A method for optimizing and regulating construction parameters of dynamic compaction for high fill subgrade

Abstract

The application relates to the technical field of roadbed construction, in particular to a high-filled roadbed dynamic compaction and pressure supplement construction parameter optimization regulation method, which comprises the following steps: according to ramming construction parameters, optimal ramming times and optimal loads are respectively obtained; based on the optimal ramming times and the optimal loads, each ramming area is differentially obtained, and a pressure supplement area is determined according to the settlement deviation and the position deviation of each ramming area; a multiple linear regression model is established by taking the settlement deviation and the position deviation as input variables and taking the load of pressure supplement ramming, the supplement ramming times and the interval time as output variables, and the to-be-optimized parameters are output; the primary and secondary influence order of each parameter in the to-be-optimized parameters is determined, and a weight distribution diagram is constructed according to the distribution position of the pressure supplement area; the ramming sequence at each position in the weight distribution diagram is determined, and the parameter optimization processing is completed. The working efficiency and scene adaptability during pressure supplement construction are improved.

Description

Technical Field

[0001] This invention relates to the field of roadbed construction technology, specifically a method for optimizing and controlling construction parameters of dynamic compaction and reinforcement in high embankment roadbeds. Background Technology

[0002] Dynamic compaction is a convenient, fast, economical, and environmentally friendly foundation treatment method, widely used in foundation treatment for various soil and rock types. When constructing high embankments, dynamic compaction is used for reinforcement, and its construction quality directly determines the long-term stability and bearing capacity of the subgrade. Traditional dynamic compaction parameters are often determined by construction personnel based on experience from similar projects, lacking objective judgment. Furthermore, neglecting the spatial arrangement of the compaction area and settlement deviations can lead to uneven settlement, increased construction costs, and low reinforcement efficiency.

[0003] For example, Chinese Patent Publication No. CN119577899A discloses a method and system for quality control of roadbed dynamic compaction construction based on the dynamic response of a tamping hammer. This control method includes the following steps: acquiring the acceleration signal of the tamping hammer during the current compaction cycle; processing the acceleration signal of the tamping hammer to obtain multiple key parameters for predicting the dynamic deformation modulus; the data processing includes sequential filtering, time-domain analysis, and frequency-domain analysis; based on the multiple key parameters and a pre-determined dynamic deformation modulus prediction model, obtaining the dynamic deformation modulus of the compaction point after the current compaction cycle; comparing the dynamic deformation modulus of the compaction point with the set design value of the dynamic deformation modulus, and based on the comparison result, sending a prompt to stop compaction or a prompt to continue compaction.

[0004] For example, Chinese Patent Publication No. CN114791767A discloses a virtual reality-based visualization management system for dynamic compaction foundations. The system includes: a foundation raw data acquisition unit, configured to acquire raw foundation data, which includes at least: area, surface unevenness, natural unit weight, dry unit weight, specific gravity, saturation, void ratio, porosity, liquid voids, and plastic limit; a data fusion unit, configured to use a preset data fusion model to fuse the acquired raw data to generate a data fusion result; and a foundation classification unit, configured to classify the foundation based on the data fusion result using preset classification rules to obtain foundation categories.

[0005] In existing technologies, the compaction process is controlled by the dynamic deformation modulus during compaction; or the compaction parameters are selected based on the difference between the current model and the ideal model under compaction analysis. However, existing technologies tend to be more theoretical engineering designs. When the settlement of the current compaction area is different, it is necessary to screen the compaction area in real time according to the compaction situation at different locations in order to determine the parameters that need to be optimized for additional pressure in the compaction area. Summary of the Invention

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a method for optimizing and controlling the construction parameters of dynamic compaction of high embankment subgrade, comprising: based on the subgrade embankment height, retrieving the compaction area at each compaction, and constructing compaction construction parameters based on the settlement, load and number of compactions corresponding to the compaction area.

[0007] Based on the obtained compaction construction parameters, the settlement-load relationship curve and the settlement-compaction number curve are obtained respectively, and the optimal number of compaction blows and the optimal load are determined.

[0008] Based on the optimal number of compaction blows and the optimal load, the differences between each compaction zone are calculated, and the additional compaction zone is determined by the settlement deviation and position deviation of each compaction zone.

[0009] Using the settlement deviation and location deviation corresponding to the compaction area as input variables, and the load, number of compaction cycles, and interval time of the compaction as output variables, a multiple linear regression model is established, and the parameters to be optimized are obtained by multiple linear regression calculation.

[0010] Based on the standardized regression coefficients of the multiple linear regression model, the order of primary and secondary influence of the parameters to be optimized is determined, and a weight distribution map is constructed based on the distribution location of the pressure compensation region.

[0011] Determine the tamping sequence at each location in the weight distribution map, and use the tamping sequence to complete the optimization process for each region.

[0012] The beneficial effects of this invention are as follows: First, this invention retrieves the compaction area based on the roadbed embankment height, and constructs compaction construction parameters by combining settlement, load, and number of compactions; at the same time, it completes parameter identification and recording according to the arrangement information and shape of each row of compaction areas, thereby strengthening the spatial correlation of parameters.

[0013] II. This invention calculates settlement deviation by using the difference between the foundation settlement corresponding to the optimal number of compaction cycles and load and the actual settlement; calculates the positional deviation of adjacent areas using the center coordinates and compaction radius of the compaction area; classifies areas and sets semantic tags for settlement / positional deviation, forming a tag combination; then constructs a settlement distribution matrix and decomposes it into sub-matrices according to deviation direction and area type; normalizes the weights of negative deviation sub-matrices to obtain settlement deviation priority weights; constructs a positional deviation matrix based on 8 neighborhoods and calculates the priority weights corresponding to positional deviations; and outputs a list of compaction areas in descending order of combined weights. The method of eliminating isolated anomalies ensures the data authenticity of the compaction areas, and the subsequent quantitative sorting according to weight normalization provides clear area priorities for subsequent processing, improving the targeting and specificity of compaction construction.

[0014] Third, this invention establishes a multiple linear regression model by using the settlement deviation and location deviation of the compaction area as input variables and the compaction load, frequency, and interval time as output variables to calculate the parameters to be optimized. Then, the regression model is re-validated using the predicted parameters to avoid conflicts between actual theory and field data. Subsequently, a multiple linear regression model is established with the compaction effect value as the dependent variable and the compaction load, frequency, and interval time as independent variables. The order of primary and secondary influence of each parameter is determined by standardizing the regression coefficients, further clarifying the control priority of each area, and determining the priority weight distribution of each point predicted under actual implementation, so as to improve the work efficiency during compaction construction.

[0015] Fourth, this invention determines the starting and ending points of each row by dividing the compaction area into rows according to their coordinates; sets the path arrangement information within each row based on the order of primary and secondary influence of parameters; connects adjacent rows through the shortest turning path to determine the compaction sequence; and selects methods for local adjustment and global topology optimization based on scenarios with consistent and inconsistent path directions to ensure the global sorting process of the current construction, enabling high-priority compaction areas to be constructed first, thereby improving the efficiency and adaptability of compaction construction. Attached Figure Description

[0016] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0017] Figure 1 This is a flowchart illustrating a method for optimizing and controlling construction parameters in dynamic compaction and reinforcement of high-fill roadbeds.

[0018] Figure 2 This is a flowchart illustrating step S3 of a method for optimizing and controlling construction parameters in dynamic compaction of high embankment roadbeds.

[0019] Figure 3 This is a flowchart illustrating step S4 of a method for optimizing and controlling construction parameters in dynamic compaction of high embankment roadbeds.

[0020] Figure 4 This is a flowchart of step S5 of a method for optimizing and controlling construction parameters of dynamic compaction and reinforcement in high embankment roadbeds.

[0021] Figure 5 This is a flowchart illustrating step S6 of a method for optimizing and controlling construction parameters in dynamic compaction of high embankment roadbeds. Detailed Implementation

[0022] The embodiments of the present invention are described in detail below. The embodiments described below are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention. Where specific techniques or conditions are not specified in the embodiments, they shall be performed in accordance with the techniques or conditions described in the literature in the art or in accordance with the product manual.

[0023] See Figure 1 A method for optimizing and controlling construction parameters of dynamic compaction for high embankment subgrade includes: S1, based on the subgrade embankment height, retrieving the compaction area at each compaction, and constructing compaction construction parameters based on the settlement, load and number of compactions corresponding to the compaction area.

[0024] S2. Based on the obtained compaction construction parameters, calculate the settlement-load relationship curve and the settlement-compaction number curve respectively, and determine the optimal number of compactions and the optimal load.

[0025] S3, based on the optimal number of compaction blows and the optimal load, calculates the differences in each compaction zone, and determines the additional compaction zone based on the settlement deviation and positional deviation of each compaction zone.

[0026] S4. Using the settlement deviation and location deviation corresponding to the compaction area as input variables, and the load, number of compaction cycles, and interval time of the compaction as output variables, a multiple linear regression model is established, and the parameters to be optimized are obtained by multiple linear regression calculation.

[0027] S5, based on the standardized regression coefficients of the multiple linear regression model, determines the order of primary and secondary influences of the parameters to be optimized, and constructs a weight distribution map based on the distribution location of the pressure compensation region.

[0028] S6, determine the tamping sequence at each location in the weight distribution map, and complete the optimization process of each region according to the tamping sequence.

[0029] The current compaction process is conducted in three passes: the first pass involves compaction at the initially selected spacing and number of passes; the second pass involves additional compaction in the middle of the first pass's compaction points; and the third pass involves low-energy compaction of certain areas. For data acquisition, a positioning sensor measures the settlement after each compaction pass, and the compaction degree is recorded using the sand cone method or a nuclear density meter. The load value is determined based on the weight of the dynamic compaction hammer. In terms of data processing, step S1 retrieves the compaction area for each pass and constructs construction parameters by combining the settlement, load, and number of compaction passes. Preliminary compaction parameters are determined by checking the dynamic compaction reinforcement height, number of layers, and the actual compaction area and number of passes for each layer. During additional compaction, the parameters are initially optimized based on the data from the first pass, and then the final additional compaction area and parameters are determined by combining the optimized parameters from the second pass.

[0030] In addition to obtaining the dynamic compaction construction parameters set above, it is also necessary to determine the soil stiffness, porosity, natural moisture content, and dry density of each filling layer through static cone penetration tests and heavy dynamic cone penetration tests. These parameters will be used as verification data after compaction to check whether the compaction meets the standards, such as compaction degree ≥96%, settlement difference stability value ≤5mm / pass, and porosity ≤22%.

[0031] The implementation method for constructing the compaction construction parameters in step S1 also includes: based on the arrangement information of each compaction area, identifying the parameters of the compaction areas in each row, and recording the compaction construction parameters corresponding to each compaction area according to the shape of the area corresponding to each compaction area. In this scheme, each compaction area is essentially an area formed by a compaction point after compaction, and each compaction area presents a specific shape, such as a circle or a square.

[0032] The method for determining the optimal number of tamping blows and the optimal load in step S2 includes: based on the settlement, number of tamping blows and load of different tamping areas, data fitting is performed using the settlement of each tamping blow to construct settlement-load relationship curves and settlement-number of tamping blows relationship curves.

[0033] The optimal load is extracted by using the rate of change of the slope between two consecutive points in the settlement-load relationship curve as a criterion. When constructing the settlement-load relationship curve, the horizontal axis represents the load and the vertical axis represents the settlement. The horizontal axis records the optimal load under multiple compaction blows by accumulating the load. By calculating the slope value of each point in the curve, when the rate of change of the slope value between two adjacent points is <5%, the corresponding load is the critical point of settlement stability. After that, it is difficult to have significant settlement even if the load is increased. The critical point at this time represents the output optimal load.

[0034] The optimal number of compaction blows is extracted by using the settlement difference between two consecutive points in the settlement-number of compaction blows curve as the criterion. The optimal number of compaction blows is determined by the difference between two adjacent settlement values, and the number of times with a settlement difference of less than 5 mm is selected as the optimal number of compaction blows.

[0035] After obtaining the optimal load and optimal number of compaction cycles under the current construction conditions, the compaction status of each compaction zone can be checked based on the settlement corresponding to the optimal load and optimal number of compaction cycles to identify locations with significant insufficient settlement, thereby helping to determine the areas that need additional compaction.

[0036] In one embodiment of the present invention, the number of tamping blows and the load obtained are used as basic optimization parameters under normal scenarios. The deviation between the actual final settlement of each tamping area and the ideal settlement is determined, thereby determining the deviation generated in each area. When calculating the current settlement deviation, it is expressed as a percentage value including positive and negative signs. That is, the part where the final settlement is less than the foundation settlement and the part that is significantly higher than the foundation settlement are identified, thereby determining the compaction status of each area at the preliminary design tamping point.

[0037] As for judging the positional deviation of the compaction area, it involves checking the spacing between adjacent compaction areas to prevent some compaction points from shifting position during actual compaction.

[0038] like Figure 2As shown, the method for determining the compaction zone in step S3 includes: S31, setting the foundation settlement based on the optimal number of compactions and the optimal load for the current compaction zone. This foundation settlement is the settlement corresponding to the optimal number of compactions and the optimal load, and will be used to compare the deviations of adjacent compaction zones. Based on the difference between the foundation settlement and the final settlement of the corresponding compaction zone, the settlement deviation of each compaction zone is obtained.

[0039] S32: Obtain the center coordinates and compaction radius of the current compaction area. Use the compaction radii between adjacent compaction areas to determine the positional deviation of the compaction area. At this point, the compaction radius represents the maximum radial range under optimal load and optimal number of compactions, with the soil compaction reaching 96%. This radius can be obtained by measuring the soil under the optimal load and optimal number of compactions obtained through test compaction in the current scenario. By comparing the effective compaction range, it can be determined whether there are gaps in the current compaction. This can lead to reinforcement blind spots, making the soil reinforcement unstable.

[0040] When setting this positional deviation, first calculate the distance between two adjacent compaction areas according to the center coordinates, and then subtract the compaction radius corresponding to the two compaction areas from this distance to determine whether there is a positional deviation in the compacted part that is not fully covered.

[0041] Since the compaction zones are distributed in rows and columns and there is a compaction pattern with intermediate insertion points, when calculating the positional deviation of the current compaction zone, the compaction zone corresponding to the current positional deviation will be marked, and the positional deviation in each direction will be marked in an array.

[0042] S33. Based on the distribution location of the current compaction area, divide each compaction area into area types, and set semantic labels according to the value range of settlement deviation and position deviation in various area types to determine the label combination form of each compaction area.

[0043] When setting the area type, select the stress concentration locations such as the roadbed slope toe, corner, edge, and above the soft soil interlayer as key areas. These areas have a high risk of instability and need to be identified and treated separately. For areas with uniform stress, such as the roadbed center and the internal homogeneous soil area, they will be regarded as regular areas. This determines the area type of the current compaction area. Then, the range of values ​​for position deviation and settlement deviation will be used as the basis for the additional compaction required in the compaction area.

[0044] For example, in critical areas, the settlement deviation can be divided into positive and negative deviation ranges. Negative deviation indicates that the compaction is different, while positive deviation indicates that there is excessive compression, that is, the actual settlement is much greater than the theoretical settlement. In this case, excessive settlement may indicate that the roadbed is over-compressed, and there is a risk of bulging.

[0045] Assuming that the acceptable settlement deviation for critical areas is defined as a range of [-10%, +5%], and using -15% and 10% as thresholds for differentiation, threshold ranges for slight and severe exceedances are set accordingly. For general areas, the acceptable settlement deviation is defined as a range of [-15%, +10%], with -20% and 15% selected as thresholds for differentiation.

[0046] As for the positional deviation, the acceptable value is set within the range of [0, 0.1m] in the critical area, and 0.2m is used as the dividing threshold to explain the slight and severe exceedances. In the regular area, the acceptable value is a positional deviation of less than 0.2m, and then 0.3m is selected as the dividing threshold to view the label description of each compaction area under the specific area type.

[0047] Meanwhile, when determining the label combination, the positional deviation and settlement deviation will be processed separately to match the direction of the two compaction zones corresponding to the positional deviation.

[0048] S34. Based on the label combination of each compaction area, the compaction areas are initially screened; for the compaction areas after the initial screening, the priority weights are determined by the settlement deviation and the position deviation, and the output list of compaction areas is determined.

[0049] During the initial screening, the semantic tags that meet the criteria are removed, and the remaining areas are arranged according to their positions.

[0050] When initially screening the compaction area, the implementation method also includes: removing the corresponding data when the semantic labels of settlement deviation and location deviation both meet the standards.

[0051] When there are substandard semantic tags for settlement deviation and location deviation, the corresponding area is regarded as the area to be verified.

[0052] For each area to be verified, determine whether the adjacent areas to be verified are all up to standard. If they are all up to standard, remove the corresponding data and use the remaining data as the initial screening area.

[0053] The initial screening requires removing regions that fully meet the criteria. For each problematic region, there must be at least one other region in its eight neighborhoods that is also anomalous. If all eight neighborhoods meet the criteria, the current data is considered an isolated outlier and needs to be removed to prevent noise from affecting subsequent calculations.

[0054] The implementation method for determining the label combination form of each compaction area in step S34 also includes: S341, when dealing with the settlement deviation, constructing a settlement distribution matrix based on the continuous distribution position of each compaction area, where each element in the matrix represents the settlement deviation at the corresponding position.

[0055] S342, decompose the settlement distribution matrix according to the direction of settlement deviation and the region type to obtain multiple settlement sub-matrices; the direction of settlement deviation represents positive deviation and negative deviation. It is necessary to determine the negative deviation region where the vibration compaction is insufficient and additional compaction is required, as well as the positive deviation region where the compression is excessive and no additional compaction is required, but other parameters need to be recorded.

[0056] S343 performs weight normalization on the settlement submatrix with negative deviation after decomposition to obtain the priority weight corresponding to the settlement deviation. The priority weight corresponding to the settlement deviation represents the normalized value of the settlement deviation in its matrix, which is used to quantify the priority of its regional pressure replenishment.

[0057] S344, when dealing with positional deviations, a positional deviation matrix is ​​constructed for the current compaction area according to its 8-neighborhood directions. The positional deviations between the current compaction area and its adjacent compaction areas in the positional deviation matrix are used as the priority weights for configuring the positional deviations of the current compaction area. Each of the adjacent compaction areas corresponding to the current compaction area in the eight directions (East / South / West / North / Southeast / Southwest / Northeast / Northwest) is combined into a separate data matrix in a 3×3 manner, and the positional deviations of these areas are superimposed on the elements in the middle of the matrix. The positional deviations ignore the influence of area type and emphasize the differences in spatial distribution. After normalizing the positional deviation matrix, the priority weights are set according to the eigenvalues ​​of all positional deviation matrices; that is, the ratio of the sum of the positional deviations of each positional deviation matrix to the sum of the positional deviations of all positional deviation matrices is used as the priority weight at this time, thus explaining the spatial clustering of positional deviations.

[0058] S345 synchronizes the priority weights of settlement deviation and location deviation to the corresponding compaction area to obtain the combined weight of each compaction area, and outputs the list of compaction areas in descending order of the value range of the combined weight.

[0059] The combined weights will be set based on the weighted sum of the two priority weights. Since the positional deviation represents a region and the settlement deviation represents a point, the settlement deviation is calculated by summing the weights according to the region corresponding to the positional deviation. After converting it into a region-level settlement deviation, the weights are normalized so that the two priority weights can correspond. Then, the priority weights of the settlement deviation and the positional deviation are weighted and summed. Weights of 0.7 and 0.3 can be used to obtain the final combined weights.

[0060] To highlight the spatial clustering effect of positional deviations, the combined weights of multiple regions can be adjusted based on the ratio of the difference between the standard deviation of a single region's positional deviation and the overall standard deviation. This results in a normalized ranking of the combined weights, which in turn emphasizes the priority of the pressure-replenishing regions.

[0061] It should be noted that step S3 will complete the preliminary screening of the compaction area based on the data of the first and second compaction under the working scenario of the dynamic compaction machine. The subsequent step S4 will determine the compaction parameters based on the data of the second compaction. The output data will be determined by the residual iteration method between the predicted value and the actual value. After optimizing the coordinates of the corresponding compaction points, the predicted data for the third compaction will be output. Step S5 will complete the weight configuration and the arrangement of the subsequent compaction implementation process based on the predicted data of the third compaction.

[0062] In one embodiment of the present invention, during the multiple linear regression calculation, all data are first normalized, and then the load, number of compaction cycles, and interval time of the additional compaction are used as parameters for independent calculation. A binary linear regression is constructed for each output variable to obtain the parameters to be optimized after the settlement deviation and position deviation occur in the current area.

[0063] For example ;in, This represents the output value of the regression equation for the i-th output variable, where i represents the index number of the output variable, and the value range is i=1, 2, 3, which correspond to the load, number of compaction blows, and interval time, respectively. This represents the constant term of the i-th output variable; This represents the regression coefficient of the i-th output variable with respect to the deviation of settlement. The regression coefficient of the i-th output variable with respect to the positional deviation can be obtained by the least squares method. This indicates a deviation in settlement. Indicates positional deviation; This represents the random error term of the i-th output variable.

[0064] After obtaining the predicted value of each output variable, the portion with a coefficient of determination greater than 0.8 is output, and the current parameter to be optimized is determined based on the region of this data. Then, the weight distribution corresponding to the compaction is determined based on the primary and secondary influence of the parameter to be optimized globally, thereby determining the order and allocation position of the compaction.

[0065] like Figure 3 As shown, when calculating the parameters to be optimized using multiple linear regression in step S4, the implementation method also includes: S41, combining the settlement deviation and position deviation of each replenishment area to obtain the predicted value of each replenishment area in the parameters to be optimized, and using the difference in the predicted values ​​of each replenishment area as the accuracy of the prediction.

[0066] S42, based on the difference in predicted values ​​of the parameters to be optimized, the model regression is reprocessed using the difference in predicted values ​​of the parameters to be optimized as part of the re-verification, and the output parameters to be optimized are obtained according to the regression processing of each model.

[0067] Steps S41 and S42 are the post-validation and iterative correction steps of the model, used to determine the difference between the predicted value and the actual value, and to determine whether there is a significant error under the preliminary multiple linear regression.

[0068] The predicted value represents the predicted value output for any pressure compensation area under multiple linear regression calculation. The difference in predicted values ​​will be used as the prediction deviation for a single area, and the deviation direction will be used to perform regression processing on the deviation of each area.

[0069] Next, the mean and standard deviation of the predicted value difference are examined, and data exceeding the mean ± 3 times the standard deviation are filtered out to remove outliers in the current regression analysis. Furthermore, the recalculated mean squared error must be less than the original mean squared error, and the coefficient of determination must be greater than the original coefficient of determination. Finally, the corresponding data is output. It should be noted that the outliers removed belong to historical data. The current model is trained using historical data, and the predicted values ​​are obtained by selecting the positional deviation and settlement deviation after the first compaction. The actual values ​​of the second compaction are used to calibrate the model, removing historical data that does not conform to the current scenario, to obtain historical data similar to the first and second compaction in the current scenario. Then, the positional deviation and settlement deviation during the second compaction are combined with historical data to predict the data for the third compaction, which is the set of parameters to be optimized under the pressure compensation treatment.

[0070] In one embodiment of the present invention, in step S5, the load, number of times and interval time during the compaction are verified, and the order of the primary and secondary influence of these three parameters in the current scenario is determined in order to determine the main situation of each parameter during compaction, clarify which parameter plays a decisive role in compaction, and thus reflect which parameters should be adjusted first in which areas.

[0071] like Figure 4 As shown, the implementation method of constructing the weight distribution map in step S5 includes: S51, based on the historical data corresponding to the current set of parameters to be optimized, configure the compensation effect value corresponding to the compensation area; the compensation effect value represents the ratio of the absolute value of the difference between the settlement deviation before and after compensation compaction to the settlement deviation before compaction, and this value represents the compaction effect presented in historical data similar to the current one.

[0072] S52 uses the compaction effect as the dependent variable and the load, number of compaction passes, and interval time of compaction as independent variables to perform multiple linear regression and obtain the standardized regression coefficients corresponding to each parameter in the parameter set to be optimized.

[0073] S53. Based on the range of standardized regression coefficients, set the order of primary and secondary influence of each parameter in the parameters to be optimized; combine the ratio of the predicted value and the maximum allowable value of the parameters to be optimized in each pressure replenishment region to complete the weight arrangement of each pressure replenishment region.

[0074] The larger the value of the standardized regression coefficient, the more significant the influence of the parameter on the current parameter. Then, the weight distribution of each pressure replenishment area under the current prediction process can be obtained.

[0075] The standardized regression coefficient can be expressed as the regression coefficient under the corresponding compaction effect. Since standardization is used in linear regression, the standardized regression coefficient is used to explain the regression coefficient obtained now so that the current calculation content is different from the regression coefficient of the previous regression calculation. It should be noted that the standardized regression coefficient represents the ratio of the regression coefficient of load, number of compaction times and interval time to the whole. That is, the regression coefficients obtained from the three factors are comprehensively identified and normalized to explain the relative size ratio between the three regression coefficients.

[0076] The standardized regression coefficient with the largest value will be regarded as having the greatest impact on the compaction process and set as the main influencing element. This will give us the part of the interval time, load and number of compaction cycles that has the greatest impact on the current compaction area. In one scenario, this can be expressed as the order of primary and secondary influence: load > number of compaction cycles > interval time.

[0077] The weight value of each compaction zone will be set by combining its standardized regression coefficient and the predicted value marked in the current set of parameters to be optimized; for example, it can be expressed as standardized regression coefficient × predicted load value / maximum allowable load of the zone; the weight corresponding to the number of compaction times is set in the same way as its load, and the weight of the interval time needs to be set by selecting the reciprocal of the ratio of the predicted value and the maximum allowable value; it can be seen that the larger the interval time, the better the compaction effect of the current zone. At this time, the ratio of the predicted value and the maximum allowable value should be consistent with the value of the number of compaction times and the load. Therefore, the reciprocal of its ratio is selected to interpret its weight, and all calculated weights are normalized to make the weight configuration of each compaction zone visible.

[0078] In one embodiment of the present invention, such as Figure 5 As shown, the implementation method for determining the compaction sequence at each position in the weight distribution map in step S6 includes: S61, obtaining the coordinate position of each compaction point in the compaction area, and determining the start and end points of the compaction point in each row; each compaction point represents the coordinates of a compaction area that needs to be compacted; if the positions of the compaction points in the compaction area are synchronized, and there is only one compaction point in a certain row, a virtual start point will be used to mark the position of the current compaction point, that is, extending one grid forward and backward along a single compaction point, and setting the start and end points in the form of a standard grid; or the current compaction point is used as the start or end point, and one point is extended forward or backward to record the position of the current compaction point.

[0079] S62, at the starting and ending points of each row, the path arrangement information of each row is set according to the order of primary and secondary influence of the parameters to be optimized; the compaction points on each row are sorted according to the compaction points between the starting and ending points, and the path arrangement information of each row is recorded.

[0080] Assuming that the load is the primary factor, the number of compaction operations is secondary, and the interval time is the least important factor, then the sorting will prioritize the load, sorting them from largest to smallest according to their weight. If the weights of the loads are the same, then sorting them from largest to smallest according to the weights of the number of compaction operations. Finally, if the weights of the number of compaction operations are the same, then sorting them from largest to smallest according to the weights of the interval time.

[0081] S63, based on the path arrangement information of the current row, determine the shortest turning path between the adjacent row and the current row, and connect the adjacent rows step by step with the shortest turning path to determine the tamping order to be executed.

[0082] If some compaction points are out of order due to weight sorting, the path arrangement information of each row can be converted into a global sorting method. In the form of a weighted undirected graph, the current row and the compaction points of adjacent rows are constructed into a global graph structure. Then, the shortest turning path is set according to the position of the compaction points, and finally the output compaction order is obtained.

[0083] Therefore, in step S63, the method for determining the tamping sequence also includes: S631, for the path arrangement information of the current row, if the direction corresponding to each tamping point in the path arrangement information is consistent, then the shortest distance between the end point of the current row and the start point or end point of the adjacent row is selected as the shortest turning path for output; if the direction of each row's tamping point is consistent with the direction corresponding to its weight, then the shortest turning path is set according to the distance between the start point of each row and the start point or end point of other rows, so that the tamping sequence can present a serpentine curved path.

[0084] S632, if the directions corresponding to each compaction point in the path arrangement information are not completely consistent, it means that according to the weight distribution, they cannot be directly connected into a single form of arrangement path, and there are conflicting compaction point pairs; obtain the total number of conflicting compaction points, and determine the node scale for correction processing based on the ratio of the number of conflicting compaction points to the total number of global compaction points.

[0085] S633: When the ratio of the number of conflict tamping points to the number of global tamping points is greater than the preset ratio, all tamping points are converted into a global topology map. Adjacent tamping points in the same row and tamping points that can be reached between adjacent rows are connected respectively. The edge weight in the same row is the moving distance of the equipment, and the edge weight in adjacent rows is the moving distance across rows plus the safety distance. The safety distance represents the safe distance when the equipment moves, such as 5 meters.

[0086] The topological connection order between each compaction point is determined, and the shortest turning path is the one with the shortest total distance between all compaction points. When constructing the global topology graph, all compaction points need to be connected, and points with high edge weights are visited first. The topological connection order indicates that the movement direction after the compaction points are connected is continuous. In this way, the shortest distance is found as the shortest turning path output when there is a conflict between compaction points.

[0087] As for the aforementioned preset ratio, it will be set based on the average number of additional compaction points when making global adjustments in historical data according to the compaction sequence, in order to limit the global adjustments of the current scheme.

[0088] S634 When the ratio of the number of conflict reinforcement points to the number of global reinforcement points is less than the preset ratio, the direction corresponding to the largest weight among the conflict reinforcement points is taken as the adjustment direction, the starting point and ending point of the current row after adjustment are determined, and the shortest distance between the ending point of the current row and the starting point or ending point of the adjacent row is taken as the shortest turning path of the output.

[0089] When there are too few compaction points requiring adjustment, the portion with the highest weight among the conflicting compaction points will be selected, and their order will be adjusted so that the compaction sequence within the current row can be gradually compacted according to the weights of load, number of compaction cycles, and interval time; this completes the adjustment in a local direction. Then, based on the actual compaction effect, steps S3-S6 are repeated continuously until the derived number of cycles, load, and other values ​​are achieved, and the soil compaction is completed under the corresponding compaction sequence.

[0090] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention, which are still covered within the protection scope of the present invention.

Claims

1. A method for optimizing and regulating construction parameters of dynamic compaction for high fill subgrade, characterized in that, include: Based on the roadbed embankment height, the compaction area at each compaction was retrieved, and the compaction construction parameters were constructed using the settlement, load and number of compactions corresponding to the compaction area. Based on the obtained compaction construction parameters, the settlement-load relationship curve and the settlement-compaction number curve are obtained respectively, and the optimal number of compaction blows and the optimal load are determined. Based on the optimal number of compaction blows and the optimal load, the differences between each compaction zone are calculated, and the additional compaction zone is determined by the settlement deviation and position deviation of each compaction zone. Using the settlement deviation and location deviation corresponding to the compaction area as input variables, and the load, number of compaction cycles, and interval time of the compaction compaction as output variables, a multiple linear regression model is established, and the parameters to be optimized are obtained by multiple linear regression calculation. Based on the standardized regression coefficients of the multiple linear regression model, the order of primary and secondary influence of the parameters to be optimized is determined, and a weight distribution map is constructed based on the distribution location of the pressure compensation region. Determine the tamping sequence at each location in the weight distribution map, and use the tamping sequence to complete the optimization process for each region; The methods for determining the optimal number of compaction blows and the optimal load include: Based on the settlement, number of tamping blows, and load in different tamping zones, data fitting was performed using the settlement amount of each tamping blow to construct settlement-load relationship curves and settlement-number of tamping blows relationship curves. The optimal load is extracted by using the rate of change of the slope between two consecutive points in the settlement-load relationship curve as a criterion. The optimal number of compaction blows is extracted by using the settlement difference between two consecutive points in the settlement-number of compaction blows curve as a criterion. The methods for constructing a weight distribution graph include: Based on the historical data corresponding to the current set of parameters to be optimized, configure the pressure compensation effect value corresponding to the pressure compensation area; Using the compaction effect as the dependent variable and the load, number of compaction passes, and interval time of compaction as independent variables, a multiple linear regression was performed to obtain the standardized regression coefficients corresponding to each parameter in the parameter set to be optimized. Based on the range of standardized regression coefficients, the order of primary and secondary influence of each parameter in the parameters to be optimized is set; and the weighting of each pressure replenishment region is completed by combining the ratio of the predicted value to the maximum allowable value of the parameters to be optimized in each pressure replenishment region. The methods for determining the tamping sequence at each location in the weight distribution diagram include: Obtain the coordinates of each compaction point in the compaction area, and determine the start and end points of each compaction point in each row; At the starting and ending points of each row, the path arrangement information for each row is set according to the order of primary and secondary influence of the parameters to be optimized. Based on the path arrangement information of the current row, determine the shortest turning path between the adjacent rows and the current row, and connect the adjacent rows step by step with the shortest turning path to determine the tamping order.

2. The method according to claim 1, wherein, Other methods for constructing compaction construction parameters include: Based on the arrangement information of each compaction zone, the parameters of the compaction zone in each row are identified, and the compaction construction parameters corresponding to each compaction zone are recorded according to the shape of the zone.

3. The method for optimizing and controlling construction parameters of dynamic compaction and supplementary compaction in high embankment subgrade according to claim 1, characterized in that, The methods for determining the pressure compensation area include: The foundation settlement is set based on the optimal number of compactions and the optimal load for the current compaction area; the settlement deviation of each compaction area is obtained based on the difference between the foundation settlement and the final settlement of the corresponding compaction area. Obtain the center coordinates and compaction radius of the current compaction area, and determine the positional deviation of the compaction area based on the compaction radii between adjacent compaction areas; Based on the current distribution of the compaction area, the area type of each compaction area is divided, and semantic labels are set according to the value range of settlement deviation and position deviation in each area type to determine the label combination form of each compaction area. Based on the label combination of each compaction area, the compaction areas are initially screened; for the compaction areas after the initial screening, the priority weights are determined by the settlement deviation and the position deviation, and the output list of compaction areas is determined.

4. The method for optimizing and controlling construction parameters of dynamic compaction and supplementary compaction in high embankment subgrade according to claim 3, characterized in that, The preliminary screening of the compaction area also includes the following methods: When the semantic labels for both settlement deviation and location deviation meet the standards, the corresponding data will be removed. When there are substandard semantic tags for settlement deviation and location deviation, the corresponding area is regarded as the area to be verified. For each area to be verified, determine whether the adjacent areas to be verified are all up to standard. If they are all up to standard, remove the corresponding data and use the remaining data as the initial screening area.

5. The method for optimizing and controlling construction parameters of dynamic compaction and supplementary compaction in high embankment subgrade according to claim 3, characterized in that, The implementation methods for determining the label combination format for each compaction zone also include: When dealing with settlement deviations, a settlement distribution matrix is ​​constructed based on the continuous distribution location of each compaction zone; The settlement distribution matrix is ​​decomposed according to the direction of settlement deviation and the region type to obtain multiple settlement sub-matrices; The weights of the submatrix with negative deviation after decomposition are normalized to obtain the priority weights corresponding to the settlement deviation. When dealing with positional deviations, a positional deviation matrix is ​​constructed for the current compaction area according to the 8-neighbor directions. The positional deviations between the current compaction area and the adjacent compaction areas in the positional deviation matrix are used as the priority weights for configuring the positional deviations of the current compaction area. The priority weights of settlement deviation and location deviation are synchronized to the corresponding compaction areas to obtain the combined weights of each compaction area. The list of compaction areas is then output in descending order of the value range of the combined weights.

6. The method for optimizing and controlling construction parameters of dynamic compaction and supplementary compaction in high embankment roadbeds according to claim 1, characterized in that, When calculating the parameters to be optimized using multiple linear regression, the implementation methods also include: By combining the settlement deviation and location deviation of each replenishment area, the predicted value of each replenishment area in the parameters to be optimized is obtained, and the difference between the predicted values ​​of each replenishment area is taken as the accuracy of the prediction. Based on the difference in predicted values ​​of the parameters to be optimized, the model regression is reprocessed using the difference in predicted values ​​of the parameters to be optimized as part of the re-verification. The regression is then processed according to the regression processing of each model to obtain the output parameters to be optimized.

7. The method for optimizing and controlling construction parameters of dynamic compaction and supplementary compaction in high embankment subgrade according to claim 1, characterized in that, Other methods for determining the order of impacts include: For the path arrangement information of the current row, if the direction of each compaction point in the path arrangement information is consistent, then the shortest distance between the end point of the current row and the start or end point of the adjacent row is selected as the shortest turning path for output. If the directions corresponding to each compaction point in the path arrangement information are not completely consistent, obtain the number of conflicting compaction points, and determine the node scale for correction processing based on the ratio of the number of conflicting compaction points to the total number of global compaction points. When the ratio of the number of conflict tamping points to the number of global tamping points is greater than the preset ratio, all tamping points are converted into a global topology map. Adjacent tamping points in the same row and reachable tamping points between adjacent rows are connected to determine the topology connection order between each tamping point. The shortest turning path is the one with the shortest total distance between each tamping point. When the ratio of the number of conflict reinforcement points to the total number of global reinforcement points is less than the preset ratio, the direction corresponding to the largest weight among the conflict reinforcement points is taken as the adjustment direction, the starting point and ending point of the current row after adjustment are determined, and the shortest turning path is taken as the shortest distance between the ending point of the current row and the starting point or ending point of the adjacent row.

Images