CN116189794A - Rammed earth water salt content measurement method - Google Patents

Rammed earth water salt content measurement method Download PDF

Info

Publication number
CN116189794A
CN116189794A CN202310139327.6A CN202310139327A CN116189794A CN 116189794 A CN116189794 A CN 116189794A CN 202310139327 A CN202310139327 A CN 202310139327A CN 116189794 A CN116189794 A CN 116189794A
Authority
CN
China
Prior art keywords
salt content
soil
rammed earth
water
particle
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.)
Pending
Application number
CN202310139327.6A
Other languages
Chinese (zh)
Inventor
王志明
徐田铖
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing University of Science and Technology
Original Assignee
Nanjing University of Science and Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Nanjing University of Science and Technology filed Critical Nanjing University of Science and Technology
Priority to CN202310139327.6A priority Critical patent/CN116189794A/en
Publication of CN116189794A publication Critical patent/CN116189794A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
    • G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
    • G16C20/20—Identification of molecular entities, parts thereof or of chemical compositions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • 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
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
    • G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
    • G16C20/70—Machine learning, data mining or chemometrics

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computing Systems (AREA)
  • Artificial Intelligence (AREA)
  • Software Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Data Mining & Analysis (AREA)
  • Health & Medical Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Computational Linguistics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Mathematical Physics (AREA)
  • Crystallography & Structural Chemistry (AREA)
  • Chemical & Material Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Molecular Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Medical Informatics (AREA)
  • Databases & Information Systems (AREA)
  • Investigating Or Analyzing Materials By The Use Of Electric Means (AREA)

Abstract

The invention discloses a method for measuring the water salt content of rammed earth, which comprises the following steps: initializing a population, compacting the population scale in a water and salt content prediction model, establishing a generalized regression network to predict the water and salt content of the compacted soil, calculating the root mean square error between the actual value and the predicted value of the water content or the salt content as the fitness of particles, updating the state of the particles, executing crossover and mutation operations, judging whether convergence conditions are met, and predicting the water and salt content of the compacted soil. The invention utilizes the generalized regression neural network optimized by the improved GA-PSO algorithm, can indirectly obtain the water content and the salt content of the rammed soil by measuring the capacitance value, the conductivity, the soil temperature and the soil density parameters of the rammed soil, avoids the defects of field sampling and time and labor waste in laboratory detection in laboratory measurement, greatly improves the modeling speed compared with the traditional GA algorithm, avoids the problem of sinking into a local optimal solution compared with the traditional PSO algorithm, and is a method with better comprehensive performance.

Description

Rammed earth water salt content measurement method
Technical Field
The invention belongs to the technical field of soil measurement, and particularly relates to a method for rapidly measuring the moisture content and the salt content of rammed earth.
Background
The earthen site is a kind of soil building site and has high history and cultural value. Because the earthen site is mostly in the natural open air environment, the earthen site is easily damaged by various human and natural factors. For the common cracks, undercuts, collapse and other diseases of the earthen site, physical reinforcement is generally adopted to carry out protection measures such as reinforcement and repair on the earthen site. The method is characterized in that the method is a common and effective method for reinforcing the branch roof by ramming, and the original structure can be effectively stabilized. In order to monitor the reinforcing effect of the rammed roof, the water content and the salt content of the rammed roof need to be measured.
At present, the method for measuring the water content and the salt content of rammed earth is mainly traditional laboratory measurement (a drying method and a soaking solution measurement method), wherein the drying method is a measurement method for determining the water content according to the change of the mass of a soil sample before and after drying, and the soaking solution measurement method is based on the principle that the soil sample is prepared into leaching solution, and the salt content of the soil is calculated by calculating the conductivity. Neither of these methods enables rapid real-time on-site measurement of parameters. Since rammed earth is a soil, many characteristics are similar to those of agro-farming soil, but there are many differences, for example, the rammed earth needs to be added with additives such as quicklime to ensure strength and rigidity, the density of the rammed earth generally meets certain standards, and the measurement of the rammed earth building requires certain continuity, real-time performance and the like. When the dielectric property measurement method is used for measuring the water content and the salt content of the rammed earth, parameters such as capacitance value, conductivity, soil temperature, soil density and the like need to be fully considered to accurately predict the water content and the salt content, so that a multiple regression algorithm of the parameters on the water content and the salt content needs to be established.
Disclosure of Invention
The invention aims to provide a multiple regression algorithm for measuring the water content and the salt content of rammed earth so as to realize the rapid and accurate measurement and prediction of the water content and the salt content of the rammed earth.
The technical solution for realizing the purpose of the invention is as follows:
the method for measuring the water salt content of the rammed earth comprises the following steps of:
step 1, initializing a population, and setting the population scale in a GA-PSO optimized GRNN model;
step 2, establishing a generalized regression network model, taking the capacitance value, the conductivity, the soil temperature and the soil density of the rammed earth as input parameters, and taking the predicted water content and the salt content of the rammed earth as output parameters;
step 3, calculating the fitness of each particle:
taking the root mean square error between the actual value and the predicted value of the moisture content or the salt content of the rammed soil as the fitness of the particles;
step 4, updating the state of the particles:
updating the speed and displacement of each particle in the particle swarm algorithm;
step 5, executing crossover and mutation operation:
introducing cross operation to the updating of the particle position and speed, and introducing mutation operation to the updating of the particle position and speed;
step 6, judging whether convergence conditions are met, if the optimized generalized regression network model meets the precision requirement, establishing a generalized regression network, and predicting the moisture content and the salt content of the rammed soil; if the precision does not meet the requirement, returning to the step 2;
and 7, predicting the water content and the salt content of the rammed earth by using the optimized generalized regression network.
Compared with the prior art, the invention has the remarkable advantages that:
(1) Inversion prediction of the rammed earth water content and the salt content can be realized through a generalized regression neural network model. The water content and the salt content of the rammed earth can be indirectly obtained by measuring the capacitance, the conductivity, the soil temperature and the soil density parameters of the rammed earth, the defects that a laboratory measurement method needs on-site sampling and detection in a laboratory is time-consuming and labor-consuming are avoided, and the water content and the salt content can be obtained on site in real time.
(2) The generalized regression neural network optimized by the improved GA-PSO algorithm improves the accuracy of the regression model, greatly improves the modeling speed compared with the independent GA algorithm, avoids the problem of sinking into a local optimal solution compared with the independent PSO algorithm, and is a method with better comprehensive performance.
Drawings
FIG. 1 is a flow chart of the GRNN algorithm for GA-PSO optimization.
Fig. 2 is a GRNN network configuration diagram.
FIG. 3 is a schematic diagram of particle update in PSO algorithm.
Fig. 4 is a graph of water cut measurement data versus predicted data.
FIG. 5 is a graph of salt content measurement data versus predicted data.
Detailed Description
The invention is further described with reference to the drawings and specific embodiments.
The method for measuring the water salt content of rammed earth in the embodiment comprises the following steps of:
step 1, initializing a population, and setting the population scale in an algorithm model.
Particle Swarm Optimization (PSO) is an optimizing algorithm with inspiration from the predation behavior of the bird swarm, has high convergence speed and has the basic idea of searching for an optimal solution through cooperation and information sharing between the swarm and the individual. The particle swarm algorithm abstracts each feasible solution into an individual particle point and extends the particle point to an N-dimensional space, so that a position vector D of each individual in the N-dimensional space is obtained i ＝(d 1 ，d 2 ，...，d N ) And a flying speed vector V i ＝(v 1 ，v 2 ，...，v N ). Each particle has a fitness, the value of which is determined by the objective function. The particles find positions in space according to conditions, and adjust own position vectors and flying speed vectors. In the process, each particle can generate an individual history optimal solution p_best, a global optimal solution g_best in the population is obtained through communication, and the position vector and the speed square vector of the particle are continuously adjusted according to the p_best and the g_best, so that the optimal solution is finally achieved.
Step 2, establishing a generalized regression network model to predict the water and salt content of the rammed soil
The generalized regression neural network (general regression neural network, GRNN) is a radial basis neural network based on mathematical statistics, is generally used for carrying out regression approximation on nonlinear functions, has a high training speed, is far smaller than a BP neural network, and has higher accuracy under the condition of a small sample. As shown in fig. 2, the GRNN network is composed of a four-layer structure including an input layer, a mode layer, a summation layer, and an output layer. The number of neurons of the input layer is the same as the vector dimension in the training sample, and the input vector X is directly output to the mode layer. In the prediction of the water and salt content of rammed soil, parameters of an input layer are set as a capacitance value, conductivity, soil temperature and soil density of the rammed soil, and an output layer is the water content and the salt content of the rammed soil.
The theoretical basis of GRNN is a nonlinear regression analysis, assuming x, y is two random variables, x 0 For x, the joint probability density of x, y is f (x, y), then there are:
Figure BDA0004087023530000031
wherein
Figure BDA0004087023530000032
For input of x 0 Prediction output of y under the condition, f (x 0 Y) is x 0 Joint probability density with y.
By using
Figure BDA0004087023530000033
Representing a sample dataset, where n is the total number of samples, x i ，y i Is the observed value for the i-th sample. Applying Parzen non-parametric estimation to the sample dataset may yield an estimated probability density function:
Figure BDA0004087023530000034
where p is the dimension of the random variable x and σ is the width coefficient (standard deviation) of the gaussian function, also known as the smoothing factor.
Will be
Figure BDA0004087023530000035
Carrying into formula (1), obtaining:
Figure BDA0004087023530000041
the number of neurons of the mode layer is the same as the number of training samples, namely, each neuron is provided with a training sample corresponding to the training samples, the Euclid distance between each neuron and all the training samples is calculated for the test samples in the mode layer, and the transfer function of the mode layer is as follows:
Figure BDA0004087023530000042
wherein X is a network input vector, X i Training samples corresponding to the ith neuron; n is the number of training samples and also the number of pattern layer neurons.
Two types of neurons are used in the summation layer to sum, the first calculation directly sums the mode layer outputs, the formula is:
Figure BDA0004087023530000043
the second calculation is to perform weighted summation on the mode layer output, multiply the connection weight of the r-th neuron in the mode layer and the t-th molecule in the summation layer by the neuron of the r-th mode layer and sum, and the formula is as follows:
Figure BDA0004087023530000044
the two types of neuron transfer functions are respectively:
Figure BDA0004087023530000045
Figure BDA0004087023530000046
in the formula SD Is a transfer function when the weight is 1, S Nt Is the transfer function of the t th neuron, w rt And s is the number of the mode layer weighted sum neurons for the connection weight between the r-th neuron in the mode layer and the t-th neuron in the combining layer.
The output layer neuron outputs are:
Figure BDA0004087023530000047
wherein q is the dimension of the output vector, and q is 2 in the prediction algorithm of the rammed earth water salt content.
In the regression model, the parameters of the input layer are the capacitance value, the conductivity, the temperature and the rammed earth density measured by the measuring module, and the output layer is the water content and the salt content. Since the weighting factor of each observed value Yi is the corresponding sample X i An index of the square of the Euclidean distance from X. The accuracy of the predicted value is therefore largely dependent on the value of the smoothing factor sigma, and when sigma is large, - (X-X) r ) T (X-X r )/2σ 2 Approaching 0, the predicted value is therefore approximately equal to the average value of the training sample dependent variable; while σ approaches 0, exp [ - (X-X) r ) T (X-X r )/2σ 2 ]Approaching to 0, the distance between Y and the training sample is very close, the predicted value has better precision only when the predicted point is in the training sample, and the predicted effect is poor when the predicted point is not in the training sample, namely the network generalization capability is poor. Therefore, selecting a proper smoothing factor is important for the generalized regression neural network, and the precision and generalization capability of the model are concerned. In order to select the most suitable smoothing factor, the traditional empirical method has low accuracy and long time consumption, so that the smoothing factor is optimized by adopting an improved GA-PSO algorithm.
Step 3, calculating the fitness of each particle
When predicting the water content or salt content of the rammed earth, the root mean square error (Root Mean Square Error, RMSE) between the actual value and the predicted value of the water content or salt content is used as the fitness of the particles, and the fitness of each particle is calculated. The root mean square error is calculated as follows:
Figure BDA0004087023530000051
wherein
Figure BDA0004087023530000052
For the predicted value of the ith sample, y i Is the actual value of the i-th sample.
Step 4, updating the state of the particles
The speed and displacement of each particle in the particle swarm algorithm are respectively carried out according to the following two formulas:
Figure BDA0004087023530000053
Figure BDA0004087023530000054
wherein i=1, 2, … M, M is the total number of particles in the population;
Figure BDA0004087023530000055
the velocity vector of the ith particle iterates in the kth and the (k+1) th rounds;
Figure BDA0004087023530000056
Is the position vector of the ith particle in the k-th round and the k+1-th round iteration respectively;
Figure BDA0004087023530000057
is the individual optimal solution for round k,>
Figure BDA0004087023530000058
is the optimal solution of the k-th round of group; omega is an inertial factor; r is (r) 1 、r 2 Is a random number between (0, 1); c 1 、c 2 Is a learning factor.
Step 5, executing crossover and mutation operations
The crossover and mutation operations in the modified GA-PSO algorithm are similar to those in the genetic algorithm.
Step 5.1, performing a interleaving operation
And introducing cross operation to the updating of the position and the speed of the particles, and obtaining an updating formula of the position and the speed of the particles, wherein the updating formula is as follows:
Figure BDA0004087023530000061
Figure BDA0004087023530000062
wherein ,
Figure BDA0004087023530000063
is the position vector of the jth particle in the k-th and k+1-th iterations, respectively;
Figure BDA0004087023530000064
The j-th particle is the velocity vector of the iteration of the kth and the k+1-th rounds; alpha 1 、α 2 Is a random number of the interval (0, 1).
Step 5.2, executing mutation operation:
the mutation operation is to mutate the individual according to a certain probability, so as to achieve the purpose of optimizing the individual and enriching the group, and the mutation operation is also an important reason that the algorithm is not easy to sink into a local optimal solution. The mutation operation is performed as follows:
Figure BDA0004087023530000065
in the formula Pm 、P′ m Individuals before and after mutation, P max 、P min The upper limit and the lower limit of the solution space are respectively, beta is variation probability, k is the current iteration number, and mg is the total iteration number.
Step 6, judging whether the convergence condition is satisfied
Stopping if the optimized GRNN model meets the precision requirement; otherwise, returning to the step 2, the training needs to be continued by adjusting parameters.
Step 7, determining a final GRNN model and predicting
The optimized GRNN model can be used for predicting the water content and the salt content of the rammed earth.
Examples
Data acquisition is carried out on self-made rammed earth samples, the capacitance value, the conductivity, the soil temperature and the soil density of the rammed earth are measured, the water content and the salt content of the rammed earth are measured by using a laboratory method, the data acquired by the experiment are predicted by using the method, and the predicted result is compared with the measured result obtained by using the laboratory method. The results are shown in fig. 4 and 5, wherein fig. 4 is the comparison of the water content measurement data and the predicted data, and fig. 5 is the comparison of the salt content measurement data and the predicted data. Experiments show that the predicted value and the measured value obtained by the method have good correlation and high accuracy.

Claims (5)

1. The method for measuring the water salt content of the rammed earth is characterized by comprising the following steps of:
step 1, initializing a population, and setting the population scale in a GA-PSO optimized GRNN model;
step 2, establishing a generalized regression network model, taking the capacitance value, the conductivity, the soil temperature and the soil density of the rammed earth as input parameters, and taking the predicted water content and the salt content of the rammed earth as output parameters;
step 3, calculating the fitness of each particle:
taking the root mean square error between the actual value and the predicted value of the moisture content or the salt content of the rammed soil as the fitness of the particles;
step 4, updating the state of the particles:
updating the speed and displacement of each particle in the particle swarm algorithm;
step 5, executing crossover and mutation operation:
introducing cross operation to the updating of the particle position and speed, and introducing mutation operation to the updating of the particle position and speed;
step 6, judging whether convergence conditions are met, if the optimized generalized regression network model meets the precision requirement, establishing a generalized regression network, and predicting the moisture content and the salt content of the rammed soil; if the precision does not meet the requirement, returning to the step 2;
and 7, predicting the water content and the salt content of the rammed earth by using the optimized generalized regression network.
2. The method for measuring the water salt content of rammed earth of claim 1, wherein the operations of crossing and mutation are performed: the method specifically comprises the following steps:
performing a crossover operation updates the formula:
Figure FDA0004087023500000011
Figure FDA0004087023500000012
wherein ,
Figure FDA0004087023500000013
is the position vector of the jth particle at the kth iteration;
Figure FDA0004087023500000014
Is the velocity vector of the jth particle at the kth iteration; alpha 1 、α 2 Random numbers for interval (0, 1);
the update formula of the mutation operation is executed as follows:
Figure FDA0004087023500000015
in the formula Pm 、P′ m Individuals before and after mutation, P max 、P min The upper and lower limits of the solution space are respectively defined, beta is the variation probability, and mg is the total iteration number.
3. The method of measuring the water salt content of rammed earth of claim 1, wherein establishing a generalized regression network model comprises:
setting a smoothing factor of the GRNN network, inputting a rammed earth capacitance value, conductivity, soil temperature and soil density, and setting and outputting the rammed earth capacitance value, conductivity, soil temperature and soil density as a rammed earth moisture content and a salt content; the input layer, the mode layer, the summation layer and the output layer are calculated.
4. The method for measuring the salt content of rammed earth according to claim 1, wherein the fitness of each particle is calculated, and the root mean square error RMSE between the actual value and the predicted value of the water content or the salt content is calculated as follows:
Figure FDA0004087023500000021
wherein
Figure FDA0004087023500000022
For the predicted value of the ith sample, y i For the actual value of the ith sample, n is the total number of samples.
5. The method of measuring the water salt content of rammed earth of claim 1, wherein the state of the particles is updated:
the following two formulas are carried out:
Figure FDA0004087023500000023
Figure FDA0004087023500000024
wherein i=1, 2, … M, M is the total number of particles in the population;
Figure FDA0004087023500000025
is the velocity vector of the ith particle at the kth iteration;
Figure FDA0004087023500000026
is the position vector of the ith particle at the kth iteration;
Figure FDA0004087023500000027
Is the individual optimal solution for round k,>
Figure FDA0004087023500000028
is the optimal solution of the k-th round of group; omega is an inertial factor; r is (r) 1 、r 2 Is a random number between (0, 1); c 1 、c 2 Is a learning factor. />
CN202310139327.6A 2023-02-20 2023-02-20 Rammed earth water salt content measurement method Pending CN116189794A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310139327.6A CN116189794A (en) 2023-02-20 2023-02-20 Rammed earth water salt content measurement method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310139327.6A CN116189794A (en) 2023-02-20 2023-02-20 Rammed earth water salt content measurement method

Publications (1)

Publication Number Publication Date
CN116189794A true CN116189794A (en) 2023-05-30

Family

ID=86437922

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310139327.6A Pending CN116189794A (en) 2023-02-20 2023-02-20 Rammed earth water salt content measurement method

Country Status (1)

Country Link
CN (1) CN116189794A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116660323A (en) * 2023-07-24 2023-08-29 四川省科源工程技术测试中心有限责任公司 Agricultural farmland saline-alkali degree sampling detection device and method

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116660323A (en) * 2023-07-24 2023-08-29 四川省科源工程技术测试中心有限责任公司 Agricultural farmland saline-alkali degree sampling detection device and method
CN116660323B (en) * 2023-07-24 2023-11-14 四川省科源工程技术测试中心有限责任公司 Agricultural farmland saline-alkali degree sampling detection device and method

Similar Documents

Publication Publication Date Title
Huan et al. Prediction of dissolved oxygen in aquaculture based on gradient boosting decision tree and long short-term memory network: A study of Chang Zhou fishery demonstration base, China
CN110084367B (en) Soil moisture content prediction method based on LSTM deep learning model
CN112288164B (en) Wind power combined prediction method considering spatial correlation and correcting numerical weather forecast
CN112733997B (en) Hydrological time series prediction optimization method based on WOA-LSTM-MC
CN113657028B (en) Online aerosol optical thickness prediction method based on multi-source information
CN116721537A (en) Urban short-time traffic flow prediction method based on GCN-IPSO-LSTM combination model
CN112991721A (en) Urban road network traffic speed prediction method based on graph convolution network node association degree
CN110401978A (en) Indoor orientation method based on neural network and particle filter multi-source fusion
CN114154401B (en) Soil erosion modulus calculation method and system based on machine learning and observation data
CN109754122A (en) A kind of Numerical Predicting Method of the BP neural network based on random forest feature extraction
CN110543978A (en) Traffic flow data prediction method and device based on wavelet neural network
CN114219131A (en) Watershed runoff prediction method based on LSTM
CN112765902B (en) Soft measurement modeling method for COD concentration in rural domestic sewage treatment process based on TentFWA-GD RBF neural network
CN115310348A (en) Stacking-based grouting amount integrated agent prediction model and prediction method
WO2022032874A1 (en) Adversarial neural network-based hydrological parameter calibration method for data region
CN116189794A (en) Rammed earth water salt content measurement method
Shang et al. Research on intelligent pest prediction of based on improved artificial neural network
WO2022032873A1 (en) Adversarial neural network-based hydrological parameter calibration method for data-lacking region
CN112163632A (en) Application of semi-supervised extreme learning machine based on bat algorithm in industrial detection
CN112614021A (en) Tunnel surrounding rock geological information prediction method based on built tunnel information intelligent identification
CN118332521B (en) Crust deformation time sequence simulation method based on particle swarm optimization random forest
Li et al. A data-driven rutting depth short-time prediction model with metaheuristic optimization for asphalt pavements based on RIOHTrack
CN117935988A (en) Method for predicting compressive strength of recycled coarse aggregate concrete based on support vector regression
CN114330815A (en) Ultra-short-term wind power prediction method and system based on improved GOA (generic object oriented architecture) optimized LSTM (least Square TM)
CN117709488B (en) Dam seepage prediction method based on RUN-XGBoost

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