CN113821863B - Method for predicting vertical ultimate bearing capacity of pile foundation - Google Patents
Method for predicting vertical ultimate bearing capacity of pile foundation Download PDFInfo
- Publication number
- CN113821863B CN113821863B CN202111381622.XA CN202111381622A CN113821863B CN 113821863 B CN113821863 B CN 113821863B CN 202111381622 A CN202111381622 A CN 202111381622A CN 113821863 B CN113821863 B CN 113821863B
- Authority
- CN
- China
- Prior art keywords
- fitness
- value
- generation
- neural network
- network model
- 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
- 238000000034 method Methods 0.000 title claims abstract description 22
- 238000003062 neural network model Methods 0.000 claims abstract description 43
- 238000012549 training Methods 0.000 claims abstract description 9
- 238000012360 testing method Methods 0.000 claims description 25
- 238000004364 calculation method Methods 0.000 claims description 13
- 230000006870 function Effects 0.000 description 10
- 230000000052 comparative effect Effects 0.000 description 7
- 230000003068 static effect Effects 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 4
- 150000001875 compounds Chemical class 0.000 description 2
- PBKSAWGZZXKEBJ-UHFFFAOYSA-N cyclopenta-1,3-diene;4-cyclopenta-2,4-dien-1-ylphenol;iron(2+) Chemical compound [Fe+2].C=1C=C[CH-]C=1.C1=CC(O)=CC=C1[C-]1C=CC=C1 PBKSAWGZZXKEBJ-UHFFFAOYSA-N 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000002068 genetic effect Effects 0.000 description 2
- 238000000691 measurement method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
Images
Classifications
-
- 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
- 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
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield 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
- G06N3/084—Backpropagation, e.g. using gradient descent
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 provides a method for predicting the vertical ultimate bearing capacity of a pile foundation, which comprises the following steps of S1, establishing an initial population; step S2, calculating the fitness value of each individual in the initial population by adopting the fitness function of the improved radial movement algorithm, determining the current optimal position by comparing the fitness values of the individuals one by one, and defining the current optimal position as the initial central position; step S3, adopting the update condition at the firstk. + -. 0.5 of the center position (x jmax‑x jmin）w k Generating a new pre-position point in the range, calculating a fitness value, and updating position information; step S4, determining a global optimal position, and calculating a fitness value of the global optimal position; and step S5, assigning the weight and the threshold value of the global optimal position to the BP neural network model, and training and simulating the BP neural network model to obtain the optimal predicted value of the vertical ultimate bearing capacity of the pile foundation. The invention is convenient for ensuring that the predicted value is closer to the true value.
Description
Technical Field
The invention relates to the technical field of pile foundation bearing capacity prediction, in particular to a method for predicting vertical ultimate bearing capacity of a pile foundation.
Background
The pile foundation is a foundation form with high bearing capacity, wide application range and long history. With the continuous development of social economy, the pile foundation is widely applied to high-rise buildings, ports and bridge engineering. When the method is applied, the vertical ultimate bearing capacity of the pile foundation is an important index for measuring the quality of the pile foundation. The vertical ultimate bearing capacity of the pile foundation is related to various factors such as the strength of a pile body, the property of soil around the pile, the construction process and the like, and a theoretical formula and a numerical calculation method which can comprehensively consider all the factors are not available. The static load test is the most direct and reliable method for measuring the bearing capacity of the pile foundation. However, because of the long time and high cost, the static test is generally used in important engineering and cannot be generally used. In addition, the load-settlement curve of some pile foundations is slow, and the limit state is difficult to reach during static load test, so that the limit bearing capacity of the pile foundations cannot be measured. And the dynamic measurement method testing principle is inconsistent with the load transmission mechanism of the pile foundation, and if the dynamic measurement method is adopted to measure the bearing capacity of the pile foundation, a certain measurement error can be generated.
The artificial neural network is the earliest method for predicting the bearing capacity of the pile foundation based on measured data. The bp (back propagation) algorithm is also called an error back propagation algorithm, and is a supervised learning algorithm in an artificial neural network. The method has strong nonlinear mapping capability, can approach any function theoretically, has strong flexibility, and can adjust parameters according to different conditions. However, the method is easy to fall into a local minimum value during prediction, cannot effectively search for the multi-peak function, and is low in convergence speed and poor in search result stability.
In summary, a method for predicting the vertical ultimate bearing capacity of the pile foundation is needed to solve the problems that the BP neural network in the prior art is easy to fall into a local minimum value, the convergence rate is low, and the search result is unstable.
Disclosure of Invention
The invention aims to provide a method for predicting the vertical ultimate bearing capacity of a pile foundation, which has the following specific technical scheme:
a method for predicting the vertical ultimate bearing capacity of a pile foundation comprises the following steps:
s1, taking the weight and the threshold of the BP neural network model as position point information of an improved radial movement algorithm, and establishing an initial population;
step S2, calculating the fitness value of each individual in the initial population by adopting the fitness function of the improved radial movement algorithm, determining the current optimal position by comparing the fitness values of the individuals one by one, and defining the current optimal position as the initial central position; the fitness function is a performance function of the BP neural network model, and specifically comprises the following steps:
wherein,Nthe number of test samples for the BP neural network model;t i for BP neural network modeliThe predicted value of the output variable of each test sample;y i for BP neural network modeliThe actual value of each test sample output variable;
step S3, adopting the update condition at the firstk. + -. 0.5 of the center position (x jmax - x jmin）w k Generating a new pre-position point in the range, calculating a fitness value, and updating position information; wherein,krefers to the current iteration number;x jminis as followsjThe minimum value of the individual weights or thresholds;x jmaxis as followsjThe maximum value of the individual weights or thresholds,w k is the inertia weight;
step S4, determining the global optimal position according to the position information updated in the step S3, and calculating the fitness value of the global optimal position; if the iteration times reach the upper limit or the fitness value of the global optimal position is less than 0.001, ending the updating; otherwise, repeating the steps S3-S4 until the updating is finished;
and step S5, assigning the weight and the threshold value of the global optimal position to the BP neural network model, training and simulating the BP neural network model to obtain an optimal predicted value of the vertical ultimate bearing capacity of the pile foundation, and ending the algorithm.
Preferably, the step S1 includes the steps of:
s1.1, calculating the total number of the weight values and the threshold values of the BP neural network model, and specifically obtaining the weight values and the total number of the threshold values through the following calculation formula:
nod=w 1+w 2+w 3formula (2)
w 1=n 1 × n 2 Formula (3)
w 2=n 2 × n 3 Formula (4)
w 3=n 2 + n 3 Formula (5)
Wherein,nodthe total number of the weight and the threshold value;w 1the number of the weights from the input layer to the hidden layer;w 2the number of weights from the hidden layer to the output layer;w 3is the total number of the threshold values;n 1the number of input layers;n 2the number of hidden layers;n 3the number of output layers;
step S1.2, determining value ranges of weight and thresholdx jmaxAndx jminwherein, 1 is less than or equal to j ≤ nod；
Step S1.3, randomly generating in the value range of the step S1.2nopAn initial position point ofnopEstablishing an initial population by using initial position points, wherein the numerical information of the initial position points is obtained by a calculation formula (6);
the calculation formula (6) isX i,j =x jmin + rand（0，1）（x jmax - x jmin）
Wherein,X i,j to generate the firstiAt the first initial positionjIndividual weights or thresholds;rand(0, 1) is a random number between 0 and 1.
Preferably, the step S3 includes the steps of:
step S3.1, determining an update condition for generating a new pre-location point, specifically,
wherein,
is referred to askIs newly generatediA first of the position pointsjIndividual weights or thresholds;Centre j k is referred to askIn the place of the centerjIndividual weights or thresholds;Grefers to the maximum number of iterations;
step S3.2, generating a new preset point by adopting the updating conditionY i k And calculating each pre-position point by the fitness function of the calculation formula (1)Y i k A corresponding fitness value;
step S3.3, updating the position information, in particularkFitness value of substitute location pointfitness（Y i k ) And a firstkFitness value of-1 generation location pointsfitness（X i k-1) Make a comparison iffitness（Y i k ）<fitness（X i k-1) Then the location point information needs to be updated, orderfitness（X i k ）=fitness（Y i k ），X i k =Y i k (ii) a Otherwise, it ordersfitness（X i k ）=fitness（X i k-1），X i k =X i k-1(ii) a Wherein,X i k is as followskGeneration by generationiA plurality of location points;X i k-1is as followskGeneration 1iA plurality of location points;Y i k is as followskGeneration of the firstiA pre-position point.
Preferably, the step S4 includes the steps of:
step S4.1, determining the current generation optimal positionRbestx k And global optimal positionGbestx k And, in particular, the updatedkIn generationfitness（X i k ) The position point of the minimum value in (1) is taken as the current generation optimum positionRbestx k (ii) a When in usekWhen =1, the current generation optimum positionRbestx 1Global optimum position for first generationGbestx 1；
Step S4.2 whenkWhen =2, the global optimum position is updated, in particular, the current generation optimum position is comparedRbestx k Fitness value offitness（Rbestx k ) And global optimum positionGbestx k-1Fitness value offitness（Gbestx k-1) (ii) a If it isfitness（Rbestx k ）<fitness（Gbestx k-1) Then the global optimum location needs to be updated, orderfitness（Gbestx k ）=fitness（Rbestx k ），Gbestx k =Rbestx k (ii) a Otherwise, it ordersfitness（Gbestx k ）=fitness（Gbestx k-1），Gbestx k =Gbestx k-1。
Preferably, in step S3, the center position follows the current generation optimum positionRbestx k And global optimal positionGbestx k The movement of (a) is moved, in particular,
Centre k +1=Centre k +0.4（Gbestx k -Centre k ）+0.5（Rbestx k -Centre k ）,
wherein,Centre k is as followskThe center position of the generation;Centre k+1is as followsk-central position of generation 1.
The technical scheme of the invention at least has the following beneficial effects:
the invention adopts a BP neural network model and an improved radial movement algorithm to be combined for use, and particularly, a performance function of the BP neural network model is taken as a fitness function of the improved radial movement algorithm
And assigning the weight and the threshold of the global optimal position obtained by the radial movement algorithm to the BP neural network model, training and simulating the BP neural network model to obtain the optimal predicted value of the vertical ultimate bearing capacity of the pile foundation, so that the defects that the BP neural network model is easy to fall into a local minimum value, the convergence speed is low and the search result is unstable can be improved, and the predicted value is convenient to be ensured to be closer to the true value.
In addition to the objects, features and advantages described above, other objects, features and advantages of the present invention are also provided. The present invention will be described in further detail below with reference to the drawings.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the invention and, together with the description, serve to explain the invention and not to limit the invention. In the drawings:
fig. 1 is a schematic flowchart of a method for predicting vertical ultimate bearing capacity of a pile foundation in embodiment 1 of the present invention;
FIG. 2 is a comparison of real and predicted values for a test sample.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments that can be derived by one of ordinary skill in the art from the embodiments given herein are intended to be within the scope of the present invention.
Example 1:
referring to fig. 1, a method for predicting vertical ultimate bearing capacity of a pile foundation includes the following steps:
s1, taking the weight and the threshold of the BP neural network model as position point information of an improved radial movement algorithm, and establishing an initial population;
step S2, calculating the fitness value of each individual in the initial population by adopting the fitness function of the improved radial movement algorithm, determining the current optimal position by comparing the fitness values of the individuals one by one, and defining the current optimal position as the initial central position; the fitness function is a performance function of the BP neural network model, and specifically comprises the following steps:
wherein,Nthe number of test samples for the BP neural network model;t i for BP neural network modeliThe predicted values of the output variables of the individual test samples, in particular,t i to model BP neural networkiAssigning the weight and the threshold value of the individual in each test sample to a predicted value obtained after the BP neural network model is operated;y i for BP neural network modeliThe actual value of each test sample output variable;
step S3, adopting the update condition at the firstk. + -. 0.5 of the center position (x jmax - x jmin）w k Generating a new pre-position point in the range, calculating a fitness value, and updating position information; wherein,krefers to the current iteration number;x jminis as followsjThe minimum value of the individual weights or thresholds;x jmaxis as followsjThe maximum value of the individual weights or thresholds,w k is the inertia weight;
step S4, determining the global optimal position according to the position information updated in the step S3, and calculating the fitness value of the global optimal position; if the iteration times reach the upper limit or the fitness value of the global optimal position is less than 0.001, ending the updating; otherwise, repeating the steps S3-S4 until the updating is finished;
and step S5, assigning the weight and the threshold value of the global optimal position to the BP neural network model, training and simulating the BP neural network model to obtain an optimal predicted value of the vertical ultimate bearing capacity of the pile foundation, and ending the algorithm.
The step S1 includes the steps of:
s1.1, calculating the total number of the weight values and the threshold values of the BP neural network model, and specifically obtaining the weight values and the total number of the threshold values through the following calculation formula:
nod=w 1+w 2+w 3formula (2)
w 1=n 1 × n 2 Formula (3)
w 2=n 2 × n 3 Formula (4)
w 3=n 2 + n 3 Formula (5)
Wherein,nodthe total number of the weight and the threshold value;w 1the number of the weights from the input layer to the hidden layer;w 2the number of weights from the hidden layer to the output layer;w 3is the total number of the threshold values;n 1the number of input layers;n 2the number of hidden layers;n 3the number of output layers;
step S1.2, determining value ranges of weight and thresholdx jmaxAndx jminwherein, 1 is less than or equal toj ≤ nod；
Step S1.3, generating randomly in the value range of the step S1.2Become intonopAn initial position point ofnopEstablishing an initial population by using initial position points, wherein the numerical information of the initial position points is obtained by a calculation formula (6);
the calculation formula (6) isX i,j =x jmin + rand（0，1）（x jmax - x jmin）
Wherein,X i,j to generate the firstiAt the first initial positionjIndividual weights or thresholds;rand(0, 1) is a random number between 0 and 1.
The step S3 includes the steps of:
step S3.1, determining an update condition for generating a new pre-location point, specifically,
wherein,
is referred to askIs newly generatediA first of the position pointsjIndividual weights or thresholds;Centre j k is referred to askIn the place of the centerjIndividual weights or thresholds;w k the inertia weight is decreased with the iteration times and is used for determining the convergence speed of the algorithm;
step S3.2, generating a new preset point by adopting the updating conditionY i k And calculating each pre-position point by the fitness function of the calculation formula (1)Y i k A corresponding fitness value;
step S3.3, updating the position information, specifically, updating the fitness value of the kth generation of pre-position pointfitness（Y i k ) And a firstkFitness value of-1 generation location pointsfitness（X i k-1) Make a comparison iffitness（Y i k ）<fitness（X i k-1) Then the location point information needs to be updated, orderfitness（X i k ）=fitness（Y i k ），X i k =Y i k(ii) a Otherwise, it ordersfitness（X i k ）=fitness（X i k-1），X i k =X i k-1(ii) a Wherein,X i k is as followskGeneration by generationiA plurality of location points;X i k-1is as followskGeneration 1iA plurality of location points;Y i k is as followskGeneration of the firstiA pre-position point.
The step S4 includes the steps of:
step S4.1, determining the current generation optimal positionRbestx k And global optimal positionGbestx k And, in particular, the updatedkIn generationfitness（X i k ) The position point of the minimum value in (1) is taken as the current generation optimum positionRbestx k (ii) a When in usekWhen =1, the current generation optimum positionRbestx 1Global optimum position for first generationGbestx 1；
Step S4.2 whenkWhen =2, the global optimum position is updated, in particular, the current generation optimum position is comparedRbestx k Fitness value offitness（Rbestx k ) And global optimum positionGbestx k-1Fitness value offitness（Gbestx k-1) (ii) a If it isfitness（Rbestx k ）<fitness（Gbestx k-1) Then the global optimum location needs to be updated, orderfitness（Gbestx k ）=fitness（Rbestx k ），Gbestx k =Rbestx k (ii) a Otherwise, it ordersfitness（Gbestx k ）=fitness（Gbestx k-1），Gbestx k =Gbestx k-1
In step S3, the center position follows the current generation optimum positionRbestx k And global optimal positionGbestx k The movement of (a) is moved, in particular,
Centre k +1=Centre k +0.4（Gbestx k -Centre k ）+0.5（Rbestx k -Centre k ）,
wherein,Centre k is as followskThe center position of the generation;Centre k+1is as followsk-central position of generation 1.
The parameters adopted by the BP neural network model in the step S5 include prediction accuracy, maximum iteration number and learning rate, and the prediction accuracy is 10-6The maximum iteration number is 1000, and the learning rate is 0.01.
The topological structure of the BP neural network model adopted in the embodiment 1 is 4-10-1, and the number of hidden layer nodes is 10 according to the optimal value determined by a trial algorithm. Reference numerals in example 1nodIs 61,nopIs 50 percent,GIs 100.
Comparative example 1:
and predicting the vertical ultimate bearing capacity of the pile foundation only by adopting a BP neural network model.
Comparative example 2:
and (3) predicting the vertical ultimate bearing capacity of the pile foundation by only adopting a genetic algorithm to optimize the BP neural network model.
32 groups of static load test data are collected for the vertical ultimate bearing capacity of the pile foundation, wherein 27 groups of data serve as a training set, and 5 groups of data serve as a test set. The static load test data were run 10 times using the prediction methods in example 1 and comparative examples 1-2, and the analysis results are detailed in table 1.
TABLE 1
| Grouping | Relative error | Mean Square Error (MSE) |
| Comparative example 1 | 25.86%-0.42% | 0.0049 |
| Comparative example 2 | 8.55%~0.48% | 0.0036 |
| Example 1 | 0.01%~2.42% | 1.12×10-7 |
As can be seen from Table 1, the relative error between the predicted value and the actual value of the BP neural network model test sample adopted in comparative example 1 is 25.86% -0.42%, the MSE is 0.0049, and the prediction result has great instability and relatively deviates from the actual value. The relative error between the predicted value and the actual value of the BP neural network model test sample optimized by the genetic algorithm adopted in the comparative example 2 is 8.55-0.48%, the MSE is 0.0036, and the prediction result has better stability and is closer to the actual value. Pile foundation vertical limit bearing adopted by embodiment 1The relative error between the predicted value and the actual value of the test sample obtained by the load force prediction method is 0.01-2.42%, and the MSE is 1.12 multiplied by 10-7The maximum relative error is below 3%, and the prediction result is very close to the true value. It can be seen that the method for predicting the vertical ultimate bearing capacity of the pile foundation adopted in the embodiment 1 has good predictability. Meanwhile, as can be seen from fig. 2, the difference between the 10 predicted values and the true values of the test samples (i.e., 5 groups of data in the test set) is not large, and the 10 prediction results are almost the same, so that the method for predicting the vertical ultimate bearing capacity of the pile foundation adopted in embodiment 1 has good stability.
In summary, the invention combines the BP neural network model with the improved radial movement algorithm, and specifically, the performance function of the BP neural network model is used as the fitness function of the improved radial movement algorithm
And assigning the weight and the threshold of the global optimal position obtained by the radial movement algorithm to the BP neural network model, training and simulating the BP neural network model to obtain the optimal predicted value of the vertical ultimate bearing capacity of the pile foundation, so that the defects that the BP neural network model is easy to fall into a local minimum value, the convergence speed is low and the search result is unstable can be improved, and the predicted value is convenient to be ensured to be closer to the true value.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (3)
1. The method for predicting the vertical ultimate bearing capacity of the pile foundation is characterized by comprising the following steps of:
step S1: acquiring test data of the vertical ultimate bearing capacity of the pile foundation, and dividing a data set into a training set and a test set;
s2, taking the weight and the threshold of the BP neural network model as position point information of a predicted value of the vertical ultimate bearing capacity of the pile foundation in the improved radial movement algorithm, and establishing an initial population;
step S3, calculating the fitness value of each individual in the initial population by adopting a fitness function of an improved radial movement algorithm, determining the optimal position of the predicted value of the vertical ultimate bearing capacity of the current pile foundation by comparing the fitness values of the individuals one by one, and defining the optimal position as the initial central position; the fitness function is a performance function of the BP neural network model, and specifically comprises the following steps:
wherein N is the number of test sets in the BP neural network model data set; t is tiOutputting a predicted value of a variable for the ith test set in the BP neural network model data set; y isiOutputting the true value of the variable for the ith test set in the BP neural network model data set;
step S4, adopting the updating condition to be +/-0.5 (x) of the k generation center positionjmax-xjmin)wkGenerating a new pre-position point in the range, calculating a fitness value, and updating position information; wherein k refers to the current iteration number; x is the number ofjminIs the minimum value of the jth weight or threshold; x is the number ofjmaxIs the maximum value of the jth weight or threshold, wkIs the inertia weight;
step S5, determining the global optimal position according to the position information updated in the step S4, and calculating the fitness value of the global optimal position; if the iteration times reach the upper limit or the fitness value of the global optimal position is less than 0.001, ending the updating; otherwise, repeating the steps S4-S5 until the updating is finished;
step S6, assigning the weight and the threshold value of the global optimal position to a BP neural network model, training and simulating the BP neural network model by adopting a training set to obtain an optimal predicted value of the vertical ultimate bearing capacity of the pile foundation, and ending the algorithm;
wherein the step S4 includes the steps of:
step S4.1, determining the update conditions for generating new pre-location points, specifically,
wherein,
is the jth weight or threshold of the ith position point newly generated by the kth generation; centrej kThe weight or threshold of the jth generation center position is referred to as the jth weight or threshold; g refers to the maximum number of iterations;
step S4.2, generating a new preset point Y by adopting the updating conditioni kAnd calculating each pre-position point Y by the fitness function of the calculation formula (1)i kA corresponding fitness value;
step S4.3, updating the position information, specifically, updating the fitness value fitness (Y) of the k-th generation pre-position pointi k) Fitness value fitness (X) with the k-1 th generation location pointi k-1) Making a comparison if fitness (Y)i k)<fitness(Xi k-1) Then the location point information needs to be updated, let fitness (X)i k)＝fitness(Yi k)，Xi k＝Yi k(ii) a Otherwise, let fitness (X)i k)＝fitness(Xi k-1)，Xi k＝Xi k-1(ii) a Wherein, Xi kIs the ith position point of the kth generation; xi k-1Is the ith position point of the k-1 generation; y isi kGenerating an ith pre-position point for the kth generation;
the step S5 includes the steps of:
step S5.1, determining the current generation optimal position RbestxkAnd global optimum position Gbest xkSpecifically, the updated fixness (X) in the k-th generationi k) The position point of the minimum value in (b) is taken as the current generation optimum position Rbestxk(ii) a When k is 1, the current generation optimum position Rbestx1Global optimum position Gbestx for first generation1；
Step S5.2, when k is 2, updating the global optimal position, specifically, comparing Rbestx of the current optimal positionkFitness value of (Rbestx)k) And global optimum position Gbest xk-1Fitness value of (Gbestx)k-1) (ii) a If fitness (Rbestx)k)<fitness(Gbestxk-1) Then the global optimum location needs to be updated, let fitness (Gbesx)k)＝fitness(Rbestxk)，Gbestxk＝Rbestxk(ii) a Otherwise, let fitness (Gbesx)k)＝fitness(Gbestxk-1)，Gbestxk＝Gbestxk-1。
2. The method for predicting pile foundation vertical ultimate bearing capacity according to claim 1, wherein the step S2 includes the steps of:
s2.1, calculating the total number of the weight values and the threshold values of the BP neural network model, and specifically obtaining the weight values and the total number of the threshold values through the following calculation formula:
nod＝w1+w2+w3formula (2)
w1＝n1×n2Formula (3)
w2＝n2×n3Formula (4)
w3＝n2+n3Formula (5)
Wherein, nod is the total number of the weight and the threshold; w is a1The number of the weights from the input layer to the hidden layer; w is a2The number of weights from the hidden layer to the output layer; w is a3Is the total number of the threshold values; n is1The number of input layers; n is2The number of hidden layers; n is3The number of output layers;
s2.2, determining the value range x of the weight and the threshold valuejmaxAnd xjminWherein j is more than or equal to 1 and less than or equal to nod;
s2.3, randomly generating nop initial position points in the value range of the step S1.2, establishing an initial population by the nop initial position points, and obtaining the numerical information of the initial position points through a calculation formula (6);
the calculation formula (6) is Xi,j＝xjmin+rand(0，1)(xjmax-xjmin)
Wherein, Xi,jGenerating a jth weight or threshold of the ith initial position point; rand (0, 1) is a random number between 0 and 1.
3. The method for predicting pile foundation vertical ultimate bearing capacity according to claim 2, wherein in step S4, Rbestx with its center position following the current generation optimal positionkAnd global optimum position Gbest xkThe movement of (a) is moved, in particular,
Centrek+1＝Centrek+0.4(Gbestxk-Centrek)+0.5(Rbestxk-Centrek),
wherein, CentrekIs the central position of the k generation; centrek+1Is the central position of the k-1 generation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111381622.XA CN113821863B (en) | 2021-11-22 | 2021-11-22 | Method for predicting vertical ultimate bearing capacity of pile foundation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111381622.XA CN113821863B (en) | 2021-11-22 | 2021-11-22 | Method for predicting vertical ultimate bearing capacity of pile foundation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113821863A CN113821863A (en) | 2021-12-21 |
| CN113821863B true CN113821863B (en) | 2022-03-01 |
Family
ID=78919393
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111381622.XA Active CN113821863B (en) | 2021-11-22 | 2021-11-22 | Method for predicting vertical ultimate bearing capacity of pile foundation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113821863B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116822328A (en) * | 2023-05-11 | 2023-09-29 | 中南大学 | Determination method for mine goaf earth surface subsidence prediction parameters |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108304674A (en) * | 2018-02-09 | 2018-07-20 | 陕西铁路工程职业技术学院 | A kind of railway prediction of soft roadbed settlement method based on BP neural network |
| CN111209708A (en) * | 2020-04-22 | 2020-05-29 | 西南交通大学 | Machine learning-based pile-soil interaction prediction analysis method |
| CN112926117A (en) * | 2021-03-01 | 2021-06-08 | 郑州大学 | BAS-BP-based pile foundation horizontal bearing capacity prediction method |
| CN113051775A (en) * | 2021-04-13 | 2021-06-29 | 中南大学 | Horizontal directional drilling track optimization method based on improved radial movement algorithm |
-
2021
- 2021-11-22 CN CN202111381622.XA patent/CN113821863B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108304674A (en) * | 2018-02-09 | 2018-07-20 | 陕西铁路工程职业技术学院 | A kind of railway prediction of soft roadbed settlement method based on BP neural network |
| CN111209708A (en) * | 2020-04-22 | 2020-05-29 | 西南交通大学 | Machine learning-based pile-soil interaction prediction analysis method |
| CN112926117A (en) * | 2021-03-01 | 2021-06-08 | 郑州大学 | BAS-BP-based pile foundation horizontal bearing capacity prediction method |
| CN113051775A (en) * | 2021-04-13 | 2021-06-29 | 中南大学 | Horizontal directional drilling track optimization method based on improved radial movement algorithm |
Non-Patent Citations (4)
| Title |
|---|
| 《Application of improved radial movement optimization for calculating the upper bound of ultimate bearing capacity of shallow foundation on unsaturated soil》;Liangxing Jin等;《Computers and Geotechnics》;20190129;第82-88页 * |
| 《The use of improved radial movement optimization to calculate the ultimate bearing capacity of a nonhomogeneous clay foundation adjacent to slopes》;Liangxing Jin等;《Computers and Geotechnics》;20191114;第1-8页 * |
| 《基于RMO-BP算法的感应电动机转子断条故障诊断》;崔京华等;《电气传动》;20191031;第96-101页 * |
| 《条形基础地基极限承载力的改进径向移动算法研究》;金亮星等;《铁道科学与工程学报》;20170228;第233-240页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113821863A (en) | 2021-12-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107316099A (en) | Ammunition Storage Reliability Forecasting Methodology based on particle group optimizing BP neural network | |
| CN104408518A (en) | Method of learning and optimizing neural network based on particle swarm optimization algorithm | |
| CN113821863B (en) | Method for predicting vertical ultimate bearing capacity of pile foundation | |
| CN112329349B (en) | Slope reliability assessment method | |
| JP3929235B2 (en) | State estimation method in distribution system | |
| CN110163743A (en) | A kind of credit-graded approach based on hyperparameter optimization | |
| CN114861519B (en) | Initial ground stress field acceleration optimization inversion method under complex geological conditions | |
| CN113919221A (en) | Fan load prediction and analysis method and device based on BP neural network and storage medium | |
| CN111008686B (en) | Grid structure optimization method for realizing reliable splitting based on node correlation | |
| CN114066250A (en) | Method, device, equipment and storage medium for measuring and calculating repair cost of power transmission project | |
| CN111639463B (en) | XGboost algorithm-based frequency characteristic prediction method for power system after disturbance | |
| CN107766887A (en) | A kind of local weighted deficiency of data mixes clustering method | |
| CN117252085A (en) | Cable-stayed bridge cable force tolerance optimization method and system based on neural network and particle swarm | |
| CN116956744A (en) | Multi-loop groove cable steady-state temperature rise prediction method based on improved particle swarm optimization | |
| Wu et al. | Multi-objective optimization design of constant stress accelerated degradation test using inverse Gaussian process | |
| CN107396375A (en) | Heterogeneous sensor network optimum target covering method based on quantum brainstorming | |
| CN114329702B (en) | High-stability method and device for calibrating and designing reaction spectrum based on improved differential evolution algorithm | |
| CN116680969A (en) | Filler evaluation parameter prediction method and device for PSO-BP algorithm | |
| CN115100233B (en) | Radar target tracking method based on generation of anti-network resampling particle filtering | |
| CN112581311B (en) | Method and system for predicting long-term output fluctuation characteristics of aggregated multiple wind power plants | |
| CN112330435A (en) | Credit risk prediction method and system for optimizing Elman neural network based on genetic algorithm | |
| Mao et al. | QoS trust rate prediction for Web services using PSO-based neural network | |
| CN110084477A (en) | A kind of cloud data center operating status appraisal procedure | |
| Rashidi et al. | Robust and adaptive tuning of power system stabilizers using artificial neural networks | |
| CN117895920B (en) | Distributed consistency Kalman filtering method for sensor network under communication link fault |
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 |