CN118171585B - Super-long pile foundation structure design and quality control method and system - Google Patents
Super-long pile foundation structure design and quality control method and system Download PDFInfo
- Publication number
- CN118171585B CN118171585B CN202410605350.4A CN202410605350A CN118171585B CN 118171585 B CN118171585 B CN 118171585B CN 202410605350 A CN202410605350 A CN 202410605350A CN 118171585 B CN118171585 B CN 118171585B
- Authority
- CN
- China
- Prior art keywords
- habitat
- habitats
- siv
- pile foundation
- fitness
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000013461 design Methods 0.000 title claims abstract description 31
- 238000003908 quality control method Methods 0.000 title claims abstract description 8
- 238000000034 method Methods 0.000 claims abstract description 56
- 238000013508 migration Methods 0.000 claims abstract description 50
- 230000005012 migration Effects 0.000 claims abstract description 50
- 238000005457 optimization Methods 0.000 claims abstract description 11
- 241000713311 Simian immunodeficiency virus Species 0.000 claims description 29
- 230000035772 mutation Effects 0.000 claims description 25
- 239000013598 vector Substances 0.000 claims description 23
- 230000008569 process Effects 0.000 claims description 22
- 238000010276 construction Methods 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 9
- 241000269799 Perca fluviatilis Species 0.000 claims description 6
- 238000002790 cross-validation Methods 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 4
- 230000009977 dual effect Effects 0.000 claims description 3
- 230000002068 genetic effect Effects 0.000 claims description 3
- 238000013507 mapping Methods 0.000 abstract 1
- 238000012163 sequencing technique Methods 0.000 description 8
- 230000004044 response Effects 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 238000011156 evaluation Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000008485 antagonism Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 239000003673 groundwater Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/13—Architectural design, e.g. computer-aided architectural design [CAAD] related to design of buildings, bridges, landscapes, production plants or roads
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
- G06N3/0455—Auto-encoder networks; Encoder-decoder networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/08—Probabilistic or stochastic CAD
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- Architecture (AREA)
- Civil Engineering (AREA)
- Structural Engineering (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to the field of pile foundation design, and particularly relates to a method and a system for structural design and quality control of an ultra-long pile foundation. The scheme combines a biological geography optimization algorithm with a USVM-RFE model, calculates the characteristic weight, and recursively removes the characteristic with smaller weight, increases the breadth of pile foundation design, and comprehensively measures the quality safety of the ultra-long pile foundation structure from the integral direction; calculating mobility by adopting a cosine migration model, mapping the fitness ratio to nonlinear probability distribution, and increasing smooth continuity; and a preferred strategy is adopted for the feature subset sorting list, and the preferred degree is controlled by adopting a preset proportion, so that the convergence speed and the searching performance of the algorithm are improved.
Description
Technical Field
The invention relates to the field of pile foundation design, in particular to a method and a system for structural design and quality control of an ultra-long pile foundation.
Background
The pile foundation is one of the most important basic forms in engineering construction, and the guarantee of the structural design and the engineering quality is an important link for guaranteeing the safety of the building. When the traditional computer-aided CAD system designs a pile foundation structure, uncertainty and risk in the construction process are not fully considered, power response caused by external loads such as earthquake, wind load and the like are not considered, and the problem of comprehensive consideration of the power response of the pile foundation is lacking; the problem of single linearity of the traditional mobility calculation method exists; the problem that the variation degree cannot be controlled by the biological geography optimization algorithm, and the optimal characteristics can disappear due to variation operation; the method has the problems of huge feature quantity and difficult feature selection.
Disclosure of Invention
Aiming at the situation, the invention provides a method and a system for designing and controlling the quality of an ultra-long pile foundation structure in order to overcome the defects of the prior art; aiming at the problems that uncertainty and risk in the construction process are not fully considered, power response caused by external loads such as earthquake, wind load and the like are not considered, and comprehensive consideration on the power response of the pile foundation is lacking in the traditional computer-aided CAD system when the pile foundation structure is designed, the scheme combines a biophysical optimization algorithm with a USVM-RFE model, calculates feature weights, simultaneously recursively removes the feature with smaller weight, increases the breadth of pile foundation design, and comprehensively measures the quality safety of the ultra-long pile foundation structure from the whole direction; aiming at the problem of single linearity of the traditional mobility calculation method, the scheme adopts a cosine migration model to calculate mobility, maps the fitness ratio to nonlinear probability distribution, and increases smooth continuity; aiming at the problem that the variation degree cannot be controlled by the biological geography optimization algorithm, and the optimal characteristic can disappear due to variation operation, the scheme adopts a preferred strategy for the characteristic subset sorting list, each iteration keeps a candidate overlength pile foundation historical design scheme with a better effect, the optimal solution in the current candidate overlength pile foundation historical design scheme is better utilized, the preferred degree is controlled by adopting a preset proportion, and the convergence speed and the searching performance of the algorithm are improved; aiming at the problems of huge feature quantity, difficult feature selection and easy occurrence of declining learning algorithm and overfitting of an ultra-long pile foundation historical design scheme, the scheme combines a biological geography optimization algorithm with a USVM-RFE model, selects the most relevant features, reduces irrelevant, noisy and redundant features, and compared with an SVM-RFE model, the USVM-RFE model reduces the requirement on the quantity of marked data sets, reduces algorithm complexity and improves feature utilization efficiency.
The technical scheme adopted by the invention is as follows: the invention provides a method and a system for designing and controlling the quality of an ultra-long pile foundation structure, wherein the method comprises the following steps:
step S1: collecting a data set, initializing a candidate scheme, and generating habitats;
step S2: evaluating fitness, namely selecting a fitness function to calculate and evaluate the fitness of all habitats according to the construction requirements, engineering budget and geological conditions of the engineering to which the ultra-long pile foundation to be designed belongs;
Step S3: sorting, namely sorting all habitats according to the descending order of the fitness, and outputting a descending order sorting table of the fitness;
step S4: migrating;
step S5: variation, generating a feature subset ordering list;
step S6: the method comprises the steps of reserving a current optimal solution, reserving habitats with preset proportions in front of a feature subset sorting list, wherein the habitats do not participate in migration and mutation operation of the next round;
Step S7: updating the fitness and fitness descending order sorting table of all habitats;
Step S8: and outputting an optimal candidate scheme of the project to which the ultra-long pile foundation to be designed belongs.
Further, in step S1, the collecting data set, initializing a candidate scheme, and generating habitat, specifically includes the following steps:
step S11: initializing parameters of a biological geography optimization algorithm, setting a maximum algebra Gmax, a maximum migration rate and an algebra G, marking the maximum migration rate as I and the maximum migration rate as E;
Step S12: collecting a data set, taking building requirements, engineering budgets and geological conditions as samples, and taking a history design scheme of an ultra-long pile foundation as a tag, wherein the data set consists of samples with corresponding tags;
step S13: initializing a candidate scheme, randomly generating a candidate super-long pile foundation design scheme obeying label distribution and corresponding prior probability according to a data set, recording the generated candidate super-long pile foundation design scheme as a candidate scheme, extracting features from the candidate scheme to generate a feature vector, recording the feature vector as a habitat, and recording any dimension in the habitat as a SIV.
Further, in step S4, the migration specifically includes the following steps:
Step S41: mobility refers to the probability that SIVs of habitats other than habitat h migrate to habitat h for any habitat h, and the mobility is calculated using a cosine migration model, with the following formula:
；
In the method, in the process of the invention, As a number of habitats,Is the firstThe mobility of the individual habitat,For the maximum mobility to be achieved,Is the firstThe fitness of the individual habitat,Maximum value for all habitat fitness;
Step S42: the migration rate is the probability that SIV of a habitat h migrates to other habitats except the habitat h for any habitat h, and the migration rate is calculated by using a cosine migration model, and the formula is as follows:
；
In the method, in the process of the invention, Is the firstThe rate of migration of individual habitats,Is the maximum migration rate;
step S43: performing migration operation, sequentially accessing all habitats according to the fitness descending order list, and recording the habitat being accessed as the habitat Randomly taking a decimal fraction of 0 to 1, and recording asIf (3)Then from the habitatSelecting one habitat from other habitats with probability proportional to mobility of other habitatsWill perch inMobility of (2) is recorded asIf (3)Then the next operation is carried out;
Step S44: replacing SIV if And is also provided withWill perch inSIV replacement to habitatIf SIV of (2)Or (b)And carrying out the next operation.
Further, in step S5, the mutation specifically includes the following steps:
step S51: initializing a counting probability, wherein the counting probability is the probability of the number of SIVs possessed by habitats, and if algebra G is equal to 0, the counting probability is assigned by using the prior probability;
step S52: the count probabilities for all habitats are updated as follows:
；
In the method, in the process of the invention, For the number of SIVs owned by the habitat,Is a habitat withThe counting probability at the time of the SIV,Is a habitat withThe counting probability at the time of the SIV,Is a habitat withThe counting probability at the time of the SIV,Is a habitat withMobility at the time of the individual SIVs,The migration rate when S SIVs are used as habitats;
Step S53: the mutation rate is the probability of binary mutation of SIV, and the mutation rate of all habitats is calculated by the following calculation formula:
；
In the method, in the process of the invention, Represent the firstThe rate of variation of the individual habitat,Represents the maximum variability of all habitats,Representing all of the habitatIs selected from the group consisting of a maximum probability,Representation ofIs a mathematical expectation of (a);
step S54: traversing all habitats to be proportional to Is a random choice of a habitat, defined as habitat；
Step S55: adopting USVM-RFE model to reduce the scale of habitat, namely the number of SIV;
Step S56: randomly selecting habitats SIV of random number;
step S57: binary variation of habitats using genetic algorithm The selected SIV in the (2) is subjected to mutation operation, a decimal from 0 to 1 is randomly taken and recorded asIf (3)The 0 in the SIV expressed in binary form is changed to 1 ifThen 1 in SIV expressed in binary form is changed to 0;
Step S58: algebraic G is incremented by one.
Further, in step S55, the use of USVM-RFE model to reduce the scale of habitat, i.e. reduce the number of SIVs, specifically includes the steps of;
step S551: the problem of reducing the habitat scale is converted into a corresponding quadratic programming QPP;
Step S552: applying a KKT condition to the dual pair of the quadratic programming QPP;
Step S553: defining a feature set as a set of SIV categories in all habitats, training and constructing USVM-RFE models by using k-fold cross-validation by using the data set and the feature set;
Step S554: the weight vector calculation formula is as follows:
；
In the method, in the process of the invention, Is the weight vector of the object,Is the total number of samples in the dataset that have tags,Is the total number of samples in the dataset without tags,Is from 1 toIs characterized by a self-increasing iteration index of (a),Is a Lagrangian multiplier vector in the range of interval 11,21,Is thatIs the first of (2)The dimensions of the dimensions,Is a label with the subscript j,Is marked by the subscriptA combined vector of the samples of (a) and SIV thereof;
Step S555: the weights are criteria for ranking, and are calculated from the weight vectors as follows:
；
In the method, in the process of the invention, Is a weight;
Step S556: deleting the SIV corresponding to the minimum weight;
Step S557: constructing a new USVM-RFE model by using SIV corresponding to the residual weight;
Step S558: repeating steps S553 to S557 until all SIVs are deleted;
Step S559: and sorting SIVs according to the deleted order, wherein the finally deleted SIVs are SIVs with the strongest correlation with the tags, and outputting a SIV sorting list.
Further, in step S8, the outputting of the optimal candidate scheme of the project to which the ultra-long pile foundation to be designed belongs specifically means that, according to the order of the descending order of fitness, the steps S2 to S7 are repeated in a loop iteration until the algebra G is greater than or equal to the maximum algebra Gmax or the fitness of the habitat is greater than a preset threshold, the feature vector of the habitat with the maximum fitness is converted into the candidate scheme and output, that is, the optimal candidate scheme is output.
The invention provides an ultra-long pile foundation structural design and quality control system, which comprises an initialization module, an evaluation fitness module, a sequencing module, a migration module, a variation module, a module for reserving the current optimal solution, an updating module and a module for outputting optimal candidate schemes;
The initialization module collects the data set, initializes the candidate scheme and generates habitat;
The adaptability evaluation module selects a fitness function to calculate and evaluate the fitness of all habitats according to the construction requirements, the project budget and the geological conditions of the project to which the ultra-long pile foundation to be designed belongs;
the sequencing module is used for sequencing all habitats according to the descending order of the fitness and outputting a descending order sequencing table of the fitness;
The migration module executes a migration operation on habitats;
The mutation module executes mutation operation on habitat;
The module for reserving the current optimal solution reserves habitats with preset proportions in front of the feature subset sorting list, and the habitats do not participate in the migration and mutation operation of the next round;
The updating module updates the fitness and fitness descending order ranking table of all habitats;
And the optimal candidate scheme outputting module outputs an optimal candidate scheme of the project to which the ultra-long pile foundation to be designed belongs.
The invention provides a method and a system for designing and controlling the quality of an ultra-long pile foundation structure, and the beneficial effects obtained by adopting the scheme are as follows:
(1) Aiming at the problems that uncertainty and risk in the construction process are not fully considered, power response caused by external loads such as earthquake, wind load and the like are not considered, and comprehensive consideration on the power response of the pile foundation is lacking in the traditional computer-aided CAD system when the pile foundation structure is designed, the scheme combines a biophysical optimization algorithm with a USVM-RFE model, calculates feature weights, simultaneously recursively removes the feature with smaller weight, increases the breadth of pile foundation design, and comprehensively measures the quality safety of the ultra-long pile foundation structure from the whole direction;
(2) Aiming at the problem of single linearity of the traditional mobility calculation method, the scheme adopts a cosine migration model to calculate mobility, maps the fitness ratio to nonlinear probability distribution, and increases smooth continuity;
(3) Aiming at the problem that the variation degree cannot be controlled by the biological geography optimization algorithm, and the optimal characteristic can disappear due to variation operation, the scheme adopts a preferred strategy for the characteristic subset sorting list, each iteration keeps a candidate overlength pile foundation historical design scheme with a better effect, the optimal solution in the current candidate overlength pile foundation historical design scheme is better utilized, the preferred degree is controlled by adopting a preset proportion, and the convergence speed and the searching performance of the algorithm are improved;
(4) Aiming at the problems of huge feature quantity, difficult feature selection and easy occurrence of declining learning algorithm and overfitting of an ultra-long pile foundation historical design scheme, the scheme combines a biological geography optimization algorithm with a USVM-RFE model, selects the most relevant features, reduces irrelevant, noisy and redundant features, and compared with an SVM-RFE model, the USVM-RFE model reduces the requirement on the quantity of marked data sets, reduces algorithm complexity and improves feature utilization efficiency.
Drawings
FIG. 1 is a schematic flow chart of a design and quality control method for ultra-long pile foundation structure provided by the invention;
fig. 2 is a flow chart of step S42 and step S5.
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the invention; all other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, the method and system for designing and controlling the quality of an ultra-long pile foundation structure provided by the invention comprise the following steps:
step S1: collecting a data set, initializing a candidate scheme, and generating habitats;
step S2: evaluating fitness, namely selecting a fitness function to calculate and evaluate the fitness of all habitats according to the construction requirements, engineering budget and geological conditions of the engineering to which the ultra-long pile foundation to be designed belongs;
Step S3: sorting, namely sorting all habitats according to the descending order of the fitness, and outputting a descending order sorting table of the fitness;
step S4: migrating;
step S5: variation, generating a feature subset ordering list;
step S6: the method comprises the steps of reserving a current optimal solution, reserving habitats with preset proportions in front of a feature subset sorting list, wherein the habitats do not participate in migration and mutation operation of the next round;
Step S7: updating the fitness and fitness descending order sorting table of all habitats;
Step S8: and outputting an optimal candidate scheme of the project to which the ultra-long pile foundation to be designed belongs.
In a second embodiment, referring to fig. 1, the method is based on the above embodiment, and in step S1, the collecting data set, initializing a candidate scheme, and generating habitat specifically includes the following steps:
step S11: initializing parameters of a biological geography optimization algorithm, setting a maximum algebra Gmax, a maximum migration rate and an algebra G, marking the maximum migration rate as I and the maximum migration rate as E;
Step S12: collecting a data set, taking building requirements, engineering budgets and geological conditions as samples, and taking a history design scheme of an ultra-long pile foundation as a tag, wherein the data set consists of samples with corresponding tags;
step S13: initializing a candidate scheme, randomly generating a candidate super-long pile foundation design scheme obeying label distribution and corresponding prior probability according to a data set, recording the generated candidate super-long pile foundation design scheme as a candidate scheme, extracting features from the candidate scheme to generate a feature vector, recording the feature vector as a habitat, and recording any dimension in the habitat as a SIV.
Embodiment three, referring to fig. 1 and 2, based on the above embodiment, in step S4, the migration specifically includes the following steps:
Step S41: mobility refers to the probability that SIVs of habitats other than habitat h migrate to habitat h for any habitat h, and the mobility is calculated using a cosine migration model, with the following formula:
；
In the method, in the process of the invention, As a number of habitats,Is the firstThe mobility of the individual habitat,For the maximum mobility to be achieved,Is the firstThe fitness of the individual habitat,Maximum value for all habitat fitness;
Step S42: the migration rate is the probability that SIV of a habitat h migrates to other habitats except the habitat h for any habitat h, and the migration rate is calculated by using a cosine migration model, and the formula is as follows:
；
In the method, in the process of the invention, Is the firstThe rate of migration of individual habitats,Is the maximum migration rate;
step S43: performing migration operation, sequentially accessing all habitats according to the fitness descending order list, and recording the habitat being accessed as the habitat Randomly taking a decimal fraction of 0 to 1, and recording asIf (3)Then from the habitatSelecting one habitat from other habitats with probability proportional to mobility of other habitatsWill perch inMobility of (2) is recorded asIf (3)Then the next operation is carried out;
Step S44: replacing SIV if And is also provided withWill perch inSIV replacement to habitatIf SIV of (2)Or (b)And carrying out the next operation.
Fourth embodiment, referring to fig. 1, based on the above embodiment, in step S5, the mutation generates a feature subset ranking list, which specifically includes the following steps:
step S51: initializing a counting probability, wherein the counting probability is the probability of the number of SIVs possessed by habitats, and if algebra G is equal to 0, the counting probability is assigned by using the prior probability;
step S52: the count probabilities for all habitats are updated as follows:
；
In the method, in the process of the invention, For the number of SIVs owned by the habitat,Is a habitat withThe counting probability at the time of the SIV,Is a habitat withThe counting probability at the time of the SIV,Is a habitat withThe counting probability at the time of the SIV,Is a habitat withMobility at the time of the individual SIVs,The migration rate when S SIVs are used as habitats;
Step S53: the mutation rate is the probability of binary mutation of SIV, and the mutation rate of all habitats is calculated by the following calculation formula:
；
In the method, in the process of the invention, Represent the firstThe rate of variation of the individual habitat,Represents the maximum variability of all habitats,Representing all of the habitatIs selected from the group consisting of a maximum probability,Representation ofIs a mathematical expectation of (a);
step S54: traversing all habitats to be proportional to Is a random choice of a habitat, defined as habitat；
Step S55: adopting USVM-RFE model to reduce the scale of habitat, namely the number of SIV;
Step S56: randomly selecting habitats SIV of random number;
step S57: binary variation of habitats using genetic algorithm The selected SIV in the (2) is subjected to mutation operation, a decimal from 0 to 1 is randomly taken and recorded asIf (3)The 0 in the SIV expressed in binary form is changed to 1 ifThen 1 in SIV expressed in binary form is changed to 0;
Step S58: algebraic G is incremented by one.
Fifth embodiment, referring to fig. 1, based on the above embodiment, in step S55, the method of reducing the size of habitat using USVM-RFE model, i.e. reducing the number of SIVs, specifically includes the following steps:
step S551: the problem of reducing the habitat scale is converted into a corresponding quadratic programming QPP;
Step S552: applying a KKT condition to the dual pair of the quadratic programming QPP;
Step S553: defining a feature set as a set of SIV categories in all habitats, training and constructing USVM-RFE models by using k-fold cross-validation by using the data set and the feature set;
Step S554: the weight vector calculation formula is as follows:
；
In the method, in the process of the invention, Is the weight vector of the object,Is the total number of samples in the dataset that have tags,Is the total number of samples in the dataset without tags,Is from 1 toIs characterized by a self-increasing iteration index of (a),Is a Lagrangian multiplier vector in the range of interval 11,21,Is thatIs the first of (2)The dimensions of the dimensions,Is a label with the subscript j,Is marked by the subscriptA combined vector of the samples of (a) and SIV thereof;
Step S555: the weights are criteria for ranking, and are calculated from the weight vectors as follows:
；
In the method, in the process of the invention, Is a weight;
Step S556: deleting the SIV corresponding to the minimum weight;
Step S557: constructing a new USVM-RFE model by using SIV corresponding to the residual weight;
Step S558: repeating steps S553 to S557 until all SIVs are deleted;
Step S559: and sorting SIVs according to the deleted order, wherein the finally deleted SIVs are SIVs with the strongest correlation with the tags, and outputting a SIV sorting list.
Embodiment six, referring to fig. 1, based on the foregoing embodiment, in step S8, the outputting an optimal candidate of the project to which the ultra-long pile foundation to be designed belongs includes the following operations: and (3) repeating the steps S2 to S7 according to the order of the fitness descending order ranking table until the algebra G is larger than or equal to the maximum algebra Gmax or the fitness of the habitat is larger than a preset threshold value, converting the feature vector of the habitat with the maximum fitness into a candidate scheme and outputting the candidate scheme, namely outputting the optimal candidate scheme.
An embodiment seven, referring to fig. 1 and fig. 2, based on the above embodiment, the system for designing and controlling the structure of an ultra-long pile foundation provided by the invention includes an initialization module, an evaluation fitness module, a sequencing module, a migration module, a mutation module, a module for retaining a current optimal solution, an update module, and a module for outputting an optimal candidate scheme;
The initialization module collects the data set, initializes the candidate scheme and generates habitat;
The adaptability evaluation module selects a fitness function to calculate and evaluate the fitness of all habitats according to the construction requirements, the project budget and the geological conditions of the project to which the ultra-long pile foundation to be designed belongs;
the sequencing module is used for sequencing all habitats according to the descending order of the fitness and outputting a descending order sequencing table of the fitness;
The migration module executes a migration operation on habitats;
The mutation module executes mutation operation on habitat;
The module for reserving the current optimal solution reserves habitats with preset proportions in front of the feature subset sorting list, and the habitats do not participate in the migration and mutation operation of the next round;
The updating module updates the fitness and fitness descending order ranking table of all habitats;
And the optimal candidate scheme outputting module outputs an optimal candidate scheme of the project to which the ultra-long pile foundation to be designed belongs.
Embodiment eight, referring to fig. 1, the embodiment is based on the above embodiment, and in step S11, the i=1, e=1, and algebraic g=0.
Embodiment nine, referring to fig. 1, which is based on the above embodiment, in step S12, the dataset will be as follows: 2:1, is divided into a training set, a verification set and a test set.
Embodiment ten, referring to fig. 1, based on the above embodiment, in step S13, candidate ultralong pile foundation design schemes and corresponding prior probabilities obeying the tag distribution are randomly generated according to the data set using the antagonism network GAN and the variational self-encoder VAE.
Embodiment eleven, referring to fig. 1, the embodiment is based on the above embodiment, and in step S2, the geological conditions include the bearing capacity of the soil layer and the groundwater level.
Embodiment twelve, see fig. 1, which is based on the above embodiment, in step S2, USVM times the accuracy of the cross-validation is used as a fitness function.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
The invention and its embodiments have been described above with no limitation, and the actual construction is not limited to the embodiments of the invention as shown in the drawings. In summary, if one of ordinary skill in the art is informed by this disclosure, a structural manner and an embodiment similar to the technical solution should not be creatively devised without departing from the gist of the present invention.
Claims (2)
1. A design and quality control method of an ultra-long pile foundation structure is characterized in that: the method comprises the following steps:
step S1: collecting a data set, initializing a candidate scheme, and generating habitats;
step S2: evaluating fitness, namely selecting a fitness function to calculate and evaluate the fitness of all habitats according to the construction requirements, engineering budget and geological conditions of the engineering to which the ultra-long pile foundation to be designed belongs;
Step S3: sorting, namely sorting all habitats according to the descending order of the fitness, and outputting a descending order sorting table of the fitness;
step S4: migrating;
step S5: variation, generating a feature subset ordering list;
step S6: the method comprises the steps of reserving a current optimal solution, reserving habitats with preset proportions in front of a feature subset sorting list, wherein the habitats do not participate in migration and mutation operation of the next round;
Step S7: updating the fitness and fitness descending order sorting table of all habitats;
step S8: outputting an optimal candidate scheme of the project to which the ultra-long pile foundation to be designed belongs;
in step S1, the initializing, generating habitat, specifically includes the following steps:
step S11: initializing parameters of a biological geography optimization algorithm, setting a maximum algebra Gmax, a maximum migration rate and an algebra G, marking the maximum migration rate as I and the maximum migration rate as E;
Step S12: collecting a data set, taking building requirements, engineering budgets and geological conditions as samples, and taking a history design scheme of an ultra-long pile foundation as a tag, wherein the data set consists of samples with corresponding tags;
Step S13: initializing a candidate scheme, randomly generating a candidate super-long pile foundation design scheme obeying label distribution and corresponding prior probability according to a data set, marking the generated candidate super-long pile foundation design scheme as a candidate scheme, extracting features from the candidate scheme to generate a feature vector, marking the feature vector as a habitat, and marking any dimension in the habitat as a SIV;
in step S5, the mutation specifically includes the following steps:
step S51: initializing a counting probability, wherein the counting probability is the probability of the number of SIVs possessed by habitats, and if algebra G is equal to 0, the counting probability is assigned by using the prior probability;
step S52: the count probabilities for all habitats are updated as follows:
；
In the method, in the process of the invention, For the number of SIVs owned by the habitat,Is a habitat withThe counting probability at the time of the SIV,Is a habitat withMobility at the time of the individual SIVs,The migration rate when S SIVs are used as habitats;
Step S53: the mutation rate is the probability of binary mutation of SIV, and the mutation rate of all habitats is calculated by the following calculation formula:
；
In the method, in the process of the invention, Represent the firstThe rate of variation of the individual habitat,Represents the maximum variability of all habitats,Representing all of the habitatIs selected from the group consisting of a maximum probability,Representation ofIs a mathematical expectation of (a);
step S54: traversing all habitats to be proportional to Is a random choice of a habitat, defined as habitat；
Step S55: adopting USVM-RFE model to reduce the scale of habitat, namely the number of SIV;
Step S56: randomly selecting habitats SIV of random number;
step S57: binary variation of habitats using genetic algorithm The selected SIV in the (2) is subjected to mutation operation, a decimal from 0 to 1 is randomly taken and recorded asIf (3)The 0 in the SIV expressed in binary form is changed to 1 ifThen 1 in SIV expressed in binary form is changed to 0;
step S58: algebraic G is increased by one;
in step S55, the adopted USVM-RFE model is adopted to reduce the scale of habitat, namely the number of SIVs is reduced, and the method specifically comprises the following steps of;
step S551: the problem of reducing the habitat scale is converted into a corresponding quadratic programming QPP;
Step S552: applying a KKT condition to the dual pair of the quadratic programming QPP;
Step S553: defining a feature set as a set of SIV categories in all habitats, training and constructing USVM-RFE models by using k-fold cross-validation by using the data set and the feature set;
Step S554: the weight vector calculation formula is as follows:
；
In the method, in the process of the invention, Is the weight vector of the object,Is the total number of samples in the dataset that have tags,Is the total number of samples in the dataset without tags,Is from 1 toIs characterized by a self-increasing iteration index of (a),Is a Lagrangian multiplier vector in the range of interval 11,21,Is thatIs the first of (2)The dimensions of the dimensions,Is marked by the subscriptIs used for the identification of the tag of (c),Is marked by the subscriptA combined vector of the samples of (a) and SIV thereof;
step S555: the weight is a ranking standard, and the weight is obtained by calculating the square sum of the weight vectors;
Step S556: deleting the SIV corresponding to the minimum weight;
Step S557: constructing a new USVM-RFE model by using SIV corresponding to the residual weight;
Step S558: repeating steps S553 to S557 until all SIVs are deleted;
Step S559: and sorting SIVs according to the deleted order, wherein the finally deleted SIVs are SIVs with the strongest correlation with the tags, and outputting a SIV sorting list.
2. The ultra-long pile foundation structure design and quality control method of claim 1, wherein: in step S4, the migration specifically includes the following steps:
Step S41: mobility refers to the probability that SIV of habitat other than habitat h migrates to habitat h for any habitat h, and the mobility is calculated by adopting a cosine migration model according to the following formula:
；
In the method, in the process of the invention, As a number of habitats,Is the firstThe mobility of the individual habitat,For the maximum mobility to be achieved,Is the firstThe fitness of the individual habitat,Maximum value for all habitat fitness;
Step S42: the migration rate is the probability that SIV of a habitat h migrates to other habitats except the habitat h for any habitat h, and the migration rate is calculated by using a cosine migration model, and the formula is as follows:
；
In the method, in the process of the invention, Is the firstThe rate of migration of individual habitats,Is the maximum migration rate;
step S43: performing migration operation, sequentially accessing all habitats according to the fitness descending order list, and recording the habitat being accessed as the habitat Randomly taking a decimal fraction of 0 to 1, and recording asIf (3)Then from the habitatSelecting one habitat from other habitats with probability proportional to mobility of other habitatsWill perch inMobility of (2) is recorded asIf (3)Then the next operation is carried out;
Step S44: replacing SIV if And is also provided withWill perch inSIV replacement to habitatIf SIV of (2)Or (b)And carrying out the next operation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410605350.4A CN118171585B (en) | 2024-05-16 | 2024-05-16 | Super-long pile foundation structure design and quality control method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410605350.4A CN118171585B (en) | 2024-05-16 | 2024-05-16 | Super-long pile foundation structure design and quality control method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118171585A CN118171585A (en) | 2024-06-11 |
| CN118171585B true CN118171585B (en) | 2024-08-27 |
Family
ID=91355021
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410605350.4A Active CN118171585B (en) | 2024-05-16 | 2024-05-16 | Super-long pile foundation structure design and quality control method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118171585B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103955754A (en) * | 2014-04-16 | 2014-07-30 | 江南大学 | Mold workshop scheduling method based on real-time production data collection |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105528643A (en) * | 2016-01-11 | 2016-04-27 | 盐城工学院 | Biogeographic optimization method for emergency service facility location problem |
| CN109388476A (en) * | 2018-10-12 | 2019-02-26 | 西安交通大学 | Docker based on multiple target biogeography optimization algorithm polymerize laying method with virtual machine |
| CN110751293B (en) * | 2019-09-29 | 2022-04-15 | 浙江财经大学 | Cloud manufacturing multi-task scheduling optimization method based on game theory |
| CN114444388A (en) * | 2022-01-16 | 2022-05-06 | 中国人民解放军空军工程大学 | Weapon target allocation optimization method based on improved biogeography |
-
2024
- 2024-05-16 CN CN202410605350.4A patent/CN118171585B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103955754A (en) * | 2014-04-16 | 2014-07-30 | 江南大学 | Mold workshop scheduling method based on real-time production data collection |
Non-Patent Citations (2)
| Title |
|---|
| A biogeography-based optimization for optimum discrete design of skeletal structures;Shahin Jalili et al.;Engineering Optimization;20151208;第1-24页 * |
| 基于改进BBO的Web文本聚类算法;李超雄;黄发良;温肖谦;;福建师范大学学报(自然科学版);20151120(第06期);第18-23页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118171585A (en) | 2024-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5888640B2 (en) | Photovoltaic power generation prediction apparatus, solar power generation prediction method, and solar power generation prediction program | |
| CN103105246A (en) | Greenhouse environment forecasting feedback method of back propagation (BP) neural network based on improvement of genetic algorithm | |
| CN112733417B (en) | Abnormal load data detection and correction method and system based on model optimization | |
| CN109215344B (en) | Method and system for urban road short-time traffic flow prediction | |
| CN111119282A (en) | Pressure monitoring point optimal arrangement method for water supply pipe network | |
| CN110059887B (en) | BP neural network risk identification method and system based on adaptive genetic algorithm | |
| CN113361761A (en) | Short-term wind power integration prediction method and system based on error correction | |
| CN112100927A (en) | Prediction method for slope deformation and soft soil foundation settlement based on GA-BP neural network | |
| CN108805346A (en) | A kind of hot continuous rolling force forecasting method based on more hidden layer extreme learning machines | |
| CN112836885B (en) | Combined load prediction method, combined load prediction device, electronic equipment and storage medium | |
| CN113240307B (en) | Power system economic dispatching method based on improved differential evolution algorithm | |
| CN117974221B (en) | Electric vehicle charging station location selection method and system based on artificial intelligence | |
| CN111008790A (en) | Hydropower station group power generation electric scheduling rule extraction method | |
| CN111832817A (en) | Small world echo state network time sequence prediction method based on MCP penalty function | |
| CN113971517A (en) | GA-LM-BP neural network-based water quality evaluation method | |
| CN114997040A (en) | Landslide simulation parameter inversion method based on improved support vector regression | |
| CN112183721B (en) | Construction method of combined hydrological prediction model based on self-adaptive differential evolution | |
| CN117810987A (en) | Prediction method, prediction device, prediction terminal and prediction storage medium for residential payload section | |
| CN111652413B (en) | Industrial power load prediction method based on multi-Agent distributed mass data processing | |
| CN118171585B (en) | Super-long pile foundation structure design and quality control method and system | |
| CN115277445B (en) | Micro-service system scheduling optimization method based on QoS (quality of service) target | |
| CN115081323A (en) | Method for solving multi-objective constrained optimization problem and storage medium thereof | |
| CN112785022B (en) | Method and system for excavating electric energy substitution potential | |
| CN114202063A (en) | Fuzzy neural network greenhouse temperature prediction method based on genetic algorithm optimization | |
| CN112330435A (en) | Credit risk prediction method and system for optimizing Elman neural network based on genetic algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |