Method and system for predicting pile-forming quality and construction optimal parameter of jet grouting pile

Abstract

The invention relates to the technical field of engineering intelligent algorithms, in particular to a jet grouting pile forming quality and construction optimal parameter prediction method and system, and the method comprises the following steps: building a data sample set based on similar historical engineering cases; the construction stratum soil bodies of the jet grouting piles are classified to obtain different stratum soil body categories, and construction process sensitive parameters are obtained; performing an orthogonal test to obtain an orthogonal test result so as to establish a jet grouting pile construction numerical simulation model to obtain a simulation data sample set; a jet grouting pile strength prediction model is constructed, and a jet grouting pile forming quality prediction result is obtained; establishing constraint conditions, and optimizing the pile-forming quality prediction result of the jet grouting pile according to an optimization algorithm to obtain an optimal solution set; and in combination with actual jet grouting pile construction conditions, optimal construction parameters are obtained according to an optimal solution screening method. According to the method, the optimal construction parameters of the jet grouting pile project can be rapidly and efficiently obtained, so that the construction scheme is optimized, and the pile forming quality of the jet grouting pile is guaranteed.

Description

Technical Field

[0001] The invention relates to the technical field of engineering intelligent algorithms, and in particular to a method and system for predicting the quality of jet grouting piles and optimal construction parameters. Background Art

[0002] High-pressure jet grouting pile technology is an efficient foundation reinforcement method, which is widely used in soft soil foundation treatment, building foundation reinforcement and tunnel construction control. High-pressure jet grouting pile technology injects cement slurry into the underground soil layer through high-pressure jet grouting, so that the soil and cement slurry are fully mixed to form a high-strength cement soil pile body. The main advantage is that the construction is flexible and can adapt to complex geological conditions. At the same time, the jet grouting pile has high compressive strength and shear strength, which effectively improves the bearing capacity and stability of the foundation. In addition, the jet grouting pile has little impact on the environment during construction, which is particularly suitable for densely urbanized areas and areas with dense underground facilities.

[0003] However, the quality and strength of jet grouting piles are affected by many factors, resulting in large discreteness and uncertainty in the strength and impermeability of jet grouting piles. Influencing factors include the jet grouting process itself, the setting of construction parameters, and on-site geological conditions. In previous technical solutions, the evaluation of pile strength and impermeability relied on on-site sampling and laboratory testing, which was not only time-consuming and labor-intensive but also had high engineering costs. Therefore, in existing construction, the determination of construction parameters of jet grouting piles mostly uses empirical formulas and simple statistical models, which makes it difficult to accurately describe the complex nonlinear relationship between construction parameters and geological conditions, which may lead to an increase in project costs and an increase in construction safety risks.

[0004] In view of this, there is an urgent need for a data-driven intelligent prediction method. The present invention conducts sensitivity analysis on construction process parameters, constructs a jet grouting pile strength prediction model, and optimizes the prediction results using an optimization algorithm. This method can quickly and efficiently obtain the optimal parameters for jet grouting pile construction, so as to optimize the construction plan and thus improve the project quality. Summary of the invention

[0005] In view of the defects in the prior art, the present invention provides a method and system for predicting the quality of jet grouting piles and optimal construction parameters.

[0006] In order to achieve the above-mentioned purpose, in the first aspect, the present invention provides a method for predicting the quality of jet grouting and the optimal construction parameters of the pile, and the method comprises the following steps: based on similar historical engineering cases, the construction stratum parameters, construction process parameters and jet grouting evaluation parameters are obtained to construct a data sample set; the construction stratum soil of the jet grouting is classified into different stratum soil categories, and the construction process sensitive parameters of the different stratum soil categories are obtained; an orthogonal test is performed on the construction process sensitive parameters to obtain orthogonal test results, so as to establish a numerical simulation model for the construction of jet grouting and obtain a simulation data sample set; based on the simulation data sample set, a jet grouting strength prediction model is respectively constructed for the different stratum soil categories, so as to obtain the prediction result of the quality of the jet grouting; constraint conditions are established, and the prediction result of the quality of the jet grouting is optimized according to the optimization algorithm, so as to obtain the optimal solution set that meets the constraint conditions; in combination with the actual jet grouting construction situation, the optimal solution set is optimized and screened according to the optimal solution screening method to obtain the optimal construction parameters. The present invention improves the prediction accuracy of the quality of jet grouting piles by constructing a data sample set, classifying stratum soil, obtaining sensitive parameters and establishing a prediction model. Based on the prediction results, an optimization algorithm is used to obtain the optimal solution set that meets the constraints, obtain the optimal construction parameters, and improve the construction efficiency. At the same time, by adjusting the construction parameters, materials are saved, the bearing capacity of the foundation is increased, the stress distribution of the pile and the soil is more reasonable, and the overall engineering construction level is significantly improved.

[0007] Optionally, the data sample set is constructed by obtaining construction stratum parameters, construction process parameters and jet grouting pile evaluation parameters based on similar historical engineering cases, including: the construction stratum parameters include the stratum distribution and soil properties of the soil in the construction area where the jet grouting pile is located, the soil properties include the physical indicators and mechanical properties of the soil, the physical indicators include natural density, moisture content, void ratio, dry density, liquid index and standard penetration number, and the mechanical properties include compression modulus, shear strength and Poisson's ratio; the construction process parameters include nozzle diameter, lead hole diameter, cement content, water-cement ratio, shotcrete flow rate, air flow rate, water flow rate and lifting speed; the jet grouting pile evaluation parameters include the unconfined compressive strength and permeability coefficient of the jet grouting pile after consolidation and forming. The present invention collects the construction stratum parameters, construction process parameters and jet grouting pile evaluation parameters of similar historical engineering cases to construct a data sample set, which provides comprehensive and accurate data support for subsequent analysis, helps to improve prediction accuracy, optimize construction parameters, and improve construction efficiency, and shows significant application value in scenes such as foundation reinforcement and waterproofing.

[0008] Optionally, the soil of the stratum for the construction of the jet grouting pile is classified into different stratum soil categories, and the sensitive parameters of the construction process of the different stratum soil categories are obtained, including: classifying the soil of the construction stratum to obtain the different stratum soil categories, including silt, silty soil, clay, plain fill and sand; performing a global sensitivity analysis on the construction process parameters of the different stratum soil categories based on the evaluation parameters of the jet grouting pile to obtain the sensitive parameters of the construction process. The present invention classifies the soil of the stratum for the construction of the jet grouting pile, and performs a global sensitivity analysis on the construction process parameters of the different stratum soil categories based on the evaluation parameters of the jet grouting pile to obtain sensitive parameters of the construction process, which helps to improve the pertinence and efficiency of the selection of construction parameters, optimize the combination of construction parameters, and thus improve the quality of the jet grouting pile and the construction efficiency; at the same time, by clarifying the sensitive parameters of different stratum soil categories, a more accurate basis for parameter adjustment is provided for actual construction.

[0009] Optionally, the orthogonal test is performed on the sensitive parameters of the construction process to obtain orthogonal test results, so as to establish a numerical simulation model for the construction of jet grouting piles and obtain a simulation data sample set, including: based on the similar historical engineering cases, the value interval of the sensitive parameters of the construction process is defined, and the target unknown parameters are expanded using the value interval to construct a simulation test case; numerical simulation is performed according to the simulation test case, so as to establish the numerical simulation model for the construction of jet grouting piles; and the evaluation parameters of the jet grouting piles are obtained according to the numerical simulation model for the construction of jet grouting piles as the simulation data sample set. The present invention effectively improves the experimental efficiency by defining the parameter value interval and expanding the construction of simulation test cases, combining numerical simulation modeling, and obtaining evaluation parameters as the simulation data sample set. At the same time, it can comprehensively evaluate the influence of different construction parameter combinations on the pile quality and construction efficiency, reveal the interaction between parameters through orthogonal experiments, optimize the parameter combination, and improve the construction effect. In addition, the numerical simulation model provides accurate data support and decision-making basis for subsequent actual construction, and enhances the adaptability and effectiveness of the construction plan.

[0010] Optionally, based on the simulation data sample set, a jet grouting pile strength prediction model is constructed for each of the different stratum soil categories, thereby obtaining a prediction result of the quality of the jet grouting piles, including: normalizing the simulation data sample set to obtain normalized data; obtaining initialization parameters of the jet grouting pile strength prediction model, and dividing the simulation data sample set into pure subsets by recursive splitting; based on the pure subsets, the jet grouting pile strength prediction model is constructed for each of the different stratum soil categories; training and verifying the jet grouting pile strength prediction model to obtain the optimal model hyperparameter combination of the jet grouting pile strength prediction model; and using the jet grouting pile strength prediction model to establish a correspondence between the construction process parameters and the jet grouting pile evaluation parameters in the different stratum soil categories as the prediction result of the quality of the jet grouting piles. The present invention eliminates the influence of different dimensions and orders of magnitude through data normalization processing, thereby improving the model convergence speed and prediction accuracy; by dividing the pure subsets, the intrinsic structure and laws of the data are revealed, providing a basis for model construction; a jet grouting pile strength prediction model is constructed, and the optimal hyperparameter combination is obtained through model training and verification, thereby improving the prediction performance; a corresponding relationship between construction process parameters and evaluation parameters is established in different stratum soil categories, thereby providing accurate data reference and decision-making basis for actual construction.

[0011] Optionally, dividing the simulated data sample set into pure subsets by recursive splitting includes: measuring the purity of the pure subsets by using a Gini index, where the Gini index includes: ; in, Pure subset The Gini index, is the number of categories, For categories, Pure subset Medium Category The present invention effectively evaluates the purity of the subset through the Gini index. The smaller the index, the higher the purity, which helps to improve the prediction accuracy of the model, reveals the intrinsic structure and rules of the data, enhances the generalization ability of the model, provides accurate data support for the construction of the jet grouting pile strength prediction model, optimizes the construction parameter selection, and is conducive to improving the construction efficiency.

[0012] Optionally, constructing the jet grouting pile strength prediction model for the different stratum soil types respectively includes: ; in, To categorize the voting results, It means to take the category that makes the function value the largest. is the number of decision trees, For decision tree, is the indicator function, For the The prediction results output by the decision tree are: is the input sample, The present invention constructs a jet grouting pile strength prediction model for different stratum soil categories, and improves the prediction accuracy through ensemble learning of multiple decision trees. The jet grouting pile strength prediction model integrates the prediction results of multiple decision trees to obtain the final prediction result, which effectively reduces the risk of overfitting and enhances the generalization ability of the model.

[0013] Optionally, the establishing of constraint conditions, optimizing the jet grouting pile quality prediction results according to an optimization algorithm to obtain an optimal solution set that satisfies the constraint conditions, includes: using the jet grouting pile evaluation parameters as the constraint conditions, including: ; in, and As constraints, is the minimum unconfined compressive strength obtained based on the jet grouting pile strength prediction model, is the minimum unconfined compressive strength allowed by the project, is the maximum permeability coefficient obtained based on the jet grouting pile strength prediction model, The invention sets the minimum unconfined compressive strength and the minimum permeability as constraints to ensure that the quality of the jet grouting pile meets the engineering requirements. The invention continuously iterates the optimal solution through the optimization algorithm. It has the advantages of strong global search capability and highly parallel processing capability, and can effectively avoid falling into the local optimal solution, realize the search of the global optimal solution, and provide accurate data support for the optimization of the construction parameters of the jet grouting pile.

[0014] Optionally, the optimal solution set is optimized and screened according to the optimal solution screening method to obtain the optimal construction parameters in combination with the actual jet grouting pile construction situation, including: obtaining the corresponding equation between the construction cost of the jet grouting pile and the construction process parameters, including: ; in, The construction cost of jet grouting piles is is the amount of dielectric material, For dielectric materials, is the unit volume cost of dielectric material, is the amount of medium material; under the constraints, according to the optimal solution screening method, a step-by-step verification mechanism is used to optimize and screen the optimal solution set to obtain the rotary grouting pile construction plan with the lowest construction cost, so as to obtain the optimal construction parameters. The present invention establishes an equation between the construction cost of the rotary grouting pile and the construction process parameters, quantifies the construction cost, provides data support for optimizing the construction parameters, and uses a step-by-step verification mechanism to screen the optimal solution set to ensure the feasibility and economy of the selected construction parameters, exclude infeasible or poorly economical solutions, and improve the reliability of the optimization results.

[0015] In the second aspect, the present invention provides a prediction system for the quality and optimal construction parameters of a jet grouting pile, the system executes the prediction method for the quality and optimal construction parameters of a jet grouting pile provided by the present invention, the system includes an input device, an output device, a processor and a memory, and its gain lies in: the hardware facilities integrated by the present invention have excellent performance, the input device, the output device, the processor and the memory are interconnected, the information transmission between the various components is smooth, and an efficient information processing system is constructed through the interaction of multiple hardware facilities. The present invention dynamically monitors the construction process of the jet grouting pile, feeds back and adjusts the construction parameters in real time, ensures the pile formation effect and construction quality, and improves the intelligent level of the jet grouting pile construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A flow chart of a method for predicting the quality of jet grouting piles and optimal construction parameters according to an embodiment of the present invention;

[0017] Figure 2 A schematic diagram of a clustering algorithm flow chart of an embodiment of the present invention;

[0018] Figure 3 A schematic diagram of a random forest algorithm flow chart of an embodiment of the present invention;

[0019] Figure 4 A schematic diagram of the archived micro-genetic algorithm flow chart of an embodiment of the present invention;

[0020] Figure 5 Schematic diagram of the construction process of the triple-tube method of jet grouting piles according to an embodiment of the present invention;

[0021] Figure 6 This is a framework diagram of a system for predicting the quality of jet grouting piles and optimal construction parameters according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The specific embodiments of the present invention will be described in detail below. It should be noted that the embodiments described herein are only for illustration and are not intended to limit the present invention. In the following description, a large number of specific details are set forth in order to provide a thorough understanding of the present invention. However, it is obvious to those of ordinary skill in the art that these specific details do not need to be adopted to implement the present invention. In other examples, in order to avoid confusing the present invention, known circuits, software or methods are not specifically described.

[0023] Throughout the specification, references to "one embodiment," "an embodiment," "an example," or "an example" mean that a particular feature, structure, or characteristic described in conjunction with the embodiment or example is included in at least one embodiment of the present invention. Therefore, the phrases "in one embodiment," "in an embodiment," "an example," or "an example" appearing in various places throughout the specification do not necessarily all refer to the same embodiment or example. In addition, particular features, structures, or characteristics may be combined in one or more embodiments or examples in any suitable combination and / or subcombination. In addition, it should be understood by those of ordinary skill in the art that the figures provided herein are for illustrative purposes and that the figures are not necessarily drawn to scale.

[0024] See also Figure 1 An embodiment of the present invention provides a method for predicting the quality of jet grouting piles and optimal construction parameters, the method comprising the following steps:

[0025] S1. Based on similar historical engineering cases, the construction stratum parameters, construction process parameters and jet grouting pile evaluation parameters are obtained to construct a data sample set.

[0026] In this embodiment, the construction stratum parameters of each construction case, the construction process scheme of the jet grouting pile and its related parameters, and the strength and impermeability of the jet grouting pile after consolidation and molding are obtained in combination with previous similar engineering schemes to obtain a data sample set. The construction stratum parameters include the stratum distribution and soil characteristics of the soil in the construction area where the jet grouting pile is located. The soil characteristics include the physical indicators and mechanical properties of the soil. The physical indicators include natural density, moisture content, porosity, dry density, liquid index and standard penetration number, and the mechanical properties include compression modulus, shear strength and Poisson's ratio; the construction process parameters include nozzle diameter, lead hole diameter, cement content, water-cement ratio, shotcrete flow rate, air flow rate, water flow rate and lifting speed; the jet grouting pile evaluation parameters include the unconfined compressive strength and permeability coefficient of the jet grouting pile after consolidation and molding.

[0027] Specifically, by setting the air flow rate and the water flow rate to 0, it is indicated that the construction process method is the single-pipe method; by setting the water flow rate to 0, it is indicated that the construction process method is the double-pipe method.

[0028] Furthermore, by obtaining 2000 sets of data samples of similar historical engineering cases, a data sample set is constructed, as shown in Table 1:

[0029] Table 1

[0030]

[0031] Among them, the ellipsis in the table represents the parameter conditions of similar historical engineering cases with serial numbers between 4 and 1997, as well as some construction geographical parameters and process parameters. Due to the large number of similar historical engineering cases, ellipsis is used to represent these conditions.

[0032] S2. Classify the soil of the construction stratum of the jet grouting pile into different stratum soil categories, and obtain sensitive parameters of the construction process of the different stratum soil categories.

[0033] Among them, S2 specifically includes the following steps:

[0034] S21. Classifying the construction stratum soil to obtain the different stratum soil types, including silt, silty soil, clay soil, plain fill soil and sandy soil.

[0035] In this embodiment, a K-medoids clustering algorithm (K-medoids for short) is selected to classify the construction stratum soil into different stratum soil categories, including silt, silty soil, clay soil, plain fill soil and sandy soil.

[0036] See also Figure 2 , the figure is a schematic diagram of the clustering algorithm process; first, the typical representative parameter samples of the above five different stratum soil categories are input as the initial cluster centers; secondly, the division criteria are established based on the Euclidean distance metric, and the remaining samples are assigned to the data clusters corresponding to the nearest neighbor cluster centers; finally, for each data cluster, the cluster center is updated by solving, forming an iterative optimization process until the convergence conditions are met.

[0037] Specifically, for each soil parameter point of a similar historical engineering case in the data sample set, the distance from the soil parameter point to the primary clustering center of the soil type in different strata is obtained to satisfy the following relationship:

[0038]

[0039] in, is the Euclidean distance, is the total number of parameters, is the index variable, For each similar historical engineering case The value of the parameter, is the first The value of a parameter.

[0040] Furthermore, based on the distance from the soil parameter points of each engineering case to the primary clustering center of different stratum soil types, the parameter points are divided into the stratum soil type corresponding to the primary clustering center of the shortest distance. A criterion function is constructed to traverse the data points in various soil types, and the data point with the minimum criterion function is obtained as the new cluster center; the criterion function satisfies the following relationship:

[0041]

[0042] in, is the criterion function, is the index variable for traversing data points, is the Euclidean distance.

[0043] Repeat the above method steps to classify the soil types of the samples in the data sample set.

[0044] S22. Based on the jet jet pile evaluation parameters, a global sensitivity analysis is performed on the construction process parameters of the different stratum soil types to obtain the construction process sensitive parameters.

[0045] In this embodiment, different stratum soil categories are obtained based on the classification of the construction stratum soil. For the construction process schemes and related parameters of the jet grouting piles in different stratum soil categories, the relevant programs of the Matrix Laboratory (MATLAB) are called, and the global sensitivity analysis of parameters is performed based on the probabilistic analysis of numerical uncertainty (PAWN) based on the unconfined compressive strength and permeability of the jet grouting piles after consolidation. Parameters with a global sensitivity index less than 0.05 are judged as insensitive parameters, and vice versa.

[0046] The sensitivity of each construction process parameter in different soil types is shown in Table 2, where “√” indicates that the parameter is sensitive and “×” indicates that the parameter is insensitive.

[0047] Table 2

[0048]

[0049] Furthermore, any other parameter sensitivity analysis method can be used instead of the PAWN parameter global sensitivity analysis method according to the difference in the number of original construction data and parameter accuracy.

[0050] S3. Conducting an orthogonal test on the sensitive parameters of the construction process to obtain orthogonal test results, so as to establish a numerical simulation model for jet grouting pile construction and obtain a simulation data sample set.

[0051] In this embodiment, orthogonal experiments are carried out on sensitive parameters of the rotary jet grouting pile construction process in different stratum soil types, and a numerical simulation model for rotary jet grouting pile construction is constructed based on the orthogonal experimental results. The unconfined compressive strength and permeability coefficient of the rotary jet grouting pile after consolidation are obtained as a simulation data sample set.

[0052] Specifically, the orthogonal test method is to define a value range for the sensitive parameters based on previous construction cases and relevant literature, and use the value range to expand each of the target unknown parameters to construct a large number of simulation test cases.

[0053] Furthermore, numerical simulations were carried out based on the software Discrete Lattice Spring Model (DLSM) and the large-scale commercial software COMSOL Multiphysics (Comsol), and then a numerical simulation model of jet grouting pile construction was constructed to obtain the unconfined compressive strength and permeability coefficient of the jet grouting pile after consolidation.

[0054] In the numerical simulation, the insensitive parameters of the jet grouting pile construction process in the stratum can be determined based on previous construction cases and relevant construction experience without participating in the orthogonal test of the stratum.

[0055] In actual engineering cases, if the global sensitivity analysis of the construction process parameters of each stratum shows that the air flow rate and the water flow rate are both insensitive parameters, the single-tube method with the lowest cost is selected in the rotary jet pile construction process, and the air flow rate and water flow rate of the rotary jet pile construction in each stratum are both taken as 0 in the numerical simulation, and the air flow rate and water flow rate do not participate in the orthogonal test; if the global sensitivity analysis of the construction process parameters in the stratum type shows that the air flow rate is a sensitive parameter, and the water flow rate is an insensitive parameter for all stratum soil bodies, then the double-tube method is selected in the rotary jet pile construction process, and the water flow rate of the rotary jet pile construction in each stratum is taken as 0 in the numerical simulation, and does not participate in the orthogonal test, and the air flow rate participates in the orthogonal test; in other cases, the triple-tube method is used by default, and the air flow rate and water flow rate of the rotary jet pile construction in each stratum are both involved in the orthogonal test in the numerical simulation.

[0056] In an optional embodiment, taking the construction of foundation pit jet grouting piles in a certain section of the Yangtze River Bridge project as an example, the designed construction pile length is 54.2m. K-medoids is used to classify the 1# stratum as silt, the 2# stratum as silty soil, and the 3# stratum as clay soil.

[0057] For the 3# stratum of clay soil type, the air flow rate parameter and water flow rate parameter in the jet grouting pile construction process parameters are both sensitive parameters. In the numerical simulation, both the air flow rate parameter and the water flow rate parameter are involved in the orthogonal test.

[0058] Specifically, the orthogonal table of sensitive parameters of the rotary jet grouting pile construction process under different construction strata in actual construction cases is shown in Table 3, where “*” indicates an insensitive parameter with a default value based on previous construction cases and relevant construction experience.

[0059] Table 3

[0060]

[0061] Specifically, the relevant construction stratum parameters and the jet grouting pile construction parameters in Table 3 were imported into the numerical simulation model of the jet grouting pile construction. The DLSM software was used to fix the upper surface of the jet grouting pile model and increase the lower surface from bottom to top at a speed of 0.01 m / s to obtain the peak value of the unconfined compressive strength of different jet grouting piles; the seepage extraction function provided by the COMSOL finite element analysis software was used to calculate the permeability coefficient of different jet grouting piles. Taking the 1# silt stratum as an example, the evaluation parameters of the jet grouting piles for each combination of construction process parameters in different stratum types were obtained through numerical simulation, as shown in Table 4:

[0062] Table 4

[0063]

[0064] S4. Based on the simulation data sample set, a jet grouting pile strength prediction model is constructed for each of the different stratum soil types, thereby obtaining a jet grouting pile quality prediction result.

[0065] Wherein, S4 specifically includes the following steps:

[0066] S41, normalizing the simulated data sample set to obtain normalized data.

[0067] Specifically, the data is normalized to satisfy the following relationship:

[0068]

[0069] in, To normalize the data, is the sample data, is the minimum value of the sample data, is the maximum value of the sample data.

[0070] S42, obtaining initialization parameters of the jet grouting pile strength prediction model, and dividing the simulation data sample set into pure subsets by recursive splitting.

[0071] Specifically, the simulated data sample set is divided into pure subsets in a recursive splitting manner based on the decision number algorithm, and the purity of the pure subset is measured using the Gini index, which satisfies the following relationship:

[0072]

[0073] in, Pure subset The Gini index, is the number of categories, For categories, Pure subset Medium Category The sample proportion.

[0074] S43. Based on the pure subset, the jet jet pile strength prediction model is constructed for each of the different stratum soil types.

[0075] In this embodiment, a rotary jet grouting pile strength prediction model is constructed using a random forest algorithm, and initialization parameters of the random forest model are selected, including the number of trees, the maximum depth of the tree, and the minimum number of samples required for node splitting.

[0076] Specifically, the jet grouting pile strength prediction model satisfies the following relationship:

[0077]

[0078] in, To categorize the voting results, It means to take the category that makes the function value the largest. is the number of decision trees, For decision tree, is the indicator function, For the The prediction results output by the decision tree are: is the input sample, For category.

[0079] Furthermore, based on the jet grouting pile strength prediction model, the corresponding relationship between the jet grouting pile construction process parameters and the unconfined compressive strength and permeability coefficient of the jet grouting pile after consolidation and forming in different stratum soil types was established as the prediction result of the jet grouting pile quality.

[0080] S44, training and verifying the jet grouting pile strength prediction model to obtain the optimal model hyperparameter combination of the jet grouting pile strength prediction model.

[0081] In this embodiment, the training data set is input into the above-mentioned jet grouting pile strength prediction model through the input layer to obtain the output result of the output layer. When the prediction accuracy of the jet grouting pile strength prediction model reaches the specified stop condition, the model training is completed; the training data set satisfies the following relationship:

[0082]

[0083] in, is the training data set, are the construction process parameters for different soil types. It is the unconfined compressive strength of the jet grouting pile after consolidation. It is the permeability coefficient of the jet grouting pile after consolidation.

[0084] Furthermore, on the validation set, the trained jet grouting pile strength prediction model was verified using the determination coefficient to evaluate the accuracy of the jet grouting pile strength prediction model; the determination coefficient satisfies the following relationship:

[0085]

[0086] in, is the coefficient of determination, is the total number of recorded data, is the index variable for recording data, For the The predicted value of the recorded data, For the The measured value of the recorded data, for The average value of the recorded data.

[0087] When building and training the model, the grid search algorithm is used to repeatedly adjust the model hyperparameters to find the hyperparameter combination that makes the determination coefficient approach 1. When the preset stop condition is reached, the parameter optimization process ends and the optimal model hyperparameter combination of the rotary jet grouting pile strength prediction model is obtained.

[0088] See also Figure 3, the figure is a schematic diagram of the random forest algorithm process; taking the construction of the jet grouting pile strength prediction model of 1# silt formation type as an example, its technical process is as follows: perform outlier detection on the simulated data sample set, eliminate missing values ​​and outlier samples, divide the processed 500 sets of training data (numbered 1-500) and 100 sets of test data (numbered 501-600) in a ratio of 5:1, and convert the data into a standard normal distribution through normalization; the model training uses the grid search algorithm to optimize the initial weight parameter combination, configure the number of decision trees to be 1000, the upper limit of the tree depth to be 20 layers, and the minimum number of node split samples to be 5. It has been verified that the prediction accuracy of the constructed jet grouting pile strength prediction model on the test set is more than 96%, which confirms the reliability of the model. Based on the same modeling process, the jet grouting pile strength prediction model of 2# silt soil formation type and 3# clay soil formation type can be established simultaneously.

[0089] S5. Establish constraint conditions, and optimize the prediction results of the jet grouting pile quality according to the optimization algorithm to obtain the optimal solution set that meets the constraint conditions.

[0090] In this embodiment, the jet grouting pile evaluation parameters are used as constraint conditions to satisfy the following relationship:

[0091]

[0092]

[0093]

[0094] in, and As constraints, is the minimum unconfined compressive strength obtained based on the jet grouting pile strength prediction model, is the minimum unconfined compressive strength allowed by the project, is the maximum permeability coefficient obtained based on the jet grouting pile strength prediction model, is the maximum permeability coefficient allowed by the project, The unconfined compressive strength of this group of jet grouting pile construction process parameters in various strata obtained based on the jet grouting pile strength prediction model. It is the permeability coefficient of this group of jet grouting pile construction process parameters in various strata obtained based on the jet grouting pile strength prediction model.

[0095] It should be noted that the minimum unconfined compressive strength allowed by the project is The maximum permeability coefficient allowed by the project is 1.5MPa. 1E-7cm / s.

[0096] Furthermore, the Archive-based Micro Genetic Algorithm (AMGA) is used as an optimization algorithm to optimize the prediction results of the quality of jet grouting piles, and the Pareto optimal solution set that meets the constraints is obtained.

[0097] Firstly, the unconfined compressive strength and permeability coefficient of the jet grouting piles in each stratum after consolidation are determined as constraints, and the construction cost of the jet grouting piles is taken as the optimization target; secondly, the basic parameters of the AMGA algorithm, such as the number of iterations, population size, crossover and mutation probability, are selected to create the initial population and initialize the archive; then, non-dominated individuals are selected through binary constraint competition; then, in the current archive, non-dominated individuals are screened out according to the dominance relationship; then, after the AMGA algorithm generates the initial chromosome, the dominance relationship between the candidate solution and the archive is dynamically detected: if the candidate solution is dominated by the archived individual, it is eliminated; if there is a non-dominated relationship, it is stored in the archive; if the candidate solution dominates the archived individual, the dominated individual is stored and removed to maintain the high-quality non-dominated solution set in the archive; finally, it is determined whether the AMGA algorithm has reached the maximum number of iterations. If not, return to the previous step; if yes, end the algorithm, and the output archive includes the Pareto optimal solution set that meets the constraints.

[0098] According to the construction process parameters of jet grouting pile, the chromosome of AMGA algorithm has 6 gene positions, namely, the diameter of the lead hole, cement content, water-cement ratio, shotcrete flow rate, air flow rate and water flow rate. The fitness of the chromosome is measured according to the optimization objectives of the unconfined compressive strength, permeability coefficient and economy of the jet grouting pile after consolidation and molding. The key operating parameters of the genetic algorithm are as follows: the population size is 30, the probability of crossover is 0.8, the probability of mutation is 0.55 and the maximum number of iterations is 3000.

[0099] See also Figure 4 , the figure is a schematic diagram of the archive micro-genetic algorithm process; the elite individuals in the archive are extracted by judging the dominance relationship. If the candidate solution is dominated by the archived individual, it is automatically eliminated. If there is a non-dominance relationship with the archived individual, it is included in the archive. If it dominates the archived individual, it replaces the dominated item, so as to achieve efficient update of the archive set; through the iterative termination condition judgment, when the preset maximum number of iterations is reached, the output archive set includes the Pareto optimal solution set. The Pareto optimal solution set is shown in Table 5:

[0100] Table 5

[0101]

[0102] S6. Based on the actual jet grouting pile construction situation, the optimal solution set is optimized and screened according to the optimal solution screening method to obtain the optimal construction parameters.

[0103] In this embodiment, by referring to relevant similar projects and market research, the corresponding equation between the construction cost of the jet grouting pile and the construction process parameters is obtained as the optimization target, satisfying the following relationship:

[0104]

[0105] in, The construction cost of jet grouting piles is is the amount of dielectric material, For dielectric materials, is the unit volume cost of dielectric material, The amount of dielectric material used.

[0106] It should be noted that the medium materials include cement, admixtures, accelerators and suspending agents; the unit volume cost includes the cost of the material itself, machinery cost, labor cost, equipment loss cost and other costs.

[0107] On the premise of meeting the constraints, the construction cost of jet grouting piles is used as a standard to evaluate the overall performance of the jet grouting pile process and related parameter selection. The smaller the function value of the jet grouting pile construction cost, the better the method.

[0108] It should be pointed out that under the given construction technology and parameter combination, the construction cost of jet grouting piles can be quantitatively calculated. The cost model has the characteristics of stratum independence, and its numerical results are not affected by changes in stratum conditions.

[0109] In this embodiment, the relevant parameters in the Pareto optimal solution set are optimized based on the actual rotary jet grouting pile construction situation. Specifically, when the parameter values ​​of the air flow rate and the water flow rate are close to zero or significantly lower than the process threshold, the single-pipe method is determined to be applicable; when only the water flow rate parameter is close to zero or significantly lower than the process threshold, and the air flow rate meets the requirements of the specification, the double-pipe method construction technology is determined to be adopted; and the diameter of the guide hole is rounded up.

[0110] In an optional embodiment, the water flow rate in the optimal solution is When the value is significantly lower than the process threshold and the air flow rate meets the requirements of the specification, the double-tube construction process is determined to be adopted; at the same time, the hole diameter is taken as 190mm based on the actual engineering parameters; after substituting the obtained optimal construction parameters into the numerical simulation model of the rotary jet pile construction, the output result is the unconfined compressive strength , permeability coefficient , which does not meet the construction requirements of this example.

[0111] Furthermore, the optimal solution screening method is implemented by a step-by-step verification mechanism to obtain the lowest-cost jet grouting pile construction plan under the constraints. Specifically, the optimization result of the global optimal solution in the Pareto optimal solution set is input into the numerical simulation model of jet grouting pile construction for calculation to obtain the response function; then, a judgment function is constructed based on the constraints to determine whether the optimal solution meets the construction feasibility requirements; if the conditions are not met, the above verification process is iteratively executed in Pareto hierarchy order until the first optimized solution that meets the conditions is obtained as the optimal construction parameter.

[0112] The above step-by-step verification mechanism is repeatedly executed in the order of the Pareto hierarchy to obtain the first optimization solution that meets the constraints as shown in Table 6:

[0113] Table 6

[0114]

[0115] After substituting the optimized solution in Table 6 into the numerical simulation model of jet grouting pile construction, the result is the unconfined compressive strength , permeability coefficient , which meets the construction requirements of this example; this optimization solution is the final construction parameter solution and serves as the optimal construction parameter.

[0116] See also Figure 5 The figure shows the schematic diagram of the triple-tube construction process of rotary jet piles; the triple-tube construction method is used, in which the entire length of the guide hole is divided into three parts: the covering layer, the effective reinforcement section and the ultra-deep guide hole; the covering layer is provided with a steel casing; in the effective reinforcement section, high-pressure "steam drum water" is used for cutting; in the ultra-deep guide hole, high-pressure mud is used for cutting and stirring; and the construction of the rotary jet pile is achieved through drilling rods and mud drainage; at this time, the construction cost of a single rotary jet pile is 13,200 yuan.

[0117] See also Figure 6 In an optional embodiment, the present invention provides a prediction system for the quality and optimal construction parameters of jet grouting piles, the system comprising an input device, an output device, a processor and a memory, the hardware facilities being interconnected, wherein the memory is used to store a computer program, the computer program comprising program instructions, the processor being configured to call the program instructions and execute the specific steps of the relevant embodiments of the method for predicting the quality and optimal construction parameters of jet grouting piles provided by the present invention. The prediction system for the quality and optimal construction parameters of jet grouting piles provided by the present invention has a complete structure, is objective and stable, and improves the overall applicability and practical application capability of the present invention.

[0118] The method and system for predicting the quality and optimal construction parameters of jet grouting piles proposed by the present invention have significant advantages, including:

[0119] 1. As the change of geological conditions will significantly affect the strength and anti-seepage characteristics of jet grouting piles, the present invention groups the performance characteristics of the strata through the K-medoids clustering algorithm, and quantitatively sorts the sensitivity of the strata parameters of each cluster based on the PAWN global sensitivity analysis, extracts the key sensitive parameter set, and establishes a partition prediction model. Through the input parameter dimensionality reduction and model structure adaptation mechanism, the prediction accuracy and engineering applicability of the optimization solution of the process parameters of the jet grouting piles are significantly improved.

[0120] 2. The present invention searches for the optimal hyperparameters of the random forest model based on a grid search algorithm, can quickly traverse hyperparameter combinations, avoid the blindness of traditional grid search and random search, effectively overcome the combinatorial explosion problem of traditional parameter optimization methods in high-dimensional space, and further improve the accuracy of the prediction model.

[0121] 3. The present invention is based on the AMGA algorithm. By setting the construction cost of the jet grouting piles as the target optimization function and setting the strength and anti-seepage performance of the jet grouting piles after consolidation and forming as constraints, the rationality of the selection of construction parameters of the jet grouting piles is comprehensively evaluated to ensure that the performance of the jet grouting piles meets the requirements of engineering specifications while reducing costs, thereby ensuring the stability and safety of the construction.

[0122] 4. The present invention takes into account the coupling relationship between different strata and jet grouting piles in actual engineering, ensuring that the optimal process parameter combination has engineering applicability under complex geological conditions where multiple strata are embedded, and significantly improves the reliability of predicting the optimal construction parameters of jet grouting piles. At the same time, the Pareto optimal solution set is optimized based on the actual engineering parameter values, and the optimized solution is substituted back into the numerical model for verification, ensuring the feasibility of the predicted solution in actual engineering.

[0123] In summary, the method and system for predicting the quality and optimal construction parameters of jet grouting piles provided by the method of the present invention combine sensitivity analysis, random forest and optimization algorithm, and improve the scientificity and effectiveness of the jet grouting pile construction process and related parameter selection in an algorithm-driven manner, thereby enhancing construction safety, reducing construction costs, and improving engineering benefits. The method of the present invention is easy to understand, simple to calculate, with a small workload, and is convenient for engineering application, providing a theoretical basis and technical support for the further development of engineering intelligent algorithm technology.

[0124] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A method for predicting the quality of jet grouting piles and optimal construction parameters, characterized in that: The steps include: Based on similar historical engineering cases, the construction stratum parameters, construction process parameters and jet grouting pile evaluation parameters were obtained to construct a data sample set; Classifying the soil of the construction stratum of the jet grouting pile to obtain different stratum soil categories, and obtaining sensitive parameters of the construction process of the different stratum soil categories; An orthogonal test is performed on the sensitive parameters of the construction process to obtain orthogonal test results, so as to establish a numerical simulation model for jet grouting pile construction and obtain a simulation data sample set; Based on the simulation data sample set, a jet grouting pile strength prediction model is constructed for each of the different stratum soil types, thereby obtaining a jet grouting pile quality prediction result; Establishing constraint conditions, and optimizing the prediction results of the jet grouting pile quality according to the optimization algorithm to obtain an optimal solution set that satisfies the constraint conditions; Combined with the actual jet grouting pile construction situation, the optimal solution set is optimized and screened according to the optimal solution screening method to obtain the optimal construction parameters.

2. The method for predicting the quality of jet grouting piles and optimal construction parameters according to claim 1, characterized in that: The data sample set is constructed by obtaining construction stratum parameters, construction process parameters and jet grouting pile evaluation parameters based on similar historical engineering cases, including: The construction stratum parameters include the stratum distribution and soil properties of the soil in the construction area where the jet grouting pile is located. The soil properties include physical indicators and mechanical properties of the soil. The physical indicators include natural density, water content, void ratio, dry density, liquid index and standard penetration number. The mechanical properties include compression modulus, shear strength and Poisson's ratio. The construction process parameters include nozzle diameter, lead hole diameter, cement content, water-cement ratio, shotcrete flow rate, air flow rate, water flow rate and lifting speed; The evaluation parameters of the jet grouting pile include the unconfined compressive strength and permeability coefficient of the jet grouting pile after consolidation and forming.

3. The method for predicting the quality of jet grouting piles and optimal construction parameters according to claim 1, characterized in that: The method of classifying the soil of the construction stratum of the jet grouting pile to obtain different soil types and obtaining sensitive parameters of the construction process of the different soil types includes: Classifying the construction stratum soil to obtain the different stratum soil types, including silt, silty soil, clay soil, plain fill soil and sandy soil; Based on the jet grouting pile evaluation parameters, a global sensitivity analysis is performed on the construction process parameters of the different stratum soil types to obtain the construction process sensitive parameters.

4. The method for predicting the quality of jet grouting piles and optimal construction parameters according to claim 1, characterized in that: The orthogonal test is performed on the sensitive parameters of the construction process to obtain orthogonal test results, so as to establish a numerical simulation model for jet grouting pile construction and obtain a simulation data sample set, including: Defining the value range of the sensitive parameters of the construction process based on the similar historical engineering cases, using the value range to expand the target unknown parameters, and constructing simulation test cases; Performing numerical simulation according to the simulation test case, thereby establishing the numerical simulation model of the jet grouting pile construction; The evaluation parameters of the jet grouting piles are obtained according to the numerical simulation model of the jet grouting pile construction as the simulation data sample set.

5. The method for predicting the quality of jet grouting piles and optimal construction parameters according to claim 1, characterized in that: The method of constructing jet grouting pile strength prediction models for different stratum soil types based on the simulation data sample set, and then obtaining jet grouting pile quality prediction results, includes: Normalizing the simulated data sample set to obtain normalized data; Obtaining initialization parameters of the jet grouting pile strength prediction model, and dividing the simulation data sample set into pure subsets by recursive splitting; Based on the pure subset, construct the jet grouting pile strength prediction model for the different stratum soil types respectively; Training and verifying the jet grouting pile strength prediction model to obtain the optimal model hyperparameter combination of the jet grouting pile strength prediction model; In the different stratum soil types, the jet grouting pile strength prediction model is used to establish a corresponding relationship between the construction process parameters and the jet grouting pile evaluation parameters as the jet grouting pile quality prediction result.

6. The method for predicting the quality of jet grouting piles and optimal construction parameters according to claim 5, characterized in that: The step of dividing the simulated data sample set into pure subsets by recursive splitting includes: The purity of the pure subset is measured using the Gini index, which includes: ； in, Pure subset The Gini index, is the number of categories, For categories, Pure subset Medium Category The sample proportion.

7. The method for predicting the quality of jet grouting piles and optimal construction parameters according to claim 5, characterized in that: The jet grouting pile strength prediction model is constructed for the different stratum soil types, including: ； in, To categorize the voting results, It means to take the category that makes the function value the largest. is the number of decision trees, For decision tree, is the indicator function, For the The prediction results output by the decision tree are: is the input sample, For category.

8. The method for predicting the quality of jet grouting piles and optimal construction parameters according to claim 1, characterized in that: The establishment of constraint conditions, and optimizing the prediction results of the jet grouting pile quality according to the optimization algorithm to obtain the optimal solution set that satisfies the constraint conditions, include: The jet grouting pile evaluation parameters are used as the constraint conditions, including: ； in, and As constraints, is the minimum unconfined compressive strength obtained based on the jet grouting pile strength prediction model, is the minimum unconfined compressive strength allowed by the project, is the maximum permeability coefficient obtained based on the jet grouting pile strength prediction model, It is the maximum permeability coefficient allowed in the project.

9. The method for predicting the quality of jet grouting piles and optimal construction parameters according to claim 1, characterized in that: The optimal solution set is optimized and screened according to the optimal solution screening method in combination with the actual jet grouting pile construction situation to obtain the optimal construction parameters, including: Obtaining the corresponding equation between the construction cost of the jet grouting pile and the construction process parameters includes: ； in, The construction cost of jet grouting piles is is the amount of dielectric material, For dielectric materials, is the unit volume cost of dielectric material, is the amount of dielectric material used; Under the constraints, according to the optimal solution screening method, a step-by-step verification mechanism is used to optimize and screen the optimal solution set to obtain the jet grouting pile construction plan with the lowest construction cost, so as to obtain the optimal construction parameters.

10. A prediction system for the quality of jet grouting piles and optimal construction parameters, characterized in that: The system includes an input device, an output device, a processor and a memory, wherein the input device, the output device, the processor and the memory are interconnected, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute the method for predicting the quality of jet grouting piles and optimal construction parameters as described in any one of claims 1-9.

Images