Pile reverse design method and equipment based on multi-objective optimization and machine learning

Abstract

The embodiments of the present application relate to the field of reverse design of pile foundations, and specifically to a method and device for reverse design of pile foundations based on multi-objective optimization and machine learning. A specific implementation of the method includes: collecting a measured data set of pile foundations; performing update processing to obtain a measured updated data set of pile foundations; performing division processing on the measured updated data set of pile foundations to obtain a measured sample data set of pile foundations and a measured test data set of pile foundations; constructing a machine learning model for predicting pile foundation bearing capacity based on the measured sample data set of pile foundations and the measured test data set of pile foundations; generating an initial pile foundation design data set; optimizing the initial pile foundation design data set based on the machine learning model for predicting pile foundation bearing capacity to obtain target pile foundation design data; and sending the target pile foundation design data to the pile foundation construction control terminal to construct the pile foundations. This implementation improves the efficiency and accuracy of pile foundation design.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of pile foundation reverse design, and specifically to a pile foundation reverse design method and equipment based on multi-objective optimization and machine learning. Background Art

[0002] As the foundation of a building, piles play an important role in bearing the load of the building and resisting the settlement of the foundation. Pile design is a key link in civil engineering and directly affects the stability and safety of the building.

[0003] However, it is found in practice that when the above method is used for reverse design of pile foundation, the following technical problems often occur:

[0004] Traditional pile design methods usually rely on experience and test data, which is not only time-consuming and labor-intensive, but also difficult to deal with complex multi-objective optimization problems. Summary of the invention

[0005] The content of this application is used to introduce concepts in a brief form, which will be described in detail in the detailed implementation section below. The content of this application is not intended to identify the key features or essential features of the technical solution claimed for protection, nor is it intended to limit the scope of the technical solution claimed for protection.

[0006] Some embodiments of the present application propose a pile reverse design method, a computer device, and a computer-readable storage medium based on multi-objective optimization and machine learning to solve one or more of the technical problems mentioned in the above background technology section.

[0007] In a first aspect, some embodiments of the present application provide a method for reverse design of pile foundations based on multi-objective optimization and machine learning, the method comprising: collecting a measured data set of pile foundations; updating each measured data of pile foundations in the above-mentioned measured data set of pile foundations to generate measured updated data of pile foundations, and obtaining a measured updated data set of pile foundations; dividing the above-mentioned measured updated data set of pile foundations to obtain a measured sample data set of pile foundations and a measured test data set of pile foundations; constructing a machine learning model for predicting pile foundation bearing capacity based on the above-mentioned measured sample data set of pile foundations and the above-mentioned measured test data set of pile foundations; generating an initial pile foundation design data set; optimizing the above-mentioned initial pile foundation design data set based on the above-mentioned machine learning model for predicting pile foundation bearing capacity to obtain target pile foundation design data; and sending the above-mentioned target pile foundation design data to a pile foundation construction control terminal to construct pile foundations.

[0008] In a second aspect, the present application also provides a computer device, comprising a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, it implements the method described in any implementation of the first aspect.

[0009] In a third aspect, the present application further provides a computer-readable storage medium, on which a computer program is stored, wherein when the computer program is executed by a processor, the method described in any implementation of the first aspect is implemented.

[0010] The above-mentioned various embodiments of the present application have the following beneficial effects: through the reverse design method of piles based on multi-objective optimization and machine learning in some embodiments of the present application, the efficiency and accuracy of pile design can be improved. Specifically, the reason for the reduction in the efficiency and accuracy of pile design is that the traditional pile design method usually relies on experience and test data, which is not only time-consuming and labor-intensive, but also difficult to deal with complex multi-objective optimization problems. Based on this, the reverse design method of piles based on multi-objective optimization and machine learning in some embodiments of the present application, first, by collecting the pile database, including the physical parameters of the pile (such as pile body diameter, pile bottom diameter, pile length) and the two key parameters of the hyperbolic pile load-displacement relationship model (the inverse of the initial slope of the hyperbola, the inverse of the limit value of the hyperbola). Preprocessing the data ensures the integrity and reliability of the data and enhances the robustness of the model. Secondly, in the selection of the machine learning model, the present invention innovatively adopts the gradient boosting algorithm. These algorithms have strong nonlinear fitting ability and robustness, and can better capture the complex load-displacement relationship in pile design. Then, a new reverse design framework is constructed using the multi-objective optimization algorithm. The multi-objective optimization algorithm has good global search capability and convergence speed, and can effectively solve multi-objective optimization problems. Through the multi-objective optimization algorithm, a set of pile parameters is found so that the predicted bearing capacity and settlement are as close to the design requirements as possible. This reverse design method breaks through the limitations of relying on experience and trial and error in traditional design, and realizes intelligent and precise design. Finally, by constructing a reasonable objective function, the optimization algorithm can find the best balance point between multiple objectives, thereby finding the optimal combination of pile parameters. This multi-objective optimization method has significant innovation and practicality in pile design. Therefore, some pile reverse design methods based on multi-objective optimization and machine learning in this application can be efficient: by constructing a machine learning model, the mechanical properties of the pile (such as bearing capacity, settlement) can be quickly predicted, significantly shortening the design cycle. There is no need to conduct multiple physical experiments, shortening the project development time and effectively improving work efficiency. The use of machine learning algorithms significantly improves the accuracy of performance prediction. The multi-objective optimization algorithm ensures that the design parameters can meet multiple performance indicators. The multi-objective optimization algorithm can simultaneously optimize multiple performance indicators of the pile and synergistically improve the overall performance of the pile. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The above and other features, advantages and aspects of the embodiments of the present application will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that components and elements are not necessarily drawn to scale.

[0012] Figure 1is a flow chart of some embodiments of a pile reverse design method based on multi-objective optimization and machine learning according to the present application;

[0013] Figure 2 are some schematic diagrams of pile standardization load-displacement hyperbolic models according to some embodiments of the pile reverse design method based on multi-objective optimization and machine learning of the present application;

[0014] Figure 3 is a flowchart of optimizing an initial pile design data set according to some embodiments of the pile reverse design method based on multi-objective optimization and machine learning of the present application;

[0015] Figure 4 It is a schematic diagram of the structure of a computer device suitable for implementing some embodiments of the present application. DETAILED DESCRIPTION

[0016] Embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are only for exemplary purposes and are not intended to limit the scope of protection of the present application.

[0017] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other.

[0018] It should be noted that the concepts such as "first" and "second" mentioned in this application are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0019] It should be noted that the modifications of "one" and "plurality" mentioned in the present application are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0020] The names of the messages or data exchanged between multiple devices in the embodiments of the present application are only used for illustrative purposes and are not used to limit the scope of these messages or data.

[0021] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments.

[0022] Figure 1The process 100 of some embodiments of the pile foundation reverse design method based on multi-objective optimization and machine learning according to the present application is shown. The pile foundation reverse design method based on multi-objective optimization and machine learning includes the following steps:

[0023] Step 101, collecting a measured data set of foundation piles.

[0024] In some embodiments, the execution subject of the pile foundation reverse design method based on multi-objective optimization and machine learning can collect the pile foundation measured data set from the storage terminal through a wired connection or a wireless connection. Among them, the above-mentioned storage terminal can be a terminal for storing the pile foundation measured data set. The pile foundation measured data in the above-mentioned pile foundation measured data set can include but are not limited to: pile foundation measured characteristic data set. The pile foundation measured data in the above-mentioned pile foundation measured data set can represent the actual measurement data of a pile foundation. The pile foundation measured characteristic data in the above-mentioned pile foundation measured characteristic data set can be but are not limited to: pile type (driven pile, cast-in-place pile), pile body size (pile body radius, pile body diameter, pile bottom diameter and pile length) and corresponding performance indicators (pile foundation measured ultimate bearing capacity, settlement, vertical effective stress).

[0025] It should be noted that the above wireless connection methods may include but are not limited to 3G / 4G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods currently known or to be developed in the future.

[0026] Step 102 : performing update processing on each pile measurement data in the pile measurement data set to generate pile measurement update data, thereby obtaining the pile measurement update data set.

[0027] In some embodiments, the execution entity may update each pile foundation measurement data in the pile foundation measurement data set to generate pile foundation measurement update data, thereby obtaining the pile foundation measurement update data set.

[0028] In some optional implementations of some embodiments, the execution subject updates each pile measurement data in the pile measurement data set to generate pile measurement update data, and obtaining the pile measurement update data set may include the following steps:

[0029] The first step is to obtain the measured data of the pile end soil corresponding to the measured data of the pile foundation. The measured data of the pile end soil corresponding to the measured data of the pile foundation can be obtained from the above-mentioned storage terminal. The above-mentioned measured data of the pile end soil can represent the measured data of the soil around the pile foundation corresponding to the above-mentioned measured data of the pile foundation. The above-mentioned measured data of the pile end soil include but are not limited to: the initial stiffness of the load-displacement relationship of the pile end, Poisson's ratio, shear modulus, failure ratio, and cohesion.

[0030] The second step is to determine the first parameter value and the second parameter value based on the measured data of the pile foundation and the measured data of the pile end soil. The first parameter value and the second parameter value can be determined by the following formula based on the measured data of the pile foundation and the measured data of the pile end soil: .

[0031] in, Indicates the first parameter value mentioned above. The initial stiffness that represents the load-displacement relationship at the pile tip. Represents pi. The measured data of pile foundation include the measured characteristic data of pile foundation which characterizes the radius of the pile body. Indicates the second parameter value mentioned above. represents Poisson's ratio. represents the shear modulus. Indicates the destruction ratio. The measured data of pile foundation include the measured characteristic data of pile foundation which characterize the ultimate bearing capacity of pile end. Indicates cohesion. Dimensionless bearing capacity coefficient representing cohesion. The measured data of pile foundations include the measured characteristic data of pile foundations that characterize the vertical effective stress. A dimensionless bearing capacity coefficient representing the effect of lateral earth pressure at the pile tip plane.

[0032] The third step is to determine the vertical load value based on the measured data of the pile foundation, the first parameter value and the second parameter value. The vertical load value can be determined by the pile foundation standardized load-displacement hyperbola model based on the first parameter value and the second parameter value.

[0033] As an example, the above pile normalized load-displacement hyperbolic model can be referred to Figure 2 Schematic diagrams of pile normalized load-displacement hyperbolic models according to some embodiments of the pile reverse design method based on multi-objective optimization and machine learning of the present application are shown. Figure 2 As shown in the figure Indicates the vertical load value. The measured data of pile foundations include measured characteristic data of pile foundations that characterize the measured ultimate bearing capacity of pile foundations. represents the pile end stress. Represents the pile end displacement.

[0034] In the fourth step, the first parameter value, the second parameter value and the vertical load value are respectively normalized to obtain the normalized value of the first parameter, the normalized value of the second parameter and the normalized value of the vertical load. The first parameter value, the second parameter value and the vertical load value can be respectively normalized by a preset normalization algorithm to obtain the normalized value of the first parameter, the normalized value of the second parameter and the normalized value of the vertical load.

[0035] As an example, the above-mentioned preset normalization algorithm may be a maximum-minimum value normalization algorithm.

[0036] The fifth step is to perform normalization processing on each of the pile measured characteristic data in the pile measured characteristic data set included in the pile measured data to generate pile measured normalized data, thereby obtaining the pile measured normalized data set. In particular, each of the pile measured characteristic data in the pile measured characteristic data set included in the pile measured data can be normalized by the preset normalization algorithm to generate pile measured normalized data, thereby obtaining the pile measured normalized data set.

[0037] The sixth step is to fuse the normalized value of the first parameter, the normalized value of the second parameter, the normalized value of the vertical load and the normalized data set of the pile foundation to obtain the measured update data of the pile foundation. The first parameter normalized value, the second parameter normalized value, the normalized value of the vertical load and the normalized data set of the pile foundation are fused to obtain the measured update data of the pile foundation, which can be: the normalized value of the first parameter, the normalized value of the second parameter, the normalized value of the vertical load and the normalized data set of the pile foundation are determined as the normalized value of the first parameter, the normalized value of the second parameter, the normalized value of the vertical load and the normalized data set of the pile foundation measured, which are included in the pile foundation measured update data.

[0038] Step 103 , dividing the pile foundation measured update data set to obtain a pile foundation measured sample data set and a pile foundation measured test data set.

[0039] In some embodiments, the execution subject may divide and process the pile foundation measured update data set to obtain a pile foundation measured sample data set and a pile foundation measured test data set. The division and processing of the pile foundation measured update data set to obtain a pile foundation measured sample data set and a pile foundation measured test data set may be: randomly selecting a preset number of pile foundation measured update data from the pile foundation measured update data set as the pile foundation measured sample data set. Then, each pile foundation measured update data in the pile foundation measured update data set except the pile foundation measured sample data set is determined as the pile foundation measured test data set.

[0040] As an example, the preset number may be the product of the number of pile foundation measurement update data in the pile foundation measurement update data set and a preset ratio. The preset ratio may be 80%.

[0041] Step 104: construct a machine learning model for predicting pile bearing capacity based on the pile measured sample data set and the pile measured test data set.

[0042] In some embodiments, the execution subject may construct a machine learning model for predicting pile bearing capacity based on the pile measured sample data set and the pile measured test data set. The construction of the machine learning model for predicting pile bearing capacity based on the pile measured sample data set and the pile measured test data set may be as follows: first, the pile measured sample data set is used as sample data, and the initial machine learning model for predicting pile bearing capacity is trained through a preset adjustment algorithm. Then, the pile measured test data set is used as test data, and the pre-trained initial machine learning model for predicting pile bearing capacity is optimized through a preset optimization algorithm. Finally, the optimized initial machine learning model for predicting pile bearing capacity is determined as the machine learning model for predicting pile bearing capacity.

[0043] As an example, the preset adjustment algorithm may be, but is not limited to, at least one of the following: a GBDT (GradientBoosting Decision Tree) algorithm, an XGBoost (eXtreme Gradient Boosting, an optimized distributed gradient boosting library) algorithm, or a LightGBM (Light Gradient Boosting Machine, a lightweight gradient boosting machine) algorithm. The preset optimization algorithm may be a Bayesian optimization algorithm.

[0044] Step 105: Generate an initial pile foundation design data set.

[0045] In some embodiments, the execution entity may generate an initial pile foundation design data set. The initial pile foundation design data may be randomly generated to obtain an initial pile foundation design data set. The initial pile foundation design data in the initial pile foundation design data set may include but is not limited to: an initial pile foundation design feature data set. The initial pile foundation design feature data in the initial pile foundation design feature data set may include but is not limited to at least one of the following: pile type (driven pile, cast-in-place pile), soil properties around the pile (cohesionless soil, cohesive soil), pile body size (pile body radius, pile body diameter, pile bottom diameter and pile length) and corresponding performance indicators (bearing capacity, settlement, vertical effective stress).

[0046] Step 106, based on the machine learning model for predicting pile bearing capacity, the initial pile design data set is optimized to obtain target pile design data.

[0047] In some embodiments, the execution entity may optimize the initial pile design data set based on the machine learning model for predicting pile bearing capacity to obtain target pile design data.

[0048] In practice, the above-mentioned machine learning model for predicting pile bearing capacity is used to optimize the above-mentioned initial pile design data set to obtain the target pile design data. The specific implementation process can be referred to Figure 3 The flowchart of optimizing the initial pile design data set according to some embodiments of the pile reverse design method based on multi-objective optimization and machine learning of the present application is shown. Figure 3 As shown, based on the machine learning model for predicting pile bearing capacity, the above-mentioned initial pile design data set is optimized to obtain the target pile design data, which can be: first, the foundation bearing capacity model is established as the first objective function, the pile side friction resistance is considered as the second objective function, and the pile end resistance is considered as the third objective function, with the maximization of bearing capacity as the first optimization goal. Then, the safety constraint, the construction constraint, and the economic constraint are used as constraints, and the stress does not exceed the allowable value is used as the second optimization goal. Then, according to the multi-objective optimization algorithm, the optimal solution (i.e., the target pile design data) is obtained according to the first optimization goal and the second optimization goal. Specifically, the multi-objective optimization algorithm process can be: the first step is to initialize the population. The second step is to evaluate the fitness to perform non-dominated sorting on the initialized population. The third step is to calculate the crowding degree of the population individuals to select the population individuals from each population individual as the parent sub-population. The fourth step is to perform crossover inheritance and mutation operations on the parent sub-population to generate a new population. The fifth step is to perform non-dominated sorting on the new population. The fifth step is to determine whether the current state meets the termination condition (for example, reaching the specified number of evolutionary generations or finding the optimal solution that meets the stop condition). When the current state does not meet the termination condition, the crowding degree of the population individuals is calculated again to select population individuals from each population individual of the new population as the parent sub-population. When the current state meets the termination condition, the optimal solution in the new population is output as the optimal result.

[0049] In some optional implementations of some embodiments, the execution subject optimizes the initial pile design data set based on the machine learning model for predicting pile bearing capacity to obtain target pile design data, which may include the following steps:

[0050] The first step is to screen and process the initial pile design data set based on the machine learning model for predicting pile bearing capacity to obtain a first pile design data set.

[0051] In the second step, for the first pile design data set, the following optimization sub-steps are performed:

[0052] The first sub-step is to perform cross-fusion processing on each first pile design data in the first pile design data set to obtain a second pile design data set. The second pile design data set can be obtained by performing cross-fusion processing on each first pile design data in the first pile design data set using a preset cross-fusion algorithm.

[0053] As an example, the above-mentioned preset cross-fusion algorithm may be a genetic algorithm.

[0054] The second sub-step is to perform mutation processing on each second pile design data in the second pile design data set to obtain a third pile design data set. The third pile design data set can be obtained by performing mutation processing on each second pile design data in the second pile design data set using a preset mutation algorithm.

[0055] As an example, the above-mentioned preset mutation algorithm may be a genetic algorithm.

[0056] The third sub-step is to fuse the second pile design data set and the third pile design data set to obtain a fused pile design data set. The above-mentioned fusing the second pile design data set and the third pile design data set to obtain a fused pile design data set may be: first, each second pile design data in the second pile design data set is determined as fused pile design data. Secondly, each third pile design data in the third pile design data set is determined as fused pile design data. Finally, each determined fused pile design data is determined as a fused pile design data set.

[0057] The fourth sub-step is to filter and process the fused pile design data based on the machine learning model for predicting pile bearing capacity, and obtain a filtered pile design data set and a target congestion value set. The specific implementation method for generating the above-mentioned filtered pile design data set and the target congestion value set and the technical effects brought about by them can refer to the relevant steps of generating the first pile design data set in the above-mentioned embodiment, and will not be repeated here.

[0058] In the third step, in response to determining that the selected pile design data set meets the preset optimization condition, the selected pile design data corresponding to the largest target congestion value in the target congestion value set is determined as the target foundation design data. The preset optimization condition may be that each target congestion value in the target congestion value set corresponding to the selected pile design data set is greater than a preset threshold.

[0059] As an example, the preset threshold may be 0.9.

[0060] Optionally, the execution subject may also determine the pile design dataset as the first pile design dataset in response to determining that the fused pile design dataset does not satisfy a preset optimization condition, so as to execute the optimization step again.

[0061] In some optional implementations of some embodiments, the execution subject screens the initial pile design data set based on the machine learning model for predicting pile bearing capacity to obtain a first pile design data set, which may include the following steps:

[0062] The first step is to perform non-dominated sorting on each initial pile design data set in the initial pile design data set based on the machine learning model for predicting pile bearing capacity, the preset objective function and the preset constraints to obtain an initial pile design data sequence. Among them, based on the machine learning model for predicting pile bearing capacity, the preset objective function and the preset constraints, a preset multi-objective optimization algorithm can be used to perform non-dominated sorting on each initial pile design data set in the initial pile design data set to obtain an initial pile design data sequence.

[0063] As an example, the preset multi-objective optimization algorithm may be a NSGA-II (nondominated sorting genetic algorithm-II, second generation non-dominated sorting genetic) algorithm. The preset objective function may include but is not limited to at least one of the following: pile side friction resistance function and pile end resistance function. The preset constraint conditions may include but are not limited to: the importance coefficient of the pile foundation structure is greater than or equal to 1, the maximum water-cement ratio of the pile foundation structure concrete is 0.60, and the construction cost of the pile foundation is less than one million.

[0064] The second step is to determine the congestion value corresponding to each initial pile design data in the initial pile design data sequence to obtain a congestion value set. The congestion value corresponding to each initial pile design data in the initial pile design data sequence can be determined by the preset multi-objective optimization algorithm to obtain a congestion value set.

[0065] The third step is to sort the initial pile design data in the initial pile design data sequence based on the above-mentioned congestion value set to obtain a target pile design data sequence. The initial pile design data in the initial pile design data sequence can be sorted in descending order of the congestion values ​​corresponding to the initial pile design data in the initial pile design data sequence to obtain the target pile design data sequence.

[0066] The fourth step is to determine the first target number of target pile design data in the target pile design data sequence as the first pile design data set, wherein the target number can be the product of the target pile design data and the preset cross ratio.

[0067] As an example, the preset cross ratio may be 0.5.

[0068] Step 107, sending the target pile design data to the pile construction control terminal to construct the piles.

[0069] In some embodiments, the execution entity may send the target pile design data to a pile construction control terminal to construct the piles, wherein the pile construction control terminal may be a terminal for controlling the construction of the piles.

[0070] The above-mentioned various embodiments of the present application have the following beneficial effects: through the reverse design method of piles based on multi-objective optimization and machine learning in some embodiments of the present application, the efficiency and accuracy of pile design can be improved. Specifically, the reason for the reduction in the efficiency and accuracy of pile design is that the traditional pile design method usually relies on experience and test data, which is not only time-consuming and labor-intensive, but also difficult to deal with complex multi-objective optimization problems. Based on this, the reverse design method of piles based on multi-objective optimization and machine learning in some embodiments of the present application, first, by collecting the pile database, including the physical parameters of the pile (such as pile body diameter, pile bottom diameter, pile length) and the two key parameters of the hyperbolic pile load-displacement relationship model (the inverse of the initial slope of the hyperbola, the inverse of the limit value of the hyperbola). Preprocessing the data ensures the integrity and reliability of the data and enhances the robustness of the model. Secondly, in the selection of the machine learning model, the present invention innovatively adopts the gradient boosting algorithm. These algorithms have strong nonlinear fitting ability and robustness, and can better capture the complex load-displacement relationship in pile design. Then, a new reverse design framework is constructed using the multi-objective optimization algorithm. The multi-objective optimization algorithm has good global search capability and convergence speed, and can effectively solve multi-objective optimization problems. Through the multi-objective optimization algorithm, a set of pile parameters is found so that the predicted bearing capacity and settlement are as close to the design requirements as possible. This reverse design method breaks through the limitations of relying on experience and trial and error in traditional design, and realizes intelligent and precise design. Finally, by constructing a reasonable objective function, the optimization algorithm can find the best balance point between multiple objectives, thereby finding the optimal combination of pile parameters. This multi-objective optimization method has significant innovation and practicality in pile design. Therefore, some pile reverse design methods based on multi-objective optimization and machine learning in this application can be efficient: by constructing a machine learning model, the mechanical properties of the pile (such as bearing capacity, settlement) can be quickly predicted, significantly shortening the design cycle. There is no need to conduct multiple physical experiments, shortening the project development time and effectively improving work efficiency. The use of machine learning algorithms significantly improves the accuracy of performance prediction. The multi-objective optimization algorithm ensures that the design parameters can meet multiple performance indicators. The multi-objective optimization algorithm can simultaneously optimize multiple performance indicators of the pile and synergistically improve the overall performance of the pile.

[0071] The present application also provides a computer device 400. Figure 4 As shown, the computer device 400 includes: a bus 401, a processor 402, a memory 403 and a communication interface 404. The processor 402, the memory 403 and the communication interface 404 communicate with each other through the bus 401. The computer device 400 can be a server or a terminal device. It should be understood that the present application does not limit the number of processors and memories in the computer device 400.

[0072] The bus 401 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 The bus 401 is represented by only one line, but it does not mean that there is only one bus or one type of bus. The bus 401 may include a path for transmitting data between various components of the computer device 400 (for example, the memory 403, the processor 402, and the communication interface 404).

[0073] The processor 402 may include any one or more of a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).

[0074] The memory 403 may include a volatile memory, such as a random access memory (RAM). The memory 403 may also include a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid state drive (SSD).

[0075] The memory 403 stores executable program codes, and the processor 402 executes the executable program codes to respectively implement the functions of the aforementioned acquisition module, sampling module, determination module, and mixing module, thereby implementing the aforementioned method for reverse design of piles based on multi-objective optimization and machine learning. That is, the memory 403 stores instructions for executing the aforementioned method for reverse design of piles based on multi-objective optimization and machine learning.

[0076] The communication interface 404 uses a transceiver module such as, but not limited to, a network interface card or a transceiver to implement communication between the computer device 400 and other devices or a communication network.

[0077] An embodiment of the present application also provides a chip, which includes a processor and a data interface. The processor reads instructions stored in a memory through the data interface to execute the above-mentioned pile reverse design method based on multi-objective optimization and machine learning.

[0078] The embodiment of the present application also provides a computer-readable storage medium. The above-mentioned computer-readable storage medium can be any available medium that can be stored by a computing device or a data storage device such as a data center containing one or more available media. The above-mentioned available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid-state hard disk). The computer-readable storage medium includes instructions, and the above-mentioned instructions instruct the computing device to execute the above-mentioned pile reverse design method based on multi-objective optimization and machine learning.

[0079] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0080] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the protection scope of the technical solutions of the embodiments of the present application.

Claims

1. A pile reverse design method based on multi-objective optimization and machine learning, comprising: Collecting a pile foundation measured data set, wherein the pile foundation measured data in the pile foundation measured data set includes: a pile foundation measured feature data set; performing updating processing on each pile measurement data in the pile measurement data set to generate pile measurement update data, thereby obtaining the pile measurement update data set; The pile foundation measured update data set is divided and processed to obtain a pile foundation measured sample data set and a pile foundation measured test data set; Based on the pile foundation measured sample data set and the pile foundation measured test data set, construct a machine learning model for predicting pile foundation bearing capacity; Generate initial pile design data set; Based on the predicted pile bearing capacity machine learning model, the initial pile design data set is optimized to obtain target pile design data; Sending the target foundation pile design data to a foundation pile construction control terminal to construct the foundation pile; The updating process of each pile measurement data in the pile measurement data set to generate pile measurement update data includes: Obtain the measured data of the pile end soil corresponding to the measured data of the foundation pile; Determining a first parameter value and a second parameter value based on the measured data of the foundation pile and the measured data of the pile end soil; Determine a vertical load value based on the measured data of the foundation pile, the first parameter value and the second parameter value; Normalizing the first parameter value, the second parameter value, and the vertical load value respectively to obtain a first parameter normalized value, a second parameter normalized value, and a vertical load normalized value; Normalizing each pile measured characteristic data in the pile measured characteristic data set included in the pile measured data to generate pile measured normalized data, thereby obtaining the pile measured normalized data set; The first parameter normalized value, the second parameter normalized value, the vertical load normalized value and the pile foundation measured normalized data set are fused to obtain pile foundation measured updated data.

2. The pile reverse design method based on multi-objective optimization and machine learning according to claim 1, wherein: The initial pile design data set is optimized based on the predicted pile bearing capacity machine learning model to obtain target pile design data, including: Based on the machine learning model for predicting pile bearing capacity, the initial pile design data set is screened to obtain a first pile design data set; For the first pile design data set, the following optimization steps are performed: Performing cross-fusion processing on each first pile design data set in the first pile design data set to obtain a second pile design data set; performing mutation processing on each second pile design data in the second pile design data set to obtain a third pile design data set; fusing the second pile design data set and the third pile design data set to obtain a fused pile design data set; Based on the machine learning model for predicting pile bearing capacity, the fused pile design data is screened and processed to obtain the screened pile design data set and the target congestion value set; In response to determining that the screened pile design data set meets the preset optimization condition, the screened pile design data corresponding to the largest target congestion value in the target congestion value set is determined as the target pile design data.

3. The pile reverse design method based on multi-objective optimization and machine learning according to claim 2, wherein: The method further comprises: In response to determining that the fused pile foundation design data set does not satisfy the preset optimization condition, the pile foundation design data set is determined as the first pile foundation design data set to perform the optimization step again.

4. The method for reverse design of pile foundation based on multi-objective optimization and machine learning according to claim 2, wherein: The method of screening the initial pile design data set based on the predicted pile bearing capacity machine learning model to obtain a first pile design data set includes: Based on the predicted pile bearing capacity machine learning model, a preset objective function and preset constraints, each initial pile design data in the initial pile design data set is subjected to non-dominated sorting processing to obtain an initial pile design data sequence; Determine the congestion value corresponding to each initial pile design data in the initial pile design data sequence to obtain a congestion value set; Based on the congestion value set, each initial pile design data in the initial pile design data sequence is sorted to obtain a target pile design data sequence; The first target number of target pile design data in the target pile design data sequence is determined as a first pile design data set.

5. A computer device, wherein: The computer device comprises a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 4 are implemented.

6. A computer-readable storage medium, wherein: The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.

Images