtain an application test performance index, and the application test performance index is compared with the performance threshold range, and the first shear wave prediction engineering model that meets the performance threshold range is applied to the actual query, and the model that does not meet the performance threshold range is rebuilt.
[0022] Furthermore, the method for obtaining the application test performance index is as follows:
[0023] Extracting performance indicators of the first shear wave prediction engineering model within a preset time interval from the performance test process, wherein the performance indicators include response time, number of concurrent users, resource utilization and error rate;
[0024] The resource utilization is the usage of system resources such as CPU, memory, disk I / O, etc. The error rate is the ratio of the number of requests in which errors occur in the first shear wave prediction engineering model to the total number of requests within a preset time interval;
[0025] Acquire a performance coefficient according to the extracted performance index, wherein the performance coefficient includes a response time coefficient and a concurrent user coefficient, wherein the response time coefficient represents the ratio of the difference between the response time and its reference value and its reference deviation, and the concurrent user coefficient represents the ratio of the difference between the number of concurrent users and its reference value and its reference deviation;
[0026] The application test performance index of the first shear wave prediction engineering model is obtained based on the obtained performance indicators and performance coefficients.
[0027] Furthermore, the application test performance index is calculated using the following formula:
[0028] ;
[0029] Where j is the number of the first shear wave prediction engineering model, , J is the total number of the first shear wave prediction engineering models, The performance index of the application test of the engineering model for the jth first shear wave prediction, is the response time coefficient of the jth first shear wave prediction engineering model, is the concurrent user coefficient of the jth first shear wave prediction engineering model, The resource utilization of the engineering model for the jth first shear wave prediction, is the error rate of the engineering model for the jth first shear wave prediction, is the effective extreme value of the error rate.
[0030] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:
[0031] 1. A shear wave prediction engineering model is constructed by acquiring historical engineering data of the engineering site, and the constructed shear wave prediction engineering model is verified to obtain the first shear wave prediction engineering model. Then, an application test is performed on the first shear wave prediction engineering model to obtain an application test performance index. Finally, the first shear wave prediction engineering model whose application test performance index meets the performance threshold range is applied to actual queries, thereby improving the query efficiency of shear wave prediction of engineering sites and effectively solving the problem of high query difficulty of shear wave prediction of engineering sites in the prior art.
[0032] 2. By obtaining the prediction verification indicators of the shear wave prediction engineering model from the model verification process, and obtaining the prediction verification coefficient based on the prediction verification indicators, and then obtaining the shear wave prediction verification index of the shear wave prediction engineering model based on the obtained prediction verification coefficient, and finally evaluating the accuracy of the prediction of the shear wave prediction engineering model through the obtained shear wave prediction verification index, the accuracy of the prediction of the shear wave prediction engineering model is digitized, thereby achieving a more accurate evaluation of the accuracy of the prediction of the shear wave prediction engineering model.
[0033] 3. By extracting the performance indicators of the first shear wave prediction engineering model within a preset time interval from the performance test process, obtaining the performance coefficient based on the extracted performance indicators, and then obtaining the application test performance index of the first shear wave prediction engineering model based on the obtained performance indicators and performance coefficient, and finally evaluating the performance of the first shear wave prediction engineering model through the obtained application test performance index, the performance of the first shear wave prediction engineering model is quantified, thereby achieving a more accurate evaluation of the performance of the first shear wave prediction engineering model. BRIEF DESCRIPTION OF THE DRAWINGS
[0034] Figure 1 A schematic diagram of the structure of the engineering site drilling vs30 query system based on shear wave prediction provided in an embodiment of the present application;
[0035] Figure 2A schematic diagram of the structure of a data acquisition module in a vs30 query system for drilling holes in an engineering site established based on shear wave prediction provided in an embodiment of the present application;
[0036] Figure 3 This is a statistical chart of changes in the shear wave prediction verification index provided in an embodiment of the present application. DETAILED DESCRIPTION
[0037] The embodiment of the present application solves the problem of difficulty in querying shear wave predictions of engineering sites in the prior art by providing an engineering site drilling VS30 query system based on shear wave predictions. The system obtains historical engineering data of the engineering site, and then uses the collected historical engineering data to perform model training to construct a shear wave prediction engineering model. The constructed shear wave prediction engineering model is then model verified to obtain a first shear wave prediction engineering model. The first shear wave prediction engineering model is then application tested to obtain an application test performance index. Finally, the first shear wave prediction engineering model whose application test performance index meets the performance threshold range is applied to actual queries, thereby improving the query efficiency of shear wave predictions of engineering sites.
[0038] The technical solution in the embodiment of the present application is to solve the problem of difficulty in querying the shear wave prediction of the above-mentioned engineering site. The overall idea is as follows:
[0039] By acquiring historical engineering data of the engineering site, a shear wave prediction engineering model is constructed and the first shear wave prediction engineering model is obtained by model verification. Then, the first shear wave prediction engineering model is applied and tested to obtain an application test performance index. Finally, the first shear wave prediction engineering model whose application test performance index meets the performance threshold range is applied to actual queries, thereby improving the query efficiency of shear wave prediction of engineering sites.
[0040] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.
[0041] like Figure 1As shown, it is a structural schematic diagram of the engineering site drilling vs30 query system based on shear wave prediction provided in an embodiment of the present application. The engineering site drilling vs30 query system based on shear wave prediction provided in an embodiment of the present application includes: a data acquisition module, a model establishment optimization module and an application testing module; wherein the data acquisition module is used to obtain historical engineering data of the engineering site, and the historical engineering data includes historical engineering site parameters and historical drilling vs30 data; the model establishment optimization module is used to use the collected historical engineering data to perform model training to construct a shear wave prediction engineering model, and to perform model verification on the constructed shear wave prediction engineering model to obtain a first shear wave prediction engineering model; the specific process of model verification is: obtaining prediction verification indicators of the shear wave prediction engineering model from the model verification process, and the prediction verification indicators include shear wave velocity and shear wave attenuation deceleration; according to the obtained prediction verification index, the shear wave prediction engineering model is iteratively calculated to obtain the shear wave prediction verification index and the average shear wave prediction verification index, which is used to describe the accuracy of the prediction of the shear wave prediction engineering model; the average shear wave prediction verification index is compared with the reference average verification value, if it is not less than the reference average verification value, the shear wave prediction engineering model is integrated with the shear wave query system for model system integration, otherwise the shear wave prediction engineering model is optimized according to the shear wave prediction verification index; the application test module is used to perform application test on the first shear wave prediction engineering model to obtain the application test performance index, and the first shear wave prediction engineering model whose application test performance index meets the performance threshold range is applied to the actual query, and the application test performance index is used to describe the performance of the first shear wave prediction engineering model.
[0042] In this embodiment, the historical engineering site parameters include the historical geographical location, the historical area size and the historical soil layer depth. The historical geographical location is determined by the Global Positioning System (GPS). The historical area size is obtained by obtaining the location coordinates of the historical geographical location, and then the historical area size is obtained by the coordinate analysis method. The coordinate analysis method takes the absolute value of the difference between the position coordinates of the corresponding coordinate axes, and multiplies the absolute values obtained to obtain the historical area size. The historical soil layer depth is obtained by using a pressure device to press a sounding rod with a sounding probe into the test soil layer, and measuring the penetration resistance of the soil through a measurement system. The historical borehole vs30 data is the borehole vs30 corresponding to the historical engineering site parameters. The historical borehole vs30 data is obtained by recording the signal received by the three-component detector. The historical borehole vs30 data is a three-component detector. The numerical value of the signal received by the device; the shear wave prediction engineering model is a model that outputs the corresponding predicted drilling VS30 data according to the input engineering site parameters; the shear wave attenuation velocity is the attenuation degree of the shear wave per unit time during the propagation process; the iterative calculation means that the shear wave prediction engineering model is verified multiple times and the shear wave prediction verification index of the shear wave prediction engineering model is obtained according to the prediction verification indicators obtained during the model verification process; the average shear wave prediction verification index is the average value of the shear wave prediction verification index, and the average value is the ratio of the sum of the shear wave prediction verification indexes obtained by iterative calculation to the number of iterations; the query efficiency of the shear wave prediction of the engineering site is improved.
[0043] Further, such as Figure 2 As shown, it is a structural schematic diagram of the data acquisition module in the engineering site drilling vs30 query system established based on shear wave prediction provided in an embodiment of the present application. The data acquisition module includes a data acquisition unit and a preprocessing unit; the data acquisition unit: is used to collect the initial historical engineering data of the engineering site and transmit the collected initial historical engineering data to the preprocessing unit; the preprocessing unit: is used to preprocess the received initial historical engineering data to obtain historical engineering data.
[0044] In this embodiment, the preprocessing includes removing outliers, filling missing values and standardization; the step of removing outliers is: first, a box plot is drawn through quartiles (Q1, Q2, Q3), where Q1 is the lower quartile (25% quartile), Q3 is the upper quartile (75% quartile), and outliers are defined as data points less than Q1-1.5IQR (IQR is the interquartile range, i.e., Q3-Q1) or greater than Q3+1.5IQR. The box plot can intuitively display the distribution of data to identify outliers, and the identified outliers are deleted; the step of filling missing values is: counting the number of missing values in the initial historical engineering data, and using the mean of non-missing data to fill the missing values; the step of standardization is: normalizing the field names to ensure that the same fields in different data sources have a unified name, converting the data into a unified data type, and converting the text data into a unified encoding format; the availability of historical engineering data collection is improved, and the query efficiency of shear wave prediction of engineering sites is further improved.
[0045] Furthermore, the method for obtaining the shear wave prediction verification index is as follows: the prediction verification coefficient is obtained according to the prediction verification index, the prediction verification coefficient includes the shear wave velocity coefficient and the shear wave attenuation velocity coefficient, the shear wave velocity coefficient is the ratio of the shear wave velocity to the corresponding reference deviation, and the shear wave attenuation velocity coefficient is the ratio of the shear wave attenuation velocity to the corresponding reference deviation; the shear wave prediction verification index of the shear wave prediction engineering model is obtained according to the obtained prediction verification coefficient, and the shear wave prediction verification index is calculated using the following formula:
[0046] ;
[0047] In the formula, f is the number of times the model is verified, , F is the total number of model validations, is the shear wave prediction validation index of the f-th model validation, is the shear wave velocity coefficient of the f-th model verification, is the reference shear wave velocity coefficient for the f-th model verification, is the shear wave attenuation velocity coefficient of the f-th model verification, is the reference shear wave attenuation velocity coefficient for the f-th model verification.
[0048] In this embodiment, the expression of the shear wave velocity coefficient is: ,in, is the shear wave velocity of the f-th model verification, is the shear wave velocity reference deviation; the expression of the shear wave attenuation velocity coefficient is ,in, is the shear wave attenuation velocity of the f-th model verification, is the reference deviation of shear wave attenuation velocity; let the first coefficient of shear wave velocity , let the first coefficient of shear wave attenuation velocity be , is the first coefficient of shear wave velocity of the f-th model verification, is the first coefficient of the shear wave attenuation velocity of the f-th model verification, so the shear wave prediction verification index can be simplified as: ;when , that is, when the shear wave attenuation velocity coefficient is not less than the reference shear wave attenuation velocity coefficient, the shear wave attenuation velocity is too fast, and the signal error received by the three-component detector is too large, which has no practical research significance. Therefore, the shear wave prediction verification index is directly zero; when , that is, when the shear wave attenuation velocity coefficient is less than the reference shear wave attenuation velocity coefficient, such as Figure 3 As shown, it is a statistical graph of the changes in the shear wave prediction verification index provided in an embodiment of the present application. It can be seen from the graph that the first shear wave velocity coefficient is positively correlated with the shear wave prediction verification index, and the first shear wave attenuation velocity coefficient is negatively correlated with the shear wave prediction verification index. When the shear wave attenuation velocity coefficient is within the reference shear wave attenuation velocity coefficient range, the error caused by shear wave attenuation is very small. When the first shear wave velocity coefficient increases, the shear wave prediction verification index increases, and the drilling vs30 data predicted by the shear wave prediction engineering model becomes more accurate. When the first shear wave attenuation velocity coefficient increases, the shear wave prediction verification index decreases, and the drilling vs30 data predicted by the shear wave prediction engineering model becomes more error-prone. The digitization of the accuracy of the prediction of the shear wave prediction engineering model is realized, and the query efficiency of the shear wave prediction of the engineering site is further improved.
[0049] It should be understood that the reference shear wave velocity coefficient is expressed by taking the average value after statistics of the shear wave velocity coefficients verified by the model, and the reference shear wave attenuation velocity coefficient is expressed by taking the average value after statistics of the shear wave attenuation velocity coefficients verified by the model; the shear wave velocity reference deviation is expressed by taking the average value after statistics of the shear wave velocity verified by the deviation calculation formula, and the shear wave attenuation velocity reference deviation is expressed by taking the average value after statistics of the shear wave attenuation velocity verified by the model; the deviation calculation formula is: , where S is the reference deviation value, i is the data number, , n is the total number of data, is the data value of the i-th data, is the average value of the data.
[0050] Furthermore, the specific process of model optimization is as follows: selecting the optimal hyperparameters of the shear wave prediction engineering model through the Bayesian optimization method in combination with the shear wave prediction verification index; obtaining historical engineering feature data and combining and cross-linking them to form first historical engineering feature data, normalizing the first historical engineering feature data, and optimizing the first historical engineering feature data whose skewness distribution exceeds the skewness threshold range to obtain second historical engineering feature data; using the obtained second historical engineering feature data to train the shear wave prediction engineering model until the obtained shear wave model optimization index is not less than the reference shear wave model optimization index, and the shear wave prediction engineering model is recorded as the first shear wave prediction engineering model, and the shear wave model optimization index is used to describe the optimization degree of the shear wave prediction engineering model.
[0051] In this embodiment, the hyperparameters include the step size of parameter update, the penalty coefficient of regularization, the number of historical engineering data, the number of iterations of model training, the number of neurons in the hidden layer and the optimizer momentum; the steps of the Bayesian optimization method are: taking the shear wave prediction verification index as the objective function, the input of the function is the hyperparameter configuration, and the output is the performance indicator of the shear wave prediction engineering model under the given hyperparameter configuration, defining the hyperparameter space and selecting the Gaussian process to estimate the objective function, the hyperparameter space is defined by determining the value range of each hyperparameter, the Gaussian process can capture the uncertainty and correlation of the function, and some initial sampling is performed in the hyperparameter space to construct an initial estimate of the Gaussian process and then update the Gaussian process, and the Gaussian process is used to predict the objective function value and its uncertainty at each point in the hyperparameter space to obtain the optimal hyperparameter; combining the historical engineering feature data, for example, combining two related features A multiplication operation is performed to cross the historical engineering feature data to capture the complex relationship between features using a crossover method; the purpose of normalization is to improve the convergence speed and performance of the shear wave prediction engineering model; the purpose of logarithmic transformation is to improve the performance of the shear wave prediction engineering model; normalization is used to make the first historical engineering feature data have the same scale and distribution, the skewness distribution is the distribution of the first historical engineering feature data, the skewness threshold range is the reference range of the skewness distribution set by the preset personnel, data optimization includes logarithmic transformation and polynomial expansion, the logarithmic transformation is used to make the first historical engineering feature data closer to the normal distribution, and the polynomial expansion is used to increase the shear wave prediction engineering model's ability to fit nonlinear relationships; the reference shear wave model optimization index is set by the preset personnel; the optimization efficiency of the shear wave prediction engineering model is improved, and the query efficiency of the shear wave prediction of the engineering site is further improved.
[0052] Furthermore, the method for obtaining the shear wave model optimization index is as follows: extract the model optimization index of the shear wave prediction engineering model from the model optimization process, the model optimization index includes complexity and data distribution, the complexity is the number of parameters that need to be trained for the shear wave prediction engineering model, and the data distribution is the degree of dispersion of the data relative to the data standard deviation; obtain the model optimization score according to the extracted model optimization index, the model optimization score includes the complexity score and the data distribution score, the complexity score is the difference between the complexity and the corresponding reference value, and the data distribution score is the difference between the data distribution and its corresponding reference value; obtain the shear wave model optimization index of the shear wave prediction engineering model according to the obtained model optimization score.
[0053] In this embodiment, the shear wave model optimization index is calculated using the following formula:
[0054] ;
[0055] In the formula, Optimize the exponent for the shear wave model, is the complexity score, is the complexity score reference deviation, is the data distribution score, is the reference deviation of the data distribution score.
[0056] Specifically, the expression of the complexity score is ,in, For complexity, is the reference complexity; the expression of the data distribution score is ,in, is the data distribution, The reference data distribution degree is used to realize the numerical optimization degree of the shear wave prediction engineering model, and further improve the query efficiency of the shear wave prediction of the engineering site.
[0057] It should be understood that the reference complexity is expressed by taking the average value after statistically analyzing the complexity of the shear wave prediction engineering model, and the reference data distribution score is expressed by taking the average value after statistically analyzing the data distribution score of the shear wave prediction engineering model; the complexity reference deviation is expressed by the deviation calculation formula after statistically analyzing the complexity of the shear wave prediction engineering model, and the data distribution score reference deviation is expressed by the deviation calculation formula after statistically analyzing the data distribution score of the shear wave prediction engineering model; the deviation calculation formula is: , where S is the reference deviation value, i is the data number, , n is the total number of data, is the data value of the i-th data, is the average value of the data.
[0058] Furthermore, the first shear wave prediction engineering model is used to obtain the predicted drilling vs30 data. The specific method for obtaining the predicted drilling vs30 data is as follows: the first shear wave prediction engineering model includes a constant velocity prediction model and a velocity prediction model; the method for calculating the predicted drilling vs30 data through the constant velocity prediction model is as follows:
[0059] ;
[0060] In the formula, e is a natural constant, p is the number of the site to be predicted, , P is the total number of sites to be predicted, is the predicted drilling vs30 data of the pth site to be predicted, is the soil layer depth of the pth site to be predicted, is the pth site to be predicted Shear wave travel time at soil depth, is the pth site to be predicted The instantaneous shear wave velocity at the depth of the soil layer; the method of predicting the borehole vs30 data through the velocity prediction model is:
[0061] ;
[0062] In the formula, p is the number of the site to be predicted, , P is the total number of sites to be predicted, is the predicted drilling vs30 data of the pth site to be predicted, is the first regression coefficient, is the second regression coefficient, is the soil layer depth of the pth site to be predicted, is the pth site to be predicted The average shear wave velocity at the depth of the soil layer.
[0063] In this embodiment, the shear wave travel time represents the propagation time of the shear wave in the soil layer. The time spent in the depth of the soil layer; In general, with the increase of depth, the shear wave velocity of the soil layer increases. Therefore, the constant velocity prediction model is more likely to underestimate the measured vs30 value of the site. However, when the depth of the wave velocity data is greater than 20m, the error of the vs30 prediction value shows a downward trend. Therefore, the constant velocity prediction model can be used to predict the vs30 data of the borehole; In the same area, the shear wave velocity of the soil layer below the surface changes relatively consistently. There is a relatively ideal linear or high-order empirical statistical relationship between the average shear wave velocity at different depths and vs30. The regression coefficient and By Depth The profile is calculated. When the error of the constant velocity prediction model is large, the velocity prediction model can be used to calculate the predicted drilling VS30 data; the accuracy of the predicted drilling VS30 data acquisition is improved, and the query efficiency of the shear wave prediction of the engineering site is further improved.
[0064] Furthermore, the specific steps of the application test are: obtaining the test drilling vs30 data corresponding to the input test engineering site parameters obtained by the first shear wave prediction engineering model within a preset time interval, and performing a performance test on the first shear wave prediction engineering model to obtain an application test performance index, and comparing the application test performance index with the performance threshold range, applying the first shear wave prediction engineering model that meets the performance threshold range to actual queries, and rebuilding the model that does not meet the performance threshold range.
[0065] In this embodiment, the performance threshold range is a performance interval set by a preset person. If the application test performance index meets the performance threshold range, it is within the performance interval; if it does not meet the range, it is outside the performance interval. This improves the application efficiency of the first shear wave prediction engineering model and further improves the query efficiency of shear wave prediction at the engineering site.
[0066] Furthermore, a method for obtaining an application test performance index is as follows: extract performance indicators of the first shear wave prediction engineering model within a preset time interval from the performance test process, the performance indicators including response time, number of concurrent users, resource utilization and error rate, the response time being the time from the first shear wave prediction engineering model receiving the site parameters to be predicted to the output of drilling vs30 data, the resource utilization being the usage of system resources such as CPU, memory, and disk I / O, the number of concurrent users being the number of users simultaneously online and interacting with the system, the error rate being the ratio of the number of requests in which errors occur to the total number of requests of the first shear wave prediction engineering model within a preset time interval; obtain a performance coefficient based on the extracted performance indicators, the performance coefficient including a response time coefficient and a concurrent user coefficient, the response time coefficient representing the ratio of the difference between the response time and its reference value and its reference deviation, the concurrent user coefficient representing the ratio of the difference between the number of concurrent users and its reference value and its reference deviation; obtain the application test performance index of the first shear wave prediction engineering model based on the obtained performance indicators and performance coefficient.
[0067] In this embodiment, the expression of the response time coefficient is: ,in, Predict the response time of the engineering model for the jth first shear wave, is the reference response time of the engineering model for predicting the jth first shear wave, is the response time reference deviation; the expression of the concurrent user coefficient is ,in, is the number of concurrent users of the jth first shear wave prediction engineering model, is the reference number of concurrent users of the jth first shear wave prediction engineering model, The reference deviation of the number of concurrent users is realized; the performance of the first shear wave prediction engineering model is digitized, and the query efficiency of the shear wave prediction of the engineering site is further improved.
[0068] It should be understood that the reference response time is expressed by taking the average value after counting the response time of the first shear wave prediction engineering model, and the reference number of concurrent users is expressed by taking the average value after counting the number of concurrent users of the first shear wave prediction engineering model; the reference deviation of the response time is expressed by counting the response time of the first shear wave prediction engineering model through the deviation calculation formula, and the reference deviation of the number of concurrent users is expressed by counting the number of concurrent users of the first shear wave prediction engineering model through the deviation calculation formula; the deviation calculation formula is: , where S is the reference deviation value, i is the data number, , n is the total number of data, is the data value of the i-th data, is the average value of the data.
[0069] Furthermore, the application test performance index is calculated using the following formula:
[0070] ;
[0071] Where j is the number of the first shear wave prediction engineering model, , J is the total number of the first shear wave prediction engineering models, The performance index of the application test of the engineering model for the jth first shear wave prediction, is the response time coefficient of the jth first shear wave prediction engineering model, is the concurrent user coefficient of the jth first shear wave prediction engineering model, The resource utilization of the engineering model for the jth first shear wave prediction, is the error rate of the engineering model for the jth first shear wave prediction, is the effective extreme value of the error rate.
[0072] In this embodiment, is the effective extreme value of the error rate, which is the maximum value allowed by the error rate. When the error rate is not less than the effective extreme value, the error rate is too high, resulting in a large error in the application test performance index, which has no practical significance. Therefore, the application test performance index is zero. When the error rate is less than the effective extreme value, the statistical table of the application test performance index is shown in Table 1:
[0073] Table 1 Statistics of changes in application test performance index
[0074] It can be seen from the above table that the response time coefficient, concurrent user coefficient and resource utilization are directly proportional to the application test performance index in a mathematical relationship, and the error rate is inversely proportional to the application test performance index. However, the application test performance index cannot be directly obtained through a single data, but needs to be obtained through a comprehensive analysis of the response time coefficient, concurrent user coefficient, resource utilization and error rate. For example, the response time coefficient of the first group of data is smaller than the response time coefficient of the fourth group of data, but the final application test performance index is larger than the fourth group, because the concurrent user coefficient, resource utilization and error rate also have an impact, so a comprehensive analysis is required; a more accurate evaluation of the performance of the first shear wave prediction engineering model is achieved, and the query efficiency of the shear wave prediction of the engineering site is further improved.
[0075] To summarize, the embodiment of the present application constructs a shear wave prediction engineering model by acquiring historical engineering data of the engineering site, and performs model verification on the constructed shear wave prediction engineering model to obtain a first shear wave prediction engineering model, then performs application testing on the first shear wave prediction engineering model to obtain an application test performance index, and finally applies the first shear wave prediction engineering model whose application test performance index meets the performance threshold range to actual queries, thereby achieving an improvement in the query efficiency of shear wave prediction of engineering sites, and effectively solving the problem of high query difficulty in shear wave prediction of engineering sites in the prior art.
[0076] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0077] The present invention is described with reference to flowcharts and / or block diagrams of systems, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0078] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0079] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0080] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.
[0081] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.
Claims
1. The engineering site drilling vs30 query system established based on shear wave prediction is characterized by: It includes data acquisition module, model building and optimization module and application testing module; Wherein, the data acquisition module is used to obtain historical engineering data of the engineering site, and the historical engineering data includes historical engineering site parameters and historical drilling vs30 data; The model building optimization module is used to use the collected historical engineering data to perform model training to build a shear wave prediction engineering model, and to perform model verification on the built shear wave prediction engineering model to obtain a first shear wave prediction engineering model; The specific process of model verification is as follows: Acquire prediction verification indicators of the shear wave prediction engineering model from the model verification process, wherein the prediction verification indicators include shear wave velocity and shear wave attenuation velocity; According to the obtained prediction verification index, the shear wave prediction engineering model is repeatedly calculated to obtain a shear wave prediction verification index and an average shear wave prediction verification index is obtained, wherein the shear wave prediction verification index is used to describe the accuracy of the prediction of the shear wave prediction engineering model; The average shear wave prediction verification index is compared with the reference average verification value. If it is not less than the reference average verification value, the shear wave prediction engineering model is integrated with the shear wave query system. Otherwise, the shear wave prediction engineering model is optimized according to the shear wave prediction verification index. The application test module is used to perform an application test on the first shear wave prediction engineering model to obtain an application test performance index, and to apply the first shear wave prediction engineering model whose application test performance index meets the performance threshold range to the actual query, wherein the application test performance index is used to describe the performance of the first shear wave prediction engineering model; The method for obtaining the shear wave prediction verification index is as follows: Acquire a prediction verification coefficient according to the prediction verification index, wherein the prediction verification coefficient includes a shear wave velocity coefficient and a shear wave attenuation velocity coefficient; The shear wave prediction verification index of the shear wave prediction engineering model is obtained according to the obtained prediction verification coefficient. The shear wave prediction verification index is calculated using the following formula: ; In the formula, e is a natural constant, f is the number of times the model is verified, , F is the total number of model validations, is the shear wave prediction validation index of the f-th model validation, is the shear wave velocity coefficient of the f-th model verification, is the reference shear wave velocity coefficient for the f-th model verification, is the shear wave attenuation velocity coefficient of the f-th model verification, is the reference shear wave attenuation velocity coefficient of the f-th model verification; The first shear wave prediction engineering model is used to obtain predicted drilling vs30 data. The specific method for obtaining the predicted drilling vs30 data is: The first shear wave prediction engineering model includes a constant velocity prediction model and a velocity prediction model; The predicted drilling vs30 data is calculated by the constant velocity prediction model as follows: ; In the formula, p is the number of the site to be predicted, , P is the total number of sites to be predicted, is the predicted drilling vs30 data of the pth site to be predicted, is the soil layer depth of the pth site to be predicted, is the pth site to be predicted Shear wave travel time at soil depth, is the pth site to be predicted The instantaneous shear wave velocity at the soil depth; The predicted drilling vs30 data is calculated by the velocity prediction model as follows: ; In the formula, p is the number of the site to be predicted, , P is the total number of sites to be predicted, is the predicted drilling vs30 data of the pth site to be predicted, is the first regression coefficient, is the second regression coefficient, is the soil layer depth of the pth site to be predicted, is the pth site to be predicted The average shear wave velocity at the depth of the soil layer; The specific steps of the application test are: Obtain the test borehole vs30 data corresponding to the input test engineering site parameters obtained by the first shear wave prediction engineering model within a preset time interval, perform a performance test on the first shear wave prediction engineering model to obtain an application test performance index, and compare the application test performance index with a performance threshold range, apply the first shear wave prediction engineering model that meets the performance threshold range to the actual query, and re-construct the model if it does not meet the performance threshold range; The method for obtaining the application test performance index is as follows: Extracting performance indicators of the first shear wave prediction engineering model within a preset time interval from the performance test process, wherein the performance indicators include response time, number of concurrent users, resource utilization and error rate; The resource utilization is the usage of CPU, memory, and disk I / O system resources, and the error rate is the ratio of the number of requests in which errors occur in the first shear wave prediction engineering model to the total number of requests within a preset time interval; Acquire a performance coefficient according to the extracted performance index, wherein the performance coefficient includes a response time coefficient and a concurrent user coefficient, wherein the response time coefficient represents the ratio of the difference between the response time and its reference value and its reference deviation, and the concurrent user coefficient represents the ratio of the difference between the number of concurrent users and its reference value and its reference deviation; Obtaining the application test performance index of the first shear wave prediction engineering model according to the obtained performance indicators and performance coefficients; The application test performance index is calculated using the following formula: ; Where j is the number of the first shear wave prediction engineering model, , J is the total number of the first shear wave prediction engineering models, The performance index of the application test of the engineering model for the jth first shear wave prediction, is the response time coefficient of the engineering model for predicting the jth first shear wave, is the concurrent user coefficient of the jth first shear wave prediction engineering model, The resource utilization of the engineering model for the jth first shear wave prediction, is the error rate of the engineering model for the jth first shear wave prediction, is the effective extreme value of the error rate.
2. The engineering site drilling vs30 query system based on shear wave prediction as claimed in claim 1 is characterized by: The data acquisition module includes a data acquisition unit and a preprocessing unit; The data collection unit is used to collect initial historical engineering data of the engineering site and transmit the collected initial historical engineering data to the preprocessing unit; The preprocessing unit is used to preprocess the received initial historical engineering data to obtain historical engineering data.
3. The engineering site drilling vs30 query system based on shear wave prediction as claimed in claim 1 is characterized in that: The specific process of model optimization is as follows: The optimal hyperparameters of the shear wave prediction engineering model are selected by Bayesian optimization method in combination with the shear wave prediction validation index; Acquire historical engineering feature data and combine and cross-form the first historical engineering feature data, normalize the first historical engineering feature data, and optimize the first historical engineering feature data whose skewness distribution exceeds the skewness threshold range to obtain the second historical engineering feature data; The shear wave prediction engineering model is trained using the obtained second historical engineering feature data until the obtained shear wave model optimization index is no less than the reference shear wave model optimization index, and the shear wave prediction engineering model is recorded as the first shear wave prediction engineering model. The shear wave model optimization index is used to describe the optimization degree of the shear wave prediction engineering model.
4. The engineering site drilling vs30 query system based on shear wave prediction as claimed in claim 3 is characterized in that: The method for obtaining the shear wave model optimization index is as follows: Extracting a model optimization index of a shear wave prediction engineering model from a model optimization process, wherein the model optimization index includes complexity and data distribution; Obtaining a model optimization score according to the extracted model optimization indicators, wherein the model optimization score includes a complexity score and a data distribution score; The shear wave model optimization index of the shear wave prediction engineering model is obtained according to the obtained model optimization score.
