Electric power engineering investigation method and system

Abstract

The invention discloses an electric power engineering investigation method and system. The method comprises the following steps: S1, acquiring soil test data and field data corresponding to the soil test data; s2, constructing a hierarchical optimization model according to preset pre-training data; s3, soil test data and field data are input into the layering optimization model, a layering optimization result is obtained, and the layering optimization result is used for representing the layering condition of the soil layer. The system comprises a server side, and a first client side, a second client side and a third client side which are respectively in communication connection with the server side. According to the method, data collection standardization, flow standardization and interaction real-time performance are achieved, the problem of noun definition difference of units is solved, the data collection accuracy and efficiency are improved, meanwhile, work flows of field data collection, field project acceptance, project management and planning and the like can be broken through, and multi-terminal collaboration is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of engineering investigation, and in particular to a method and system for electric power engineering investigation. Background Art

[0002] Engineering survey is a key task in the early stage of power engineering construction, which aims to provide comprehensive and accurate basic information such as topography and geology for the design and construction of the project, so as to ensure the scientificity and rationality of the engineering design. In addition, it can also effectively guide the subsequent construction and allow construction personnel to be familiar with the actual situation on site in advance, so as to formulate targeted construction preparation plans, such as foundation construction plans, earthwork allocation plans, etc., to provide strong support for construction safety and quality.

[0003] In the existing power engineering survey operations, there are problems such as low standardization of data collection, non-standard collection process, lack of real-time collection interaction, and deviations in the definition of professional terms in various units, which leads to complex data collection. At the same time, traditional survey methods are difficult to achieve efficient data transmission and distribution and storage, and have deficiencies in security, confidentiality, system operation and maintenance, etc. Summary of the invention

[0004] The present invention provides a comprehensive, efficient, safe and adaptable electric power engineering survey method and system, which can solve at least one of the above technical problems.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0006] A method for power engineering survey comprises the following steps:

[0007] S1. Obtain geotechnical test data and field data corresponding to the geotechnical test data;

[0008] S2. Build a hierarchical optimization model based on the preset pre-training data;

[0009] S3. Input the geotechnical test data and the field data into the stratified optimization model to obtain stratified optimization results, which are used to characterize the stratification of soil layers.

[0010] Furthermore, in S2, the pre-training data includes pore pressure static penetration data and corresponding drilling data, the hierarchical optimization model includes a soil classification sub-model, a boundary recognition sub-model and a 3D interpolation sub-model, and the hierarchical optimization model construction process includes:

[0011] S2.1. Remove the pore pressure static penetration data near the soil layer boundary and determine the affected area according to different soil types;

[0012] S2.2, after removing the boundary influence area, identify the homogeneous soil unit of preset size by calculating the coefficient of variation of normalized cone parameter and soil behavior type index;

[0013] Q tn is the normalized cone parameter, and the calculation formula is:

[0014]

[0015] Among them, p a is the atmospheric pressure, p a =101.3kPa,σ v0 is the total cover pressure, σ′ vo =σ vo -u 0 ,u 0 is the hydrostatic pore pressure, q t is the corrected cone tip resistance;

[0016] I c is the soil behavior type index, and the calculation formula is:

[0017]

[0018] Among them, F r is another normalized cone parameter, F r =f s / (q t -σ v0 ), f s is the sleeve friction;

[0019] Normalized cone parameter Q tn The coefficient of variation COV(logQ tn )for:

[0020]

[0021] Soil Behavior Type Index I c The coefficient of variation COV(I c )for:

[0022]

[0023] Normalized cone parameter Q tn The coefficient of variation COV(logQ tn ) and soil behavior type index I c The coefficient of variation COV(I c ) corresponds to the pore pressure static penetration data, and the COV (logQ tn ) and COV(I c) is less than a preset coefficient of variation value, and the borehole data corresponding to the pore pressure static penetration test data is used to obtain a homogeneous soil unit;

[0024] S2.3. Confirm the input features of the soil classification sub-model and build the soil classification sub-model based on the input features and pre-training data. The input features include:

[0025] Normalized cone parameter Q tn 、F r , B q , the original cone parameter q t , R f 、u 2 , stress-related parameter σ v0 、u 0 , as well as the standard deviation and local deviation of each normalized cone parameter;

[0026] The local deviation calculation formula is:

[0027] The formula for calculating standard deviation is:

[0028] Among them, x i is the normalized cone parameter of the ith sampling point in the homogeneous soil unit, n is the total number of sampling points in the homogeneous soil unit;

[0029] S2.4, evaluate the soil classification sub-model and optimize it based on the evaluation results;

[0030] Among them, the evaluation indicators for evaluating the soil classification sub-model include:

[0031] Overall accuracy

[0032] Single class accuracy

[0033] kappa coefficient

[0034] Among them, N i is the total number of homogeneous soil units in a single soil type, n i is the number of correctly classified homogeneous soil units in a single soil type, P o and P e Observed and expected chances of agreement, respectively, based on the evaluation metrics for the combination of input features and the range of coefficients of variation of the Normalized Cone Parameter and Soil Behavior Type Index adjusted to complete.

[0035] Furthermore, the hierarchical optimization model building process also includes:

[0036] S2.5. Construct a boundary identification sub-model based on the wavelet transform modulus maximum method to identify the soil behavior type index I c The wavelet transform modulus maximum is used to locate the boundary;

[0037] S2.6, by scanning Q tn The curve uses a moving window to scan downward from the initial boundary. When the Q at the bottom of the moving window tn When the value is not greater than the top, the transition zone is considered to be over, and the top position of the moving window is used as the corrected boundary;

[0038] S2.7. Construct a 3D interpolation sub-model. The 3D interpolation sub-model is used to process the soil classification results output by the soil classification sub-model to obtain the final hierarchical optimization results.

[0039] Furthermore, constructing the 3D interpolation sub-model in S2.7 further includes:

[0040] S2.7.1. Establish a grid based on the sampling point distribution of the pore pressure static penetration data, and the grid point spacing is close to the sampling interval of the horizontal pore pressure static penetration data;

[0041] S2.7.2. Use the pre-trained data to train the 3D interpolation sub-model, initially select a diffusion coefficient with a small value, predict the soil type at the grid point, then use the predicted results and the same diffusion coefficient to train a new model to predict the soil type at the pore pressure static penetration test sampling point, and calculate the difference between the predicted and original results;

[0042] S2.7.3. Repeat the above process, gradually increase the diffusion coefficient, draw a relationship graph between the correct recognition rate and the diffusion coefficient, and finally determine the optimal diffusion coefficient to complete the construction of the 3D interpolation sub-model.

[0043] Furthermore, the S3 further includes:

[0044] S3.1, inputting the geotechnical test data and the field data into the hierarchical optimization model, and obtaining confidence indicators corresponding to the geotechnical test data and the field data respectively;

[0045] S3.2. Based on the criterion that the confidence index is greater than a preset confidence threshold, the geotechnical test data and the field data are screened, and the screened geotechnical test data and the field data are input into the hierarchical optimization model again to obtain the hierarchical optimization result.

[0046] A power engineering survey system, applicable to the power engineering survey method, comprising a server end and a first client end, a second client end and a third client end respectively connected to the server end in communication;

[0047] The first client is used to call the server to create an engineering project and configure project information for the engineering project. The first client is also used to obtain geotechnical test data and upload the geotechnical test data to the server;

[0048] The second client is used to obtain the project information from the server, and collect field data according to the project information and upload it to the server;

[0049] The third client is used to obtain the project information and the field data from the server, compare the project information and the field data according to a preset acceptance process to obtain an acceptance result, and upload the acceptance result to the server;

[0050] The server side analyzes and obtains layered data based on the project information, the field data, and the geotechnical test data corresponding to the field data.

[0051] Furthermore, the project information includes personnel information, planning information, data templates and a data dictionary corresponding to the data templates;

[0052] The personnel information is used to characterize the personnel involved in the engineering project, the planning information is used to characterize the situation of hole layout and sampling at the engineering survey site, the data template is used to characterize the data that needs to be collected during the field engineering survey of the engineering project, and the data dictionary is used to characterize the data content and fields with the smallest granularity in the data template.

[0053] Furthermore, the data fields included in the field data are determined according to the data dictionary, and the field data are used to characterize the situation at the engineering survey site.

[0054] Furthermore, the geotechnical test data is used to characterize data obtained after geotechnical testing of samples collected in the field, and the geotechnical test data corresponds to the field data.

[0055] Furthermore, it also includes a fourth client, which is communicatively connected to the server, and the server obtains visualization results based on the project information, the field data and the layered data. The fourth client is used to obtain the visualization results from the server and display them.

[0056] The beneficial effects of the present invention are embodied in:

[0057] 1. This power engineering survey method realizes the standardization of collected data, normalization of processes and real-time interaction through data templates and data dictionaries, solves the problem of differences in the definitions of terms in various units, and improves the accuracy and efficiency of data collection.

[0058] 2. This power engineering survey system places the main computing and storage parts in the cloud by setting up a server and multiple clients, and the client flexibly adopts a variety of forms, so that the system can open up work processes such as field data collection, field engineering acceptance, engineering management and planning, and realize multi-terminal collaboration. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] The drawings described herein are used to provide further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute improper limitations on the present application.

[0060] Figure 1 It is a schematic diagram of the overall process of the electric power engineering survey method according to an embodiment of the present invention.

[0061] Figure 2 It is a structural diagram of a hierarchical optimization model according to an embodiment of the present invention.

[0062] Figure 3 It is a schematic diagram of the overall structure of a power engineering survey system according to an embodiment of the present invention.

[0063] Figure 4 It is a schematic diagram of the overall structure of a power engineering survey system according to another embodiment of the present invention.

[0064] Figure 5 It is a schematic diagram of the operation and maintenance framework of the server side provided by a specific example of the present invention.

[0065] Figure 6 It is a schematic diagram of the relationship between a data dictionary and a data template provided by a specific example of the present invention.

[0066] Figure 7 It is a structural block diagram of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0067] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. In the absence of conflict, the embodiments in this application and the features in the embodiments can be combined with each other. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0068] It should be noted that if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or suggesting their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the meaning of "and / or" appearing in the full text includes three parallel schemes. Taking "A and / or B" as an example, it includes scheme A, or scheme B, or a scheme that satisfies both A and B. In addition, "multiple" refers to more than two.

[0069] See also Figure 1 The present invention provides a method for power engineering survey, comprising the following steps:

[0070] S1. Obtain geotechnical test data and field data corresponding to the geotechnical test data;

[0071] S2. Build a hierarchical optimization model based on the preset pre-training data;

[0072] S3. Input the geotechnical test data and the field data into the stratified optimization model to obtain stratified optimization results, which are used to characterize the stratification of soil layers.

[0073] It is understandable that in the prior art, samples collected on site need to be subjected to geotechnical tests to obtain engineering characteristic indicators such as physical properties and mechanical properties, and soil stratification is performed based on these indicators, and finally the soil stratification situation at the work site is learned. However, due to the complexity of quality control factors in actual engineering, the samples sent for testing may not correspond to the sampling location on site, which leads to problems with the validity of the geotechnical test results. In addition, the indicators obtained based on geotechnical tests often require a lot of time for manual analysis to obtain the final stratification results, which also leads to low work efficiency. Therefore, the present invention introduces a stratified optimization model, which can judge whether the geotechnical test data corresponds to the field data on site based on a large amount of pre-training data to improve the validity of the results. In addition, the stratified optimization model can automatically calculate the stratification results, thereby greatly improving the efficiency of calculating soil stratification.

[0074] It can be understood that the data processing method of the present invention can determine whether the geotechnical test data corresponds to the on-site field data to improve the validity of the results. In addition, it can automatically calculate the stratification situation corresponding to the sample based on the geotechnical test data. Combined with the project information corresponding to multiple samples (such as the hole sampling positions corresponding to the samples) and field data, it can calculate and present the overall stratification situation in the working area, that is, obtain a visual stratification result.

[0075] In a specific example, the system will answer the relationship between the test data and depth in the current area based on experience. When the geotechnical test data is obviously different from the field data, the user will be reminded to pay attention to the authenticity of the data here.

[0076] See also Figure 2 , further, in S2, the pre-training data includes pore pressure static penetration data and corresponding drilling data, the hierarchical optimization model includes a soil classification sub-model, a boundary recognition sub-model and a 3D interpolation sub-model, and the hierarchical optimization model construction process includes:

[0077] S2.1. Remove the pore pressure static penetration data near the soil layer boundary and determine the affected area according to different soil types;

[0078] S2.2, after removing the boundary influence area, identify the homogeneous soil unit of preset size by calculating the coefficient of variation of normalized cone parameter and soil behavior type index;

[0079] Q tn is the normalized cone parameter, and the calculation formula is:

[0080]

[0081] Among them, p a is the atmospheric pressure, p a =101.3kPa,σ v0 is the total cover pressure, σ′ vo =σ vo -u 0 ,u 0 is the hydrostatic pore pressure, q t is the corrected cone tip resistance;

[0082] I c is the soil behavior type index, and the calculation formula is:

[0083]

[0084] Among them, F r is another normalized cone parameter, F r =f s / (q t -σ v0 ), f s is the sleeve friction;

[0085] Normalized cone parameter Q tn The coefficient of variation COV(logQ tn )for:

[0086]

[0087] Soil Behavior Type Index I c The coefficient of variation COV(I c )for:

[0088]

[0089] Normalized cone parameter Q tn The coefficient of variation COV(logQ tn ) and soil behavior type index I c The coefficient of variation COV(I c ) corresponds to the pore pressure static penetration data, and the COV (logQ tn ) and COV(I c ) is less than a preset coefficient of variation value, and the borehole data corresponding to the pore pressure static penetration test data is used to obtain a homogeneous soil unit;

[0090] S2.3. Confirm the input features of the soil classification sub-model and build the soil classification sub-model based on the input features and pre-training data. The input features include:

[0091] Normalized cone parameter Q tn 、F r , B q , the original cone parameter q t , R f 、u 2 , stress-related parameter σ v0 、u 0 , as well as the standard deviation and local deviation of each normalized cone parameter;

[0092] The local deviation calculation formula is:

[0093] The formula for calculating standard deviation is:

[0094] Among them, x i is the normalized cone parameter of the ith sampling point in the homogeneous soil unit, n is the total number of sampling points in the homogeneous soil unit;

[0095] S2.4, evaluate the soil classification sub-model and optimize it based on the evaluation results;

[0096] Among them, the evaluation indicators for evaluating the soil classification sub-model include:

[0097] Overall accuracy

[0098] Single class accuracy

[0099] kappa coefficient

[0100] Among them, Ni is the total number of homogeneous soil units in a single soil type, n i is the number of correctly classified homogeneous soil units in a single soil type, P o and P e Observed and expected chances of agreement, respectively, based on the evaluation metrics for the combination of input features and the range of coefficients of variation of the Normalized Cone Parameter and Soil Behavior Type Index adjusted to complete.

[0101] In a specific example, after removing the boundary influence area, a 0.3 m long homogeneous soil unit is identified and logQ is required rn and I C The coefficient of variation is less than 0.1.

[0102] It is understandable that although the current model has identified a series of input features, different combinations can be further explored or other possible relevant pore pressure static penetration test data features can be added. For example, it is possible to study whether some parameters related to the characteristics of the soil particle distribution curve can be introduced. If some features are found to have potential advantages in distinguishing different soil types, they can be included in the input features. By comparing the performance of the model under different combinations of input features, a better set of input features can be determined.

[0103] In a specific example, the random forest algorithm is used to deal with the imbalance of training data categories in the soil classification sub-model. Specifically, by sampling the training data with replacement, multiple sub-datasets are constructed, and each sub-dataset is used to train a decision tree. For the construction of each decision tree, when the node is split, a part of the features are randomly selected from the input features for evaluation, and the best split features and split points are selected so that the decision tree can classify the data. In this way, each decision tree learns the different features and patterns of the data, thereby improving the model's ability to recognize different soil types, especially for categories with a small number of samples, the classification accuracy can be improved through the comprehensive judgment of multiple decision trees.

[0104] It is understandable that for the random forest algorithm, the number of trees and the number of features selected at each node are key parameters. Increasing the number of trees may improve the stability and accuracy of the model, but it will also increase the computational cost and training time. Adjusting the number of features selected at each node will affect the diversity and generalization ability of the model. Through methods such as cross-validation, the model can be trained and the performance can be evaluated under different parameter settings to find more appropriate parameter values. For example, you can try to adjust the number of trees between 200-500 in steps of 50, and the number of features selected at each node between 3-7, observe the changing trend of the model performance, and determine the optimal parameter combination.

[0105] It is understandable that the coefficient of variation of the normalized cone parameter and the soil behavior type index of the homogeneous soil unit can be appropriately adjusted. If the threshold is appropriately relaxed, more data may be included, but some noise data may be introduced; if the threshold is tightened, the amount of data may be reduced but the homogeneity of the data may be improved. By comparing the performance of the model under different thresholds, the optimal data processing parameters can be determined. At the same time, the influence area range of removing data near the boundary of the soil layer can also be further optimized according to different soil types. If it is found that some soil types do not work well under the current influence area setting, their boundary range can be appropriately adjusted to obtain data that better represents the characteristics of the soil type for model training.

[0106] It is understandable that the trained model can be evaluated using a separate validation set data, and the performance of the model can be measured by calculating indicators such as overall accuracy (OA), single-class accuracy (SA), and kappa coefficient (K). If the model performance is not ideal, the model parameters (such as the number of trees, the number of features selected for each node) may be further adjusted or the data may be further processed (such as re-screening features, adjusting the data sampling method, etc.), and then trained and evaluated again until the model performance reaches a satisfactory effect. Through continuous iteration and optimization, the random forest model can effectively deal with the class imbalance problem in soil classification and accurately classify the soil.

[0107] It can be understood that by setting the soil classification sub-model, the soil classification results can be obtained in advance, so that based on the results, clearer and more accurate soil boundaries can be obtained through the boundary recognition sub-model, and finally a more accurate layered optimization result can be predicted through the 3D interpolation model.

[0108] See also Figure 2 In this embodiment, the hierarchical optimization model building process further includes:

[0109] S2.5. Construct a boundary identification sub-model based on the wavelet transform modulus maximum method to identify the soil behavior type index I c The wavelet transform modulus maximum is used to locate the boundary;

[0110] S2.6, by scanning Q tn The curve uses a moving window to scan downward from the initial boundary. When the Q at the bottom of the moving window tn When the value is not greater than the top, the transition zone is considered to be over, and the top position of the moving window is used as the corrected boundary;

[0111] S2.7. Construct a 3D interpolation sub-model. The 3D interpolation sub-model is used to process the soil classification results output by the soil classification sub-model to obtain the final hierarchical optimization results.

[0112] See also Figure 2 In this embodiment, constructing the 3D interpolation sub-model in S2.7 further includes:

[0113] S2.7.1. Establish a grid based on the sampling point distribution of the pore pressure static penetration data, and the grid point spacing is close to the sampling interval of the horizontal pore pressure static penetration data;

[0114] S2.7.2. Use the pre-trained data to train the 3D interpolation sub-model, initially select a diffusion coefficient with a small value, predict the soil type at the grid point, then use the predicted results and the same diffusion coefficient to train a new model to predict the soil type at the pore pressure static penetration test sampling point, and calculate the difference between the predicted and original results;

[0115] S2.7.3. Repeat the above process, gradually increase the diffusion coefficient, draw a relationship graph between the correct recognition rate and the diffusion coefficient, and finally determine the optimal diffusion coefficient to complete the construction of the 3D interpolation sub-model.

[0116] In the radial base layer of the 3D interpolation sub-model, the distance dist = |X-IW(j)| (j = 1, ..., n) between the input vector X and the training vector IW(j) is calculated, and then the distance dist is adjusted according to the adjustment coefficient, n 1 = dist × b, radial base output

[0117] Wherein, variable X represents the input vector, IW(j) represents the training vector, n represents the number of training vectors, s represents the diffusion coefficient, b is the adjustment coefficient calculated based on s, and n 1 is the adjusted distance.

[0118] At the output layer, calculate n 2 =LW×a 1 , and then the prediction of soil type T is obtained by normalization through the linear transfer function;

[0119] Among them, LW is the weight matrix of the output layer, n 2 is the intermediate calculation result, and T is the final hierarchical optimization result.

[0120] Specifically, suppose the output layer has n neurons, and the corresponding weight vector is w = [w 1 ,w 2 ,…,w n ], input vector n 2 =[n 21 ,n 22 ,…,n 2n ];

[0121] First calculate T i =w i ×n 2i, (i=1,2,…,n), and then normalize it, let S=∑ i-1 T i , then the vector of predicted soil type T is expressed as

[0122] Finally, the final soil type is determined based on the matching degree between T and the preset soil type code. For example, if there are m soil types, each type corresponds to a specific coding vector c j , (j=1,2,…,m), by calculating T and each c j Some distance metric (such as Euclidean distance or cosine similarity, etc.), select the c with the smallest distance j The corresponding soil type is taken as the final soil type prediction result, that is, the hierarchical optimization result.

[0123] In this embodiment, S3 further includes:

[0124] S3.1, inputting the geotechnical test data and the field data into the hierarchical optimization model, and obtaining confidence indicators corresponding to the geotechnical test data and the field data respectively;

[0125] S3.2. Based on the criterion that the confidence index is greater than a preset confidence threshold, the geotechnical test data and the field data are screened, and the screened geotechnical test data and the field data are input into the hierarchical optimization model again to obtain the hierarchical optimization result.

[0126] Specifically, through data learning, we can understand the correspondence between geotechnical test data and field soil layer depth, such as the approximate range of parameters of soils at different depths in the same area. Generally speaking, bearing capacity tests, consolidation tests, shear tests, and particle analysis tests have a relatively obvious correlation in the same area. Therefore, the geotechnical test data involved in stratification calculation is a combination of a series of geotechnical test results that are highly correlated with soil stratification.

[0127] Specifically, the preset pre-training data includes a series of geotechnical test results and corresponding stratification results of areas with environments similar to the area.

[0128] Specifically, the confidence index calculation of geotechnical test data and field data can be calculated based on the following method:

[0129] ①, Based on correlation analysis: Calculate the correlation coefficient between the geotechnical test data and the relevant indicators in the field soil layer data, such as the Pearson correlation coefficient or the Spearman's rank correlation coefficient. The correlation coefficient can reflect the linear or nonlinear correlation between the two, and the value range is between -1 and 1. The closer the absolute value is to 1, the stronger the correlation is.

[0130] Specifically, for two variables X and Y, the calculation formula of the Pearson correlation coefficient r is:

[0131]

[0132] Where n is the number of samples, x i and i are the i-th observation values ​​of variables X and Y, respectively. and are the means of variables X and Y respectively.

[0133] Specifically, in actual calculations, statistical software or programming languages ​​can be used to complete the calculation of the Pearson correlation coefficient. For example, in Python, numpy and pandas libraries can be used to read data, and then the pearsonr function in the scipy.stats library can be used to calculate the Pearson correlation coefficient.

[0134] It is understandable that the value range of the Pearson correlation coefficient is between -1 and 1. When r = 1, it means that the two variables are completely positively correlated, that is, when one variable increases, the other variable also increases strictly; when r = -1, it means that the two variables are completely negatively correlated, that is, when one variable increases, the other variable strictly decreases; when r = 0, it means that there is no linear correlation between the two variables. Generally speaking, the closer |r| is to 1, the stronger the correlation is; the closer |r| is to 0, the weaker the correlation is.

[0135] Specifically, for two variables X and Y, the calculation formula of Spearman's rank correlation coefficient ρ is:

[0136]

[0137] Where n is the number of samples, d i is the difference in rank between the i-th observation of variables X and Y.

[0138] Specifically, when calculating the Spearman rank correlation coefficient, you first need to convert the data of the two variables into corresponding rank data, and then calculate it according to the above formula. For example, in Python, you can use the spearmanr function in the scipy.stats library to calculate the Spearman rank correlation coefficient.

[0139] It is understandable that the Spearman rank correlation coefficient also ranges from -1 to 1, and its interpretation is similar to that of the Pearson correlation coefficient. However, unlike the Pearson correlation coefficient, the Spearman rank correlation coefficient does not depend on the specific values ​​of the variables, but is based on the rank order of the variables. Therefore, it can better reflect the monotonic relationship between variables and is also sensitive to nonlinear relationships.

[0140] In one embodiment, the confidence index calculation of the geotechnical test data and the field data is achieved by simultaneously calculating the Pearson correlation coefficient and the Spearman rank correlation coefficient and then selecting one of them after comparison.

[0141] After calculating the Pearson correlation coefficient and the Spearman rank correlation coefficient, the two can be compared. If the results of the two coefficients are relatively consistent, it means that the linear and monotonic relationship between the variables is relatively clear; if there is a large difference between the two, it is necessary to further analyze the data characteristics and the relationship between the variables. Generally speaking, when the data meets the normal distribution and there is a linear relationship between the variables, the Pearson correlation coefficient is more appropriate; when the data does not meet the normal distribution or there may be a nonlinear relationship between the variables, the Spearman rank correlation coefficient may more accurately reflect the degree of association between the variables.

[0142] ② Based on regression analysis: Establish a regression equation between geotechnical test data and field soil layer data, fit the data by least squares method, and obtain the regression coefficient and the goodness of fit index of the equation, such as the determination coefficient R 2 . R 2 It indicates the degree of fit of the regression equation to the data, and its value is between 0 and 1. The closer it is to 1, the better the fit effect is, that is, the higher the consistency between the geotechnical test data and the field soil layer data.

[0143] ③ Based on hypothesis testing: Propose a hypothesis about the relationship between geotechnical test data and field soil layer data, such as whether the means and variances of the two are equal, and use appropriate hypothesis testing methods, such as t-test, F-test, etc. to test. By calculating the test statistic and the corresponding p-value, it is determined whether the null hypothesis is rejected, thereby evaluating whether the difference between the two is significant.

[0144] It is understandable that the confidence index calculation based on correlation analysis may also be performed in other ways according to actual conditions, and the embodiment of the present invention does not specifically limit this.

[0145] It is understandable that the larger the sample size, the more reliable the calculated confidence result. Generally speaking, a sufficient sample size is required to ensure the validity of the statistical analysis.

[0146] Specifically, the field data includes the cone tip resistance and side wall resistance of the soil layer on the probe of the static sounding instrument measured by the static sounding instrument. The preset pre-training data also includes a series of stratification results and the corresponding cone tip resistance curve and side wall resistance curve. The stratification of the soil can be known by analyzing the changes in the cone tip resistance curve and the side wall resistance curve.

[0147] For example, different types of soils have different characteristics in their cone tip resistance curves and side wall resistance curves:

[0148] (1) Fill: When testing plain fill composed mainly of clay and mixed fill composed mainly of domestic waste, the values ​​change irregularly and often show sudden changes. Since they are located on the surface, they are easier to judge.

[0149] (2) Clay: The value of the cone tip resistance is relatively flat, with slow fluctuations and slight peaks in some areas. The value of the side wall resistance has a slight peak, which is on the right side of the value and at a large distance.

[0150] (3) Silty clay: The cone tip resistance value is relatively flat, with slow fluctuations and slight peaks in some areas. The side wall resistance value has slight peaks in some areas and is closer to the cone tip resistance value than clay. Most of it is located to the right of the cone tip resistance value. When the soil quality is uneven, it partially crosses the cone tip resistance value.

[0151] (4) Silt: The cone tip resistance value is relatively large, and the value is short sawtooth-shaped with gentle tooth peaks. The side wall resistance value is generally located to the right of the cone tip resistance value, with large local intervals, but occasionally it is also interspersed with the cone tip resistance value.

[0152] (5) Sandy soil: The cone tip resistance value is relatively large and presents a long sawtooth shape. The side wall resistance value is generally close to the cone tip resistance value, and the peak value is mostly located to the left of the cone tip resistance value. When the sandy soil particles are uneven, the cone tip resistance value and the side wall resistance value have more sharp teeth, and locally present an irregular, broken, large sawtooth shape.

[0153] See also Figure 3 , an embodiment of the present invention further provides a power engineering survey system, which is applicable to the power engineering survey method, and includes a server end and a first client end, a second client end and a third client end respectively connected to the server end in communication;

[0154] The first client is used to call the server to create an engineering project and configure project information for the engineering project. The first client is also used to obtain geotechnical test data and upload the geotechnical test data to the server;

[0155] The second client is used to obtain the project information from the server, and collect field data according to the project information and upload it to the server;

[0156] The third client is used to obtain the project information and the field data from the server, compare the project information and the field data according to a preset acceptance process to obtain an acceptance result, and upload the acceptance result to the server;

[0157] The server side analyzes and obtains layered data based on the project information, the field data, and the geotechnical test data corresponding to the field data.

[0158] In this embodiment, the project information includes personnel information, planning information, a data template and a data dictionary corresponding to the data template;

[0159] The personnel information is used to characterize the personnel involved in the engineering project, the planning information is used to characterize the situation of hole layout and sampling at the engineering survey site, the data template is used to characterize the data that needs to be collected during the field engineering survey of the engineering project, and the data dictionary is used to characterize the data content and fields with the smallest granularity in the data template.

[0160] In this embodiment, the data fields included in the field data are determined according to the data dictionary, and the field data is used to characterize the situation at the engineering survey site.

[0161] In this embodiment, the geotechnical test data is used to characterize the data obtained after the samples collected in the field undergo geotechnical tests, and the geotechnical test data corresponds to the field data.

[0162] In this embodiment, the server side is deployed on a private cloud server, and uses a distributed container management platform as an operating framework to deploy services in a container cluster manner; the data storage on the server side adopts an object storage method; the server side communicates with the first client, the second client, and the third client based on the session initiation protocol.

[0163] Specifically, the server-side service is deployed in a private cloud server, using the Kubernetes framework and Docker-based business deployment, and the entire business is hosted on the private cloud server. It includes Nginx (responsible for reverse proxy and load balancing), Kubernetes (container orchestration tool), business service container (using Docker), basic services (MySQL and object storage services) and security services (using self-built framework security policies, intranet access between services, and SSL encryption for communication).

[0164] Specifically, the communication protocol adopted between the server and the first client, the second client and the third client includes SIP (Session Initiation Protocol) protocol.

[0165] See also Figure 4 , an embodiment of the present invention further provides a power engineering survey system, which is applicable to the power engineering survey method, including a server end and a first client end, a second client end, a third client end and a fourth client end respectively connected to the server end in communication;

[0166] The server side analyzes and obtains layered data based on the project information, the field data, and the geotechnical test data corresponding to the field data, and the server side obtains visualization results based on the project information, the field data, and the layered data;

[0167] The first client is used to call the server to create an engineering project and configure project information for the engineering project. The first client is also used to obtain geotechnical test data and upload the geotechnical test data to the server;

[0168] The second client is used to obtain the project information from the server, and collect field data according to the project information and upload it to the server;

[0169] The third client is used to obtain the project information and the field data from the server, compare the project information and the field data according to a preset acceptance process to obtain an acceptance result, and upload the acceptance result to the server;

[0170] The fourth client is used to obtain the visualization result from the server and display it.

[0171] It can be understood that, in this embodiment, the server communicates with the first client, the second client, the third client and the fourth client based on the Session Initiation Protocol.

[0172] It can be understood that, in this embodiment, the communication protocol adopted between the server and the fourth client includes the SIP (Session Initiation Protocol) protocol.

[0173] Specifically, the visualization results include but are not limited to project type, project status, project progress, exploration hole statistics, sampling and in-situ test statistics, etc. The fourth client can be the same web client as the first client, or it can be an independent data screen to display relevant data on a large screen.

[0174] Specifically, the server uses data cache + object storage technology to achieve efficient data transmission and distribution and storage. Data is not displayed in real time, and field data is processed in a timed and summarized manner. After collection, it is sorted, archived, and cleaned for large-screen display. The display content is concise and clear and can be from simple to complex.

[0175] See also Figure 5 The present invention provides a specific example of a server-side operation and maintenance framework: using the Kubernetes (K8s) framework and docker-based business deployment, the entire business is hosted on a private cloud server.

[0176] The entire cluster includes the following modules:

[0177] Nginx: The entrance for traffic to enter the cluster. It is independently deployed on external machines and is mainly responsible for reverse proxy and load balancing, directing traffic to services within the cluster. Nginx has the advantages of being lightweight and high-performance, can be maintenance-free for a long time, and provides reliable forwarding services.

[0178] Kubernetes: A container orchestration tool for clusters, responsible for automated deployment, scaling, and management of cluster containers.

[0179] Business service container: The service runs in the container image, using Docker as the container service, and is used in conjunction with K8s to ensure excellent performance while being lightweight and flexible enough to reduce operation and maintenance costs.

[0180] Basic services: MySQL and object storage services, mainly used to store data and unformatted files.

[0181] Security service: We adopt a self-built framework security strategy. All services are accessed through the intranet, and business communications are carried out externally through a reverse proxy. Communications are encrypted using SSL. Management strategies are divided into different people and levels to ensure data security to the greatest extent possible.

[0182] MiniO: It is a high-performance, distributed object storage system. Users can expand storage capacity and performance by increasing the number of nodes. It supports multi-node deployment to form a distributed storage cluster.

[0183] In one embodiment, the first client is a web page, which is used to implement functions such as project management, data display, and achievement generation, and can display different menu bars after different roles log in based on the authorization mode of user-role-authority.

[0184] Specifically, the web version includes project overview, project planning, my project, data dictionary, template management, labor management, and user management modules.

[0185] It should be noted that, in actual operation, each unit has deviations in the definition of its own professional terms. The design and regulatory departments are less affected because they are not involved in actual field tasks. The field departments have some problems in this regard. Due to different standards, it is difficult to have the same system to meet the demands of all survey companies. For the same line laying project, some units call it "line engineering", while some units do not distinguish and still call it "ordinary engineering". This type of problem will make data collection extremely complicated. In order to solve this problem, the power engineering survey system of the present invention introduces the concepts of data dictionary and data template.

[0186] See also Figure 6 , the present invention explains the relationship between the data dictionary and the data template as follows:

[0187] The data dictionary defines the content and field name of each data item to be collected, such as "density", "color", "smell", etc. The data template combines the items in the data dictionary to form a collection template of a type. During collection, the corresponding template fields are displayed according to the different collection types.

[0188] In one embodiment, the second client is an APP client installed on a mobile electronic device, such as an APP client installed on a smart phone. The APP client is responsible for collecting field data, and the data is stored offline in a local database and can be uploaded to the cloud when the network is good. The cataloger obtains engineering planning data by scanning or searching, and quickly enters information based on the data dictionary and data template in the system during collection.

[0189] Specifically, the APP includes project selection, project details, drilling details, round entry, template collection, and local management modules.

[0190] Specifically, the logical structure of the data collected by the APP covers role permissions, engineering planning data (including tower location, drilling information, etc.), engineering tower location drilling data (detailed records of various collection parameters), etc. The physical structure maps the database table fields with Java object attributes through a specific resultMap configuration to ensure accurate data storage and reading.

[0191] In one embodiment, the third client is a small program client installed on a mobile electronic device. The small program client is mainly used for field acceptance work, and displays engineering collection and process information according to the acceptance process to assist users in comparison.

[0192] In one embodiment, the first client is also used to obtain geotechnical test data and upload the geotechnical test data to the server, the geotechnical test data is used to characterize the data obtained after the samples collected in the field undergo geotechnical testing, and the geotechnical test data corresponds to the field data.

[0193] It is understandable that geotechnical engineering tests, that is, geotechnical tests, require indoor tests on the collected rock and soil samples to obtain their physical properties, mechanical properties and other engineering characteristic indicators, provide data support for foundation design, and serve as a basis for soil stratification. In the power engineering survey system of the embodiment of the present invention, the samples collected at the field site need to undergo geotechnical tests to obtain their physical properties, mechanical properties and other engineering characteristic indicators, and soil stratification is performed based on these indicators, and finally the soil stratification situation at the work site is learned.

[0194] In one embodiment, the server obtains the hierarchical data according to the project information, the field data and the geotechnical test data corresponding to the field data.

[0195] Specifically, the power engineering survey system of the embodiment of the present invention can determine whether the geotechnical test data corresponds to the on-site field data to improve the validity of the results. In addition, it can automatically calculate the stratification situation corresponding to the sample based on the geotechnical test data. Combined with the project information corresponding to multiple samples (such as the hole sampling positions corresponding to the samples) and field data, it can calculate and present the overall stratification situation in the operating area, that is, obtain a visual stratification result.

[0196] In one embodiment, the electric power engineering survey system of the present invention can automatically generate CAD drawings based on the layering results, project information and field data to improve the design efficiency of the staff.

[0197] In one embodiment, the second client is provided with a local database for storing field data. When the second client communicates abnormally with the server, the field data collected by the second client is stored in the local database; after the second client and the server resume communication, the field data is uploaded to the server by the second client. Therefore, when the second client of the embodiment of the present invention is unable to communicate with the server due to poor network conditions, it can still collect data normally, thereby improving the stability of the system.

[0198] Specifically, the APP verifies user permissions, and only authorized catalogers can download project planning data and upload collection data. The web side is based on the RBAC (Role-Based Access Control) model, where users are associated with roles, roles are associated with permissions, and administrators can customize roles and bind users to complete authorization.

[0199] Furthermore, in a specific example, the collected data is stored offline in the local database of the APP, and the cataloger manually selects the tower position and borehole to be collected and uploads them to the server. When uploading the collected data, the cataloger's upload permission will be checked, and only the cataloger bound to this project can upload the collected data.

[0200] The electric power engineering survey system and application scenarios described in the embodiments of the present invention are intended to more clearly illustrate the technical solutions of the embodiments of the present invention, and do not constitute a limitation on the technical solutions provided by the embodiments of the present invention. Those skilled in the art will appreciate that with the evolution of the electric power engineering survey system and the emergence of new application scenarios, the technical solutions provided by the embodiments of the present invention are also applicable to similar technical problems.

[0201] It can be understood by those skilled in the art that Figure 3 and Figure 4 The electric power engineering survey system shown in the figure does not constitute a limitation on the embodiments of the present invention, and may include more or fewer components / modules than those shown in the figure, or a combination of certain components / modules, or a different arrangement of components / modules.

[0202] An embodiment of the present invention further provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes the steps of the above-mentioned power engineering survey method.

[0203] See also Figure 7 An embodiment of the present invention further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the above-mentioned power engineering survey method.

[0204] The embodiment of the present invention further provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute the steps of the above-mentioned power engineering survey method.

[0205] It is understandable that the system, device and storage medium provided in the embodiments of the present invention correspond to the method provided in the embodiments of the present invention, and the explanation, examples and beneficial effects of the relevant contents can refer to the corresponding parts in the above-mentioned power engineering survey method.

[0206] It should be noted that those skilled in the art can understand that all or part of the steps implemented in the embodiments of the present invention can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using hardware, it can be implemented in whole or in part in the form of purchasing standard parts or modified parts. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a DVD), or a semiconductor medium (eg, a solid state disk (SSD)).

[0207] In summary, the present invention provides a comprehensive, efficient, safe and adaptable electric power engineering survey method and system to solve the problems of low standardization of data collection, non-standard collection process, lack of real-time collection interaction, and complex data collection caused by deviations in the definition of professional terms of each unit when conducting electric power engineering survey operations in the prior art. The present invention places the main computing and storage parts in the cloud by setting up a server side and multiple clients, and the client flexibly adopts multiple forms, so that the system can open up the work processes such as field data collection, field engineering acceptance, engineering management and planning, and realize multi-terminal collaboration; at the same time, through data templates and data dictionaries, the standardization of collected data, normalization of processes and real-time interaction are realized, the problem of differences in the definition of terms of each unit is solved, and the accuracy and efficiency of data collection are improved.

[0208] It should be understood that the examples and implementation modes described herein are for illustrative purposes only and are not intended to limit the present invention. Those skilled in the art may make various modifications or changes based on the examples and implementation modes. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for surveying electric power engineering, characterized in that: The following steps are involved: S1. Obtain geotechnical test data and field data corresponding to the geotechnical test data; S2. Build a hierarchical optimization model based on the preset pre-training data; S3. Input the geotechnical test data and the field data into the stratified optimization model to obtain stratified optimization results, which are used to characterize the stratification of soil layers.

2. The electric power engineering survey method according to claim 1, characterized in that: In S2, the pre-training data includes pore pressure static penetration data and corresponding drilling data, the hierarchical optimization model includes a soil classification sub-model, a boundary recognition sub-model and a 3D interpolation sub-model, and the hierarchical optimization model construction process includes: S2.

1. Remove the pore pressure static penetration data near the soil layer boundary and determine the affected area according to different soil types; S2.2, after removing the boundary influence area, identify the homogeneous soil unit of preset size by calculating the coefficient of variation of the normalized cone parameter and the soil behavior type index; Q tn is the normalized cone parameter, and the calculation formula is: Among them, p a is the atmospheric pressure, p a =101.3kPa,σ v0 is the total cover pressure, σ′ vo =σ vo -u0, u0 is the hydrostatic pore pressure, q t is the corrected cone tip resistance; I c is the soil behavior type index, and the calculation formula is: Among them, F r is another normalized cone parameter, F r =f s / (q t -σ v0 ), f s is the sleeve friction; Normalized cone parameter Q tn The coefficient of variation COV(logQ tn )for: Soil Behavior Type Index I c The coefficient of variation COV(I c )for: Normalized cone parameter Q tn The coefficient of variation COV(logQ tn ) and soil behavior type index I c The coefficient of variation COV(I c ) corresponds to the pore pressure static penetration data, and the COV (logQ tn ) and COV(I c ) is less than a preset coefficient of variation value, and the borehole data corresponding to the pore pressure static penetration test data is used to obtain a homogeneous soil unit; S2.

3. Confirm the input features of the soil classification sub-model and build the soil classification sub-model based on the input features and pre-training data. The input features include: Normalized cone parameter Q tn 、F r , B q , the original cone parameter q t , R f , u2, stress-related parameters σ v0 , u0, and the standard deviation and local deviation of each normalized cone parameter; The local deviation calculation formula is: The formula for calculating standard deviation is: Among them, x i is the normalized cone parameter of the ith sampling point in the homogeneous soil unit, n is the total number of sampling points in the homogeneous soil unit; S2.4, evaluate the soil classification sub-model and optimize it based on the evaluation results; Among them, the evaluation indicators for evaluating the soil classification sub-model include: Overall accuracy Single class accuracy kappa coefficient Among them, N i is the total number of homogeneous soil units in a single soil type, n i is the number of correctly classified homogeneous soil units in a single soil type, P o and P e Observed and expected chances of agreement, respectively, based on the evaluation metrics for the combination of input features and the range of coefficients of variation of the Normalized Cone Parameter and Soil Behavior Type Index adjusted to complete.

3. The electric power engineering survey method according to claim 2, characterized in that: The hierarchical optimization model building process also includes: S2.

5. Construct a boundary identification sub-model based on the wavelet transform modulus maximum method to identify the soil behavior type index I c The wavelet transform modulus maximum is used to locate the boundary; S2.6, by scanning Q tn The curve uses a moving window to scan downward from the initial boundary. When the Q at the bottom of the moving window tn When the value is not greater than the top, the transition zone is considered to be over, and the top position of the moving window is used as the corrected boundary; S2.

7. Construct a 3D interpolation sub-model. The 3D interpolation sub-model is used to process the soil classification results output by the soil classification sub-model to obtain the final hierarchical optimization results.

4. The electric power engineering survey method according to claim 3, characterized in that: The constructing of the 3D interpolation sub-model in S2.7 further includes: S2.7.

1. Establish a grid based on the sampling point distribution of the pore pressure static penetration data, and the grid point spacing is close to the sampling interval of the horizontal pore pressure static penetration data; S2.7.

2. Use the pre-trained data to train the 3D interpolation sub-model, initially select a diffusion coefficient with a small value, predict the soil type at the grid point, then use the predicted results and the same diffusion coefficient to train a new model to predict the soil type at the pore pressure static penetration test sampling point, and calculate the difference between the predicted and original results; S2.7.

3. Repeat the above process, gradually increase the diffusion coefficient, draw a relationship graph between the correct recognition rate and the diffusion coefficient, and finally determine the optimal diffusion coefficient to complete the construction of the 3D interpolation sub-model.

5. The electric power engineering survey method according to claim 1, characterized in that: The S3 further comprises: S3.1, inputting the geotechnical test data and the field data into the hierarchical optimization model, and obtaining confidence indicators corresponding to the geotechnical test data and the field data respectively; S3.

2. Based on the criterion that the confidence index is greater than a preset confidence threshold, the geotechnical test data and the field data are screened, and the screened geotechnical test data and the field data are input into the hierarchical optimization model again to obtain the hierarchical optimization result.

6. A power engineering survey system, applicable to the power engineering survey method according to any one of claims 1 to 5, characterized in that: It includes a server end and a first client end, a second client end and a third client end which are respectively connected to the server end for communication; The first client is used to call the server to create an engineering project and configure project information for the engineering project. The first client is also used to obtain geotechnical test data and upload the geotechnical test data to the server; The second client is used to obtain the project information from the server, and collect field data according to the project information and upload it to the server; The third client is used to obtain the project information and the field data from the server, compare the project information and the field data according to a preset acceptance process to obtain an acceptance result, and upload the acceptance result to the server; The server side analyzes and obtains layered data based on the project information, the field data, and the geotechnical test data corresponding to the field data.

7. The electric power engineering survey system according to claim 6, characterized in that: The project information includes personnel information, planning information, data templates and data dictionaries corresponding to the data templates; The personnel information is used to characterize the personnel involved in the engineering project, the planning information is used to characterize the situation of hole layout and sampling at the engineering survey site, the data template is used to characterize the data that needs to be collected during the field engineering survey of the engineering project, and the data dictionary is used to characterize the data content and fields with the smallest granularity in the data template.

8. The electric power engineering survey system according to claim 7, characterized in that: The data fields included in the field data are determined according to the data dictionary, and the field data are used to characterize the situation at the engineering survey site.

9. The electric power engineering survey system according to claim 6, characterized in that: The geotechnical test data is used to characterize the data obtained after the samples collected in the field undergo geotechnical testing, and the geotechnical test data corresponds to the field data.

10. The electric power engineering survey system according to claim 6, characterized in that: It also includes a fourth client, which is communicatively connected to the server, and the server obtains visualization results based on the project information, the field data and the layered data. The fourth client is used to obtain the visualization results from the server and display them.

Images