CN117574781B - Intelligent prediction method and system for security risk of surrounding rock of underground factory building of pumped storage power station - Google Patents
Intelligent prediction method and system for security risk of surrounding rock of underground factory building of pumped storage power station Download PDFInfo
- Publication number
- CN117574781B CN117574781B CN202410050822.4A CN202410050822A CN117574781B CN 117574781 B CN117574781 B CN 117574781B CN 202410050822 A CN202410050822 A CN 202410050822A CN 117574781 B CN117574781 B CN 117574781B
- Authority
- CN
- China
- Prior art keywords
- surrounding rock
- solution
- optimization algorithm
- lstm
- neural network
- 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
- 239000011435 rock Substances 0.000 title claims abstract description 198
- 238000003860 storage Methods 0.000 title claims abstract description 48
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 71
- 238000003062 neural network model Methods 0.000 claims abstract description 71
- 238000012544 monitoring process Methods 0.000 claims abstract description 57
- 238000004458 analytical method Methods 0.000 claims abstract description 44
- 238000011156 evaluation Methods 0.000 claims abstract description 13
- 238000005457 optimization Methods 0.000 claims description 90
- 238000009826 distribution Methods 0.000 claims description 47
- 230000006870 function Effects 0.000 claims description 15
- 238000009412 basement excavation Methods 0.000 claims description 13
- 238000012163 sequencing technique Methods 0.000 claims description 13
- 239000013598 vector Substances 0.000 claims description 13
- 238000004891 communication Methods 0.000 claims description 9
- 230000007246 mechanism Effects 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 9
- 238000011161 development Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 230000035772 mutation Effects 0.000 claims description 6
- 238000012216 screening Methods 0.000 claims description 6
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 4
- 238000013461 design Methods 0.000 abstract description 6
- 238000010276 construction Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 4
- 230000007774 longterm Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000001174 ascending effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000005422 blasting Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000009423 ventilation 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
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A pumped storage power station underground powerhouse surrounding rock safety risk intelligent prediction method comprises the following steps: acquiring time sequence monitoring data of surrounding rocks of a ground factory building of the pumped storage power station; substituting the time sequence monitoring data into the corrected LSTM-CNN neural network model to predict the deformation quantity of the surrounding rock, and obtaining a prediction result of the deformation quantity of the surrounding rock; and performing surrounding rock stability analysis according to the surrounding rock deformation prediction result and the surrounding rock stability evaluation index to obtain surrounding rock stability analysis data. The design predicts the safety risk of surrounding rocks of the underground factory building of the pumped storage power station in advance, improves the prospective of risk prediction, and greatly avoids the possible safety problem.
Description
Technical Field
The invention relates to the field of pumped storage power stations, in particular to an intelligent prediction method and system for security risk of surrounding rock of an underground factory building of a pumped storage power station.
Background
The pumped storage power station is generally used for peak regulation, frequency modulation, phase modulation and accident standby of a power grid. Because the pumped storage power station is mostly in an underground factory building cavern group structure, the geological condition is complex, the construction site is long, the potential safety hazards of the caverns, the high side slope, the deep foundation pit and the like are many, and the safety risks exist in the construction processes of the cavern blasting, the side slope supporting, the foundation pit excavation and the like. A plurality of underground caverns such as a factory entering traffic hole, a ventilation and safety hole, a tail water hole and the like of some pumped storage power stations are intersected by a fracture zone, the geological condition of the fracture zone is poor, and collapse easily occurs in the process of excavating the underground caverns. The precedent of local block falling of rocks in tunnels occurs at the construction site of the factory entering traffic hole of the pumped storage power station, so that analysis of stability of surrounding rocks and risk prediction are necessary classes for construction of the pumped storage power station. Surrounding rock deformation monitoring is an important aspect of surrounding rock stability determination, and surrounding rock deformation is actually a long-term change process. In actual engineering, measured data are generally obtained through embedded monitoring instruments and equipment, and basic information is provided for surrounding rock stability judgment and deformation analysis. However, due to the limitation of multiple factors such as the service life of monitoring instruments and equipment, the monitoring environmental conditions and the like, long-term effective monitoring data are generally difficult to obtain, and great inconvenience is brought to long-term stability analysis and risk early warning of tunnels.
The Chinese patent application with the application number of CN202310110122.5 and the application date of 2023, 2 and 14 discloses an underground cavity surrounding rock deformation monitoring and early warning method and system, wherein the surrounding rock deformation threshold value after the underground cavity is excavated is determined by considering the rock mechanical property and the variability of the rock mechanical property, surrounding rock deformation information monitored and calculated by a laser convergence instrument is transmitted to a ground data center, the ground data center compares the surrounding rock deformation information with the surrounding rock deformation threshold value, early warning instructions with different grades are sent, and the early warning instructions are sent to a mobile phone of an underground constructor, so that the underground constructor can take corresponding measures in time. The monitoring and early warning method and the system disclosed by the invention can be used for continuously monitoring surrounding rock deformation after the underground cavity is excavated in real time on line, and rapidly sending early warning instructions of different grades to underground constructors according to surrounding rock deformation information and the preset surrounding rock deformation threshold value considering rock property variability, so that the underground constructors can take corresponding measures in time, thereby improving the construction safety and monitoring efficiency of the underground cavity and reducing the safety risk of the underground cavity. The method can realize real-time surrounding rock deformation monitoring and early warning, but lacks foresight and cannot early warn in advance.
The disclosure of this background section is only intended to increase the understanding of the general background of the present patent application and should not be taken as an admission or any form of suggestion that this information forms the prior art already known to a person of ordinary skill in the art.
Disclosure of Invention
The invention aims to solve the problem that early warning cannot be performed in advance in the prior art, and provides an intelligent prediction method and system for safety risk of surrounding rock of a pumped storage power station underground factory building.
In order to achieve the above object, the technical solution of the present invention is: a pumped storage power station underground powerhouse surrounding rock safety risk intelligent prediction method comprises the following steps:
acquiring time sequence monitoring data of surrounding rocks of a ground factory building of the pumped storage power station;
substituting the time sequence monitoring data into the corrected LSTM-CNN neural network model to predict the deformation quantity of the surrounding rock, and obtaining a prediction result of the deformation quantity of the surrounding rock;
performing surrounding rock stability analysis according to the surrounding rock deformation prediction result and the surrounding rock stability evaluation index to obtain surrounding rock stability analysis data;
and carrying out surrounding rock safety risk prediction through surrounding rock stability analysis data.
The timing monitoring data includes at least one of: excavation progress, surrounding rock deformation quantity, surrounding rock lithology, rock mass strength and surrounding rock quality score.
The timing monitoring data includes at least one of: excavation progress, surrounding rock deformation, surrounding rock lithology, rock mass strength and surrounding rock quality score;
substituting the time sequence monitoring data into the corrected LSTM-CNN neural network model to predict the deformation quantity of the surrounding rock, and obtaining a prediction result of the deformation quantity of the surrounding rock, wherein the method comprises the following steps:
an LSTM-CNN neural network model is established, the LSTM-CNN neural network model is trained based on the preprocessed time sequence monitoring data, and parameters of the LSTM-CNN neural network model are optimized through a mixed optimization algorithm consisting of an exponential distribution optimization algorithm and a convolution optimization algorithm, so that a surrounding rock deformation quantity prediction model is obtained and used for predicting future deformation quantity of surrounding rock.
The parameters for optimizing the LSTM-CNN neural network model by the mixed optimization algorithm consisting of the exponential distribution optimization algorithm and the convolution optimization algorithm are specifically as follows:
randomly initializing N initial solutions of an exponential distribution optimization algorithm in a solution space, wherein the dimension of the solution is D;
calculating and sequencing the fitness value of each current solution by taking the minimum mean square error of the LSTM-CNN neural network model as a fitness function;
with random probabilityEntering a development stage of an exponential distribution optimization algorithm, calculating a guide solution based on a fitness value sequencing result, and obtaining a new solution based on the guide solution;
entering an exploration stage of an exponential distribution optimization algorithm by using random probability 1-r, and obtaining a new solution based on the average value of the current solution;
calculating the fitness value of each new solution at present, and screening out the optimal solution;
introducing a solution quality enhancement mechanism of a convolution optimization algorithm into an exponential distribution optimization algorithm, and carrying out Gaussian variation with non-inertial weight on a D-dimensional search space of an optimal solution dimension by dimension to obtain the optimal solution;
and updating the global optimal solution, and entering the next iteration until the mean square error of the LSTM-CNN neural network model is minimum.
The formula for obtaining a new solution based on the guide solution is as follows:
；
wherein,is->New solution of the individual solution->To guide solution(s)>Rand is a random number between (0, 1), a +.>Is->Memory matrix of individual->Is->Historical optimal solution of the individual solutions, +.>Is [0,1 ]]Random numbers in between;
the formula for obtaining a new solution based on the average value of the current solution is as follows:
；
wherein,represents->For multiple iterations->Average of the solutions;
；
；
wherein,two solutions selected randomly from the current solutions;
the D-dimensional search space of the optimal solution is subjected to Gaussian variation with non-inertial weight in a dimension-by-dimension manner, and the optimal solution is obtained by the formula:
；
；
wherein,position vectors of d-th dimension of the optimal solution before and after mutation in the t-th iteration, +.>Wherein T is the maximum number of iterations, < ->For fitness function>For the updated optimal solution, ++>For random numbers satisfying a standard normal distribution, +.>The position vector of the d-th dimension which is the history optimal solution.
The intelligent prediction system for the safety risk of the surrounding rock of the underground factory building of the pumped storage power station comprises a data acquisition module, a deformation amount prediction module, a stability analysis module and a risk prediction module, wherein the data acquisition module is in signal connection with the deformation amount prediction module, the deformation amount prediction module is in signal connection with the stability analysis module, and the stability analysis module is in signal connection with the risk prediction module:
and a data acquisition module: the time sequence monitoring data of surrounding rock of the underground factory building of the pumped storage power station are obtained;
the deformation prediction module: substituting the time sequence monitoring data into the corrected LSTM-CNN neural network model to predict the deformation quantity of the surrounding rock, so as to obtain a prediction result of the deformation quantity of the surrounding rock;
stability analysis module: the method comprises the steps of performing surrounding rock stability analysis according to a surrounding rock deformation prediction result and a surrounding rock stability evaluation index to obtain surrounding rock stability analysis data;
risk prediction module: the method is used for predicting the security risk of the surrounding rock through the analysis data of the stability of the surrounding rock.
The timing monitoring data includes at least one of: excavation progress, surrounding rock deformation, surrounding rock lithology, rock mass strength and surrounding rock quality score;
substituting the time sequence monitoring data into the corrected LSTM-CNN neural network model to predict the deformation quantity of the surrounding rock, and obtaining a prediction result of the deformation quantity of the surrounding rock, wherein the method comprises the following steps:
an LSTM-CNN neural network model is established, the LSTM-CNN neural network model is trained based on the preprocessed time sequence monitoring data, and parameters of the LSTM-CNN neural network model are optimized through a mixed optimization algorithm consisting of an exponential distribution optimization algorithm and a convolution optimization algorithm, so that a surrounding rock deformation quantity prediction model is obtained and used for predicting future deformation quantity of surrounding rock.
The parameters for optimizing the LSTM-CNN neural network model by the mixed optimization algorithm consisting of the exponential distribution optimization algorithm and the convolution optimization algorithm are specifically as follows:
randomly initializing N initial solutions of an exponential distribution optimization algorithm in a solution space, wherein the dimension of the solution is D;
calculating and sequencing the fitness value of each current solution by taking the minimum mean square error of the LSTM-CNN neural network model as a fitness function;
with random probabilityEntering a development stage of an exponential distribution optimization algorithm, calculating a guide solution based on a fitness value sequencing result, and obtaining a new solution based on the guide solution;
entering an exploration stage of an exponential distribution optimization algorithm by using random probability 1-r, and obtaining a new solution based on the average value of the current solution;
calculating the fitness value of each new solution at present, and screening out the optimal solution;
introducing a solution quality enhancement mechanism of a convolution optimization algorithm into an exponential distribution optimization algorithm, and carrying out Gaussian variation with non-inertial weight on a D-dimensional search space of an optimal solution dimension by dimension to obtain the optimal solution;
and updating the global optimal solution, and entering the next iteration until the mean square error of the LSTM-CNN neural network model is minimum.
The formula for obtaining a new solution based on the guide solution is as follows:
；
wherein,is->New solution of the individual solution->To guide solution(s)>Rand is a random number between (0, 1), a +.>Is->Memory matrix of individual->Is->Historical optimal solution of the individual solutions, +.>Is [0,1 ]]Random numbers in between.
The formula for obtaining a new solution based on the average value of the current solution is as follows:
；
wherein,represents->For multiple iterations->Average of the solutions;
；
；
wherein,two solutions selected randomly from the current solution.
The D-dimensional search space of the optimal solution is subjected to Gaussian variation with non-inertial weight in a dimension-by-dimension manner, and the optimal solution is obtained by the formula:
；
；
wherein,position vectors of d-th dimension of the optimal solution before and after mutation in the t-th iteration, +.>Wherein T is the maximum number of iterations, < ->For fitness function>For the updated optimal solution, ++>For random numbers satisfying a standard normal distribution, +.>The position vector of the d-th dimension which is the history optimal solution.
Intelligent prediction equipment for safety risk of surrounding rock of underground factory building of pumped storage power station comprises: at least one processor, at least one memory, a communication interface, and a bus;
the processor, the memory and the communication interface complete communication with each other through the bus;
the memory stores program instructions executable by the processor that the processor invokes to implement the method as described above.
A computer readable storage medium storing computer instructions that cause a computer to implement a method as described above.
Compared with the prior art, the invention has the beneficial effects that:
1. according to the intelligent prediction method and system for surrounding rock safety risk of the underground powerhouse of the pumped storage powerhouse, parameters of an LSTM-CNN neural network model are corrected through a mixed optimization algorithm, preprocessed time sequence monitoring data are substituted into the corrected LSTM-CNN neural network model to conduct surrounding rock deformation prediction, surrounding rock deformation prediction results are obtained, when the intelligent prediction method and system are applied, parameters of the neural network model are corrected through the mixed algorithm, future surrounding rock deformation prediction is conducted based on historical time sequence monitoring data, and the surrounding rock deformation prediction results are used for stability analysis to predict surrounding rock safety risk of the underground powerhouse of the pumped storage powerhouse in advance, prospective of risk prediction is improved, and safety problems possibly occurring are avoided greatly. Therefore, the design can early warn in advance, and is high in prospective.
2. According to the intelligent prediction method and system for the surrounding rock safety risk of the underground powerhouse of the pumped storage power station, the parameters of the LSTM-CNN neural network model are optimized through the mixed optimization algorithm consisting of the exponential distribution optimization algorithm and the convolution optimization algorithm, and when the intelligent prediction method and system are applied, the parameters of the LSTM-CNN neural network model are optimized through the mixed optimization algorithm consisting of the exponential distribution optimization algorithm and the convolution optimization algorithm, so that the influence of using experience parameters on the prediction precision of the neural network model is reduced, and the prediction accuracy of the surrounding rock deformation prediction model is improved. Therefore, the design has high prediction accuracy and high efficiency.
3. According to the intelligent prediction method and system for the surrounding rock safety risk of the underground powerhouse of the pumped storage power station, a solution quality enhancement mechanism of a convolution optimization algorithm is introduced into an exponential distribution optimization algorithm, gaussian variation with non-inertial weight is carried out on a D-dimensional search space of an optimal solution in a dimension-by-dimension mode, the optimal solution is obtained, and when the method and system are applied, the solution quality enhancement mechanism of the convolution optimization algorithm is introduced into the exponential distribution optimization algorithm, gaussian variation with non-inertial weight is carried out on the search space of the optimal solution in the dimension-by-dimension mode, so that the algorithm is prevented from falling into a local optimal solution, and the parameter optimization capability is improved. Therefore, the optimization capability of the design parameters is improved, and the optimization effect is good.
Drawings
Fig. 1 is a flow chart of the present invention.
Fig. 2 is a system block diagram of the present invention.
Fig. 3 is a block diagram of the structure of the present invention.
Detailed Description
The invention is described in further detail below with reference to the accompanying drawings and detailed description.
Example 1:
referring to fig. 1, an intelligent prediction method for security risk of surrounding rock of a basement of a pumped storage power station includes:
s1, acquiring time sequence monitoring data of surrounding rocks of a ground factory building of a pumped storage power station, and preprocessing the time sequence monitoring data to obtain preprocessed time sequence monitoring data;
s2, pre-training an LSTM-CNN neural network model, correcting parameters of the LSTM-CNN neural network model through a mixed optimization algorithm, substituting the preprocessed time sequence monitoring data into the corrected LSTM-CNN neural network model to predict surrounding rock deformation, and obtaining a surrounding rock deformation prediction result;
s3, performing surrounding rock stability analysis according to the surrounding rock deformation prediction result and the surrounding rock stability evaluation index to obtain surrounding rock stability analysis data;
s4, performing surrounding rock safety risk prediction through surrounding rock stability analysis data.
When in application, the method comprises the following steps:
for surrounding rock, the excavation progress and the characteristics of the surrounding rock can have certain influence on surrounding rock deformation, so that a target monitoring area is divided into a plurality of monitoring sections, the excavation progress is recorded, a measuring point is selected on each section to be provided with a displacement sensor, surrounding rock deformation data are collected for a period of time, corresponding surrounding rock characteristics, rock strength and surrounding rock quality scores are analyzed, the recorded excavation progress, the collected surrounding rock deformation, the analyzed surrounding rock characteristics, rock strength and surrounding rock quality scores and the like form time sequence monitoring data together. Specifically, deleting and filling monitoring data which do not meet requirements to obtain equidistant time sequences, taking a fixed time interval as a time step number, then performing ADF (automatic frequency correction) inspection to judge whether stability requirements are met, building an LSTM-CNN neural network model, inserting an LSTM layer between an embedding layer and a CNN layer on the basis of a textCNN model structure, encoding the input time sequence monitoring data through the LSTM layer, outputting each time step, wherein the output of each time step comprises not only characteristic vector information of the current time but also characteristic information of front and rear time data, performing deep-level characteristic extraction through CNN, performing time sequence prediction better, training the LSTM-CNN neural network model based on the preprocessed time sequence monitoring data, optimizing parameters of the LSTM-CNN neural network model through a mixed optimization algorithm consisting of an exponential distribution optimization algorithm and a convolution optimization algorithm, obtaining a surrounding rock deformation quantity prediction model, and predicting the future deformation quantity of the surrounding rock based on the surrounding rock deformation quantity prediction model.
Example 2:
example 2 is substantially the same as example 1 except that:
the intelligent prediction method for surrounding rock safety risk of a pumped storage power station underground factory building comprises the steps of pre-training a neural network model, correcting parameters of an LSTM-CNN neural network model through a hybrid optimization algorithm, substituting processed time sequence monitoring data into the corrected LSTM-CNN neural network model to predict surrounding rock deformation, and obtaining a surrounding rock deformation prediction result specifically as follows:
s21, training an LSTM-CNN neural network model by taking monitoring data of P continuous time steps in time sequence monitoring data as input and taking surrounding rock deformation of the P+1th time step as output;
s22, optimizing parameters of the LSTM-CNN neural network model by a hybrid optimization algorithm consisting of an exponential distribution optimization algorithm and a convolution optimization algorithm;
s221, randomly initializing N initial solutions of an exponential distribution optimization algorithm in a solution space;
setting a parameter range of an LSTM-CNN neural network model as a solution space [ L, U ]]Randomly initializing N initial positions as N initial solutions in a solution space，；
Wherein,d is the dimension of each solution, and the dimension D of the solution is the same as the number of parameters of the LSTM-CNN neural network model to be optimized; each solution represents a set of parameter combinations of the LSTM-CNN neural network model;
s222, calculating and sequencing the fitness value of each current solution by taking the minimum mean square error of the LSTM-CNN neural network model as a fitness function;
and taking the mean square error between the predicted value and the true value of the LSTM-CNN neural network model as an adaptability function of an exponential distribution optimization algorithm, and solving the minimum adaptability function value to obtain the optimal parameter combination of the LSTM-CNN neural network model.
Calculating the fitness value of each current solution through a fitness function, and arranging the solutions in ascending order, wherein the quality of the solutions which are ranked earlier is better;
s223, generating a random number r between (0 and 1), if r is more than 0.5, entering a development stage of an exponential distribution optimization algorithm, calculating a guide solution based on a fitness value sequencing result, obtaining a new solution based on the guide solution, obtaining first three preferred solutions of an ascending sequence sequencing result, and obtaining an average value of the three preferred solutions as the guide solution;
；
wherein,is->New solution of the individual solution->To guide solution(s)>Rand is a random number between (0, 1), a +.>Is->Memory matrix of individual->Is->Historical optimal solution of the individual solutions, +.>Is [0,1 ]]Random numbers in between;
s224, if r is less than or equal to 0.5, entering an exploration stage of an exponential distribution optimization algorithm, and obtaining a new solution based on the average value of the current solution;
calculating the average of N solutions at the t-th iterationA new solution is obtained based on the average of the current solutions, the formula being:
；
wherein, ；
；
wherein,two solutions selected randomly from the current solutions;
s225, calculating the fitness value of each new solution at present, and screening out an optimal solution;
s226, introducing a solution quality enhancement mechanism of a convolution optimization algorithm into an exponential distribution optimization algorithm, and carrying out Gaussian variation with non-inertial weight on a D-dimensional search space of an optimal solution in a dimension-by-dimension manner to obtain the optimal solution;
；
wherein,position vectors of d-th dimension of the optimal solution before and after mutation in the t-th iteration, +.>Wherein T is the maximum number of iterations, < ->For fitness function>For obtaining the optimal solution after, randn is a random number satisfying the normal distribution of the standard,/->A position vector of a d-th dimension which is a history optimal solution;
s227, returning to the step S222 to obtain a global better solution, and entering the next iteration until the minimum mean square error of the LSTM-CNN neural network model is reached.
Example 3:
example 3 is substantially the same as example 1 except that:
the intelligent prediction method for surrounding rock safety risk of underground factory building of pumped storage power station includes that surrounding rock deformation is an important index for surrounding rock stability analysis, a surrounding rock deformation prediction result can be used as one of surrounding rock stability evaluation indexes, and surrounding rock stability evaluation is carried out by establishing a surrounding rock stability analysis model through a hierarchical analysis method in combination with indexes such as temperature, humidity, pressure, underground water inflow influence, surrounding rock crack opening, actual excavation width, actual chamber height and the like.
Besides the stability analysis means, different multistage deformation threshold values can be set directly according to different excavation progress, surrounding rock lithology, rock mass strength and surrounding rock quality scores, future surrounding rock deformation prediction results are compared with the multistage deformation threshold values respectively, and surrounding rock stability is divided.
According to the method, surrounding rock safety risk prediction is carried out according to surrounding rock stability dividing results, potential risk early warning is timely carried out when poor stability is found, future surrounding rock deformation amount prediction is carried out based on historical time sequence monitoring data, an exponential distribution optimization algorithm and a convolution optimization algorithm are mixed in a model training process to form a mixed optimization algorithm to correct parameters of a neural network model, a solution quality enhancement mechanism of the convolution optimization algorithm is used for avoiding local optimization in each iteration by improving the solution quality, surrounding rock deformation amount prediction results are used for stability analysis, potential safety risk of surrounding rock of a pumped storage power station underground powerhouse can be predicted in advance, the prospective of risk prediction is improved, and time is striven for safety planning and risk investigation.
Example 4:
as shown in fig. 2, a prediction system of a pumped storage power station underground plant surrounding rock safety risk intelligent prediction method comprises a data acquisition module, a deformation amount prediction module, a stability analysis module and a risk prediction module, wherein the data acquisition module is in signal connection with the deformation amount prediction module, the deformation amount prediction module is in signal connection with the stability analysis module, and the stability analysis module is in signal connection with the risk prediction module:
and a data acquisition module: the method comprises the steps of acquiring time sequence monitoring data of surrounding rocks of a ground factory building of a pumped storage power station, and preprocessing the time sequence monitoring data to obtain preprocessed time sequence monitoring data;
the deformation prediction module: the method comprises the steps of pre-training an LSTM-CNN neural network model, correcting parameters of the LSTM-CNN neural network model through a mixed optimization algorithm, substituting the pre-processed time sequence monitoring data into the corrected LSTM-CNN neural network model to predict surrounding rock deformation, and obtaining a surrounding rock deformation prediction result;
stability analysis module: the method comprises the steps of performing surrounding rock stability analysis according to a surrounding rock deformation prediction result and a surrounding rock stability evaluation index to obtain surrounding rock stability analysis data;
risk prediction module: the method is used for predicting the security risk of the surrounding rock through the analysis data of the stability of the surrounding rock.
The method comprises the steps of pre-training an LSTM-CNN neural network model, correcting parameters of the LSTM-CNN neural network model through a mixed optimization algorithm, substituting processed time sequence monitoring data into the corrected LSTM-CNN neural network model to predict surrounding rock deformation, and obtaining a surrounding rock deformation prediction result specifically as follows:
an LSTM-CNN neural network model is established, the LSTM-CNN neural network model is trained based on the preprocessed time sequence monitoring data, and parameters of the LSTM-CNN neural network model are optimized through a mixed optimization algorithm consisting of an exponential distribution optimization algorithm and a convolution optimization algorithm, so that a surrounding rock deformation quantity prediction model is obtained and used for predicting future deformation quantity of surrounding rock.
The parameters for optimizing the LSTM-CNN neural network model by the mixed optimization algorithm consisting of the exponential distribution optimization algorithm and the convolution optimization algorithm are specifically as follows:
randomly initializing N initial solutions of an exponential distribution optimization algorithm in a solution space, wherein the dimension of the solution is D;
calculating and sequencing the fitness value of each current solution by taking the minimum mean square error of the LSTM-CNN neural network model as a fitness function;
with random probabilityEntering a development stage of an exponential distribution optimization algorithm, calculating a guide solution based on a fitness value sequencing result, and obtaining a new solution based on the guide solution;
entering an exploration stage of an exponential distribution optimization algorithm by using random probability 1-r, and obtaining a new solution based on the average value of the current solution;
calculating the fitness value of each new solution at present, and screening out the optimal solution;
introducing a solution quality enhancement mechanism of a convolution optimization algorithm into an exponential distribution optimization algorithm, and carrying out Gaussian variation with non-inertial weight on a D-dimensional search space of an optimal solution dimension by dimension to obtain the optimal solution;
and updating the global optimal solution, and entering the next iteration until the mean square error of the LSTM-CNN neural network model is minimum.
The formula for obtaining a new solution based on the guide solution is as follows:
；
wherein,is->New solution of the individual solution->To guide solution(s)>Rand is a random number between (0, 1), a +.>Is->Memory matrix of individual->Is->Historical optimal solution of the individual solutions, +.>Is [0,1 ]]Random numbers in between.
The formula for obtaining a new solution based on the average value of the current solution is as follows:
；
wherein,represents->For multiple iterations->Average of the solutions;
；
；
wherein,two solutions selected randomly from the current solution.
The D-dimensional search space of the optimal solution is subjected to Gaussian variation with non-inertial weight in a dimension-by-dimension manner, and the optimal solution is obtained by the formula:
；
；
wherein,position vectors of d-th dimension of the optimal solution before and after mutation in the t-th iteration, +.>Wherein T is the maximum number of iterations, < ->For fitness function>For the updated optimal solution, ++>For random numbers satisfying a standard normal distribution, +.>The position vector of the d-th dimension which is the history optimal solution.
Example 5:
as shown in fig. 3, the design further discloses an intelligent prediction device for security risk of surrounding rock of underground powerhouse of pumped storage power station, comprising: at least one processor, at least one memory, a communication interface, and a bus; the processor, the memory and the communication interface complete communication with each other through the bus; the memory stores program instructions executable by the processor that the processor invokes to implement the aforementioned methods of the present invention.
Example 6:
the present design also discloses a computer readable storage medium storing computer instructions for causing the computer to implement all or part of the steps of the method according to the embodiments of the present invention, the storage medium comprising: a usb disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In general, the computer instructions to implement the methods of the present invention may be carried in any combination of one or more computer-readable storage media. The non-transitory computer-readable storage medium may include any computer-readable medium, except the signal itself in temporary propagation.
The computer readable storage medium can be, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples of the computer-readable storage medium include the non-exhaustive list of: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory RAM, a read-only memory (ROM), an erasable programmable read-only memory (EKROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
The computer program code for carrying out operations of the present invention may be written in one or more programming languages, or combinations thereof, including an object oriented programming language such as Java, SMalltalk, C ++ and conventional procedural programming languages, such as the "C" language or similar programming languages, particularly Python languages suitable for neural network computing and TensorFlow, pyTorch-based platform frameworks may be used. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of remote computers, the remote computer may be connected to the user computer through any number of types of networks, including a Local Area Network (LAN) or a Wide Area Network (WAN), or connected to an external computer through the Internet using, for example, an Internet service provider.
The above-mentioned devices and non-transitory computer readable storage medium can refer to a specific description of a pumped storage power station underground building surrounding rock safety risk intelligent prediction method and beneficial effects, and are not described herein.
While embodiments of the present invention have been shown and described above, it should be understood that the above embodiments are illustrative and not to be construed as limiting the invention, and that variations, modifications, alternatives, and variations may be made to the above embodiments by one of ordinary skill in the art within the scope of the invention.
Claims (4)
1. The intelligent prediction method for the safety risk of the surrounding rock of the underground powerhouse of the pumped storage power station is characterized by comprising the following steps:
acquiring time sequence monitoring data of surrounding rocks of a ground factory building of the pumped storage power station;
substituting the time sequence monitoring data into the corrected LSTM-CNN neural network model to predict the deformation quantity of the surrounding rock, and obtaining a prediction result of the deformation quantity of the surrounding rock;
performing surrounding rock stability analysis according to the surrounding rock deformation prediction result and the surrounding rock stability evaluation index to obtain surrounding rock stability analysis data;
performing surrounding rock safety risk prediction through surrounding rock stability analysis data;
the time sequence monitoring data comprise excavation progress, surrounding rock deformation quantity, surrounding rock lithology, rock mass strength and surrounding rock quality score;
substituting the time sequence monitoring data into the corrected LSTM-CNN neural network model to predict the deformation quantity of the surrounding rock, and obtaining a prediction result of the deformation quantity of the surrounding rock, wherein the method comprises the following steps:
establishing an LSTM-CNN neural network model, wherein the structure of the LSTM-CNN neural network model is to insert an LSTM layer between an ebedding layer and a CNN layer of a textCNN model structure;
taking monitoring data of continuous P time steps in the preprocessed time sequence monitoring data as input, taking surrounding rock deformation quantity of P+1th time step as output, training the LSTM-CNN neural network model, and optimizing parameters of the LSTM-CNN neural network model through a mixed optimization algorithm consisting of an exponential distribution optimization algorithm and a convolution optimization algorithm to obtain a surrounding rock deformation quantity prediction model for predicting future deformation quantity of surrounding rock;
the parameters for optimizing the LSTM-CNN neural network model by the mixed optimization algorithm consisting of the exponential distribution optimization algorithm and the convolution optimization algorithm are specifically as follows:
randomly initializing N initial solutions of an exponential distribution optimization algorithm in a solution space, wherein the dimension of the solution is D;
calculating and sequencing the fitness value of each current solution by taking the minimum mean square error of the LSTM-CNN neural network model as a fitness function;
entering a development stage of an exponential distribution optimization algorithm by using a random probability r epsilon (0, 1), calculating a guide solution based on a fitness value sequencing result, and obtaining a new solution based on the guide solution;
entering an exploration stage of an exponential distribution optimization algorithm by using random probability 1-r, and obtaining a new solution based on the average value of the current solution;
calculating the fitness value of each new solution at present, and screening out the optimal solution;
introducing a solution quality enhancement mechanism of a convolution optimization algorithm into an exponential distribution optimization algorithm, and carrying out Gaussian variation with non-inertial weight on a D-dimensional search space of an optimal solution dimension by dimension to obtain the optimal solution;
updating the global optimal solution, and entering the next iteration until the mean square error of the LSTM-CNN neural network model is minimum;
the formula for obtaining a new solution based on the guide solution is as follows:
wherein,for the new solution of the ith solution, < +.>For the guided solution, λ=2×rand-1, rand is a random number between (0, 1), +.>Is the memory matrix of the ith individual, +.>For the historical optimal solution of the ith solution, σ is [0,1]Random numbers in between;
the formula for obtaining a new solution based on the average value of the current solution is as follows:
wherein,for the new solution of the ith solution, < +.>For the history optimal solution of the ith solution, < +.>Represents the average of N solutions at the t-th iteration;
wherein,two solutions selected randomly from the current solutions;
the D-dimensional search space of the optimal solution is subjected to Gaussian variation with non-inertial weight in a dimension-by-dimension manner, and the optimal solution is obtained by the formula:
wherein,position vector of d-th dimension of optimal solution before and after mutation in T-th iteration, ω=1- (T/T), respectively 2 Wherein T is the maximum iteration number, f is the fitness function, ++>For the updated optimal solution, randn is a solution meeting the standard normal distributionRandom number (R)/(R)>A position vector of a d-th dimension which is a history optimal solution;
the surrounding rock stability analysis according to the surrounding rock deformation prediction result and the surrounding rock stability evaluation index specifically comprises the following steps: and taking the deformation quantity prediction result of the surrounding rock as one of surrounding rock stability evaluation indexes, and establishing a surrounding rock stability analysis model by a hierarchical analysis method by combining the indexes of temperature, humidity, pressure, underground water inflow influence, surrounding rock crack opening, actual excavation width and actual chamber height to evaluate the surrounding rock stability.
2. Intelligent prediction system of pumped storage power station underground powerhouse surrounding rock safety risk, which is characterized by comprising:
and a data acquisition module: the time sequence monitoring data of surrounding rock of the underground factory building of the pumped storage power station are obtained;
the deformation prediction module: substituting the time sequence monitoring data into the corrected LSTM-CNN neural network model to predict the deformation quantity of the surrounding rock, so as to obtain a prediction result of the deformation quantity of the surrounding rock;
stability analysis module: the method comprises the steps of performing surrounding rock stability analysis according to a surrounding rock deformation prediction result and a surrounding rock stability evaluation index to obtain surrounding rock stability analysis data;
risk prediction module: the method comprises the steps of performing surrounding rock safety risk prediction through surrounding rock stability analysis data;
the time sequence monitoring data comprise excavation progress, surrounding rock deformation quantity, surrounding rock lithology, rock mass strength and surrounding rock quality score;
substituting the time sequence monitoring data into the corrected LSTM-CNN neural network model to predict the deformation quantity of the surrounding rock, and obtaining a prediction result of the deformation quantity of the surrounding rock, wherein the method comprises the following steps:
an LSTM-CNN neural network model is established, and an LSTM layer is inserted between the embedding layer and the CNN layer on the basis of the structure of the textCNN model;
taking monitoring data of continuous P time steps in the preprocessed time sequence monitoring data as input, taking surrounding rock deformation quantity of P+1th time step as output, training the LSTM-CNN neural network model, and optimizing parameters of the LSTM-CNN neural network model through a mixed optimization algorithm consisting of an exponential distribution optimization algorithm and a convolution optimization algorithm to obtain a surrounding rock deformation quantity prediction model for predicting future deformation quantity of surrounding rock;
the parameters for optimizing the LSTM-CNN neural network model by the mixed optimization algorithm consisting of the exponential distribution optimization algorithm and the convolution optimization algorithm are specifically as follows:
randomly initializing N initial solutions of an exponential distribution optimization algorithm in a solution space, wherein the dimension of the solution is D;
calculating and sequencing the fitness value of each current solution by taking the minimum mean square error of the LSTM-CNN neural network model as a fitness function;
entering a development stage of an exponential distribution optimization algorithm by using a random probability r epsilon (0, 1), calculating a guide solution based on a fitness value sequencing result, and obtaining a new solution based on the guide solution;
entering an exploration stage of an exponential distribution optimization algorithm by using random probability 1-r, and obtaining a new solution based on the average value of the current solution;
calculating the fitness value of each new solution at present, and screening out the optimal solution;
introducing a solution quality enhancement mechanism of a convolution optimization algorithm into an exponential distribution optimization algorithm, and carrying out Gaussian variation with non-inertial weight on a D-dimensional search space of an optimal solution dimension by dimension to obtain the optimal solution;
updating the global optimal solution, and entering the next iteration until the mean square error of the LSTM-CNN neural network model is minimum;
the formula for obtaining a new solution based on the guide solution is as follows:
wherein,for the new solution of the ith solution, < +.>For the guided solution, λ=2×rand-1, rand is a random number between (0, 1), +.>Is the memory matrix of the ith individual, +.>For the historical optimal solution of the ith solution, σ is [0,1]Random numbers in between;
the formula for obtaining a new solution based on the average value of the current solution is as follows:
wherein,for the new solution of the ith solution, < +.>For the history optimal solution of the ith solution, < +.>Represents the average of N solutions at the t-th iteration;
wherein,two solutions selected randomly from the current solutions;
the D-dimensional search space of the optimal solution is subjected to Gaussian variation with non-inertial weight in a dimension-by-dimension manner, and the optimal solution is obtained by the formula:
wherein,position vectors of d-th dimension of the optimal solution before and after mutation in the T-th iteration, ω=1- (T-T), respectively 2 Wherein T is the maximum iteration number, f is the fitness function, ++>For the updated optimal solution, randn is a random number satisfying the normal distribution of the standard, ++>A position vector of a d-th dimension which is a history optimal solution;
the surrounding rock stability analysis according to the surrounding rock deformation prediction result and the surrounding rock stability evaluation index specifically comprises the following steps: and taking the deformation quantity prediction result of the surrounding rock as one of surrounding rock stability evaluation indexes, and establishing a surrounding rock stability analysis model by a hierarchical analysis method by combining the indexes of temperature, humidity, pressure, underground water inflow influence, surrounding rock crack opening, actual excavation width and actual chamber height to evaluate the surrounding rock stability.
3. Intelligent prediction equipment of pumped storage power station underground powerhouse surrounding rock safety risk, its characterized in that includes: at least one processor, at least one memory, a communication interface, and a bus;
the processor, the memory and the communication interface complete communication with each other through the bus;
the memory stores program instructions executable by the processor, the processor invoking the program instructions to implement the method of claim 1.
4. A computer readable storage medium storing computer instructions for causing a computer to implement the method of claim 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410050822.4A CN117574781B (en) | 2024-01-15 | 2024-01-15 | Intelligent prediction method and system for security risk of surrounding rock of underground factory building of pumped storage power station |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410050822.4A CN117574781B (en) | 2024-01-15 | 2024-01-15 | Intelligent prediction method and system for security risk of surrounding rock of underground factory building of pumped storage power station |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117574781A CN117574781A (en) | 2024-02-20 |
| CN117574781B true CN117574781B (en) | 2024-04-16 |
Family
ID=89890338
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410050822.4A Active CN117574781B (en) | 2024-01-15 | 2024-01-15 | Intelligent prediction method and system for security risk of surrounding rock of underground factory building of pumped storage power station |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117574781B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111291934A (en) * | 2020-02-18 | 2020-06-16 | 山东科技大学 | Surrounding rock real-time grading prediction and self-checking method in tunnel construction process |
| CN112597694A (en) * | 2020-05-28 | 2021-04-02 | 中国人民解放军陆军工程大学 | Neural network-based surrounding rock deformation prediction system and prediction method |
| CN115049173A (en) * | 2022-08-17 | 2022-09-13 | 中国石油大学(华东) | Deep learning and Eaton method coupling driving stratum pore pressure prediction method |
| CN115929406A (en) * | 2022-12-07 | 2023-04-07 | 河海大学 | Neural network-based surrounding rock stability prediction method, system and storage medium |
| CN116644970A (en) * | 2023-05-10 | 2023-08-25 | 福建工程学院 | Photovoltaic power prediction method based on VMD decomposition and lamination deep learning |
| CN117113842A (en) * | 2023-08-29 | 2023-11-24 | 福建理工大学 | Tunnel surrounding rock deformation prediction method based on PSO-LSTM model |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9646115B2 (en) * | 2012-04-17 | 2017-05-09 | Schlumberger Technology Corporation | Determining a limit of failure in a wellbore wall |
-
2024
- 2024-01-15 CN CN202410050822.4A patent/CN117574781B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111291934A (en) * | 2020-02-18 | 2020-06-16 | 山东科技大学 | Surrounding rock real-time grading prediction and self-checking method in tunnel construction process |
| CN112597694A (en) * | 2020-05-28 | 2021-04-02 | 中国人民解放军陆军工程大学 | Neural network-based surrounding rock deformation prediction system and prediction method |
| CN115049173A (en) * | 2022-08-17 | 2022-09-13 | 中国石油大学(华东) | Deep learning and Eaton method coupling driving stratum pore pressure prediction method |
| CN115929406A (en) * | 2022-12-07 | 2023-04-07 | 河海大学 | Neural network-based surrounding rock stability prediction method, system and storage medium |
| CN116644970A (en) * | 2023-05-10 | 2023-08-25 | 福建工程学院 | Photovoltaic power prediction method based on VMD decomposition and lamination deep learning |
| CN117113842A (en) * | 2023-08-29 | 2023-11-24 | 福建理工大学 | Tunnel surrounding rock deformation prediction method based on PSO-LSTM model |
Non-Patent Citations (6)
| Title |
|---|
| COA-CNN-LSTM: Coati optimization algorithm-based hybrid deep learning model for PV/wind power forecasting in smart grid applications;Abou Houran, Mohamad, et al.;《Applied Energy》;20231231;第349卷;121638 * |
| Evolving CNN-LSTM Models for Time Series Prediction Using Enhanced Grey Wolf Optimizer.《IEEE Access》.2020,第8卷161519-161541. * |
| Exponential distribution optimizer (EDO): a novel math-inspired algorithm for global optimization and engineering problems;Abdel-Basset, M., El-Shahat, D., Jameel, M. et al.;《Artif Intell Rev 56》;20230131;9329–9400 * |
| 基于二维卷积运算的智能优化算法;陈克伟;魏曙光;张嘉曦;《装甲兵学报》;20230131;第2卷(第1期);102-108 * |
| 水电站地下厂房的围岩稳定分析与变形预测;金长宇;《中国博士学位论文全文数据库 工程科技Ⅱ辑》;20080229;C037-6 * |
| 雷健雄，王黎理.《装甲兵学报》.机械工业出版社,2019,(第2版),110. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117574781A (en) | 2024-02-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wu et al. | Rock burst prediction probability model based on case analysis | |
| Haas et al. | Updating the decision aids for tunneling | |
| US11989655B2 (en) | Prediction method, device and system for rock mass instability stages | |
| EP3246858A1 (en) | Computer implemented method for generating a field development plan (fdp) for the exploitation of oil and gas reservoirs | |
| CN113806843B (en) | Deformation analysis system and method based on dynamic fluctuation of sedimentation tank bottom | |
| CN117196350B (en) | Mine geological environment characteristic monitoring and recovery treatment method and system | |
| CN111305899B (en) | Method for determining removal length of temporary support for construction of subway station arch cover method | |
| CN115130375A (en) | Rock burst intensity prediction method | |
| CN111583067A (en) | Urban underground large space construction safety early warning and emergency decision-making method and system | |
| CN111144636B (en) | Slope deformation prediction method | |
| CN116050603A (en) | Method and equipment for predicting and optimizing deformation of undercut tunnel based on hybrid intelligent method | |
| CN115688251A (en) | Earthquake multi-occurrence-zone tunnel risk decision method and system based on deep learning | |
| Xue et al. | PREDICTION OF SLOPE STABILITY BASED ON GA-BP HYBRID ALGORITHM. | |
| CN117574781B (en) | Intelligent prediction method and system for security risk of surrounding rock of underground factory building of pumped storage power station | |
| US10997326B2 (en) | Time-to-finish simulation forecaster | |
| CN104598705A (en) | Method and device for recognizing underground material layer | |
| CN112926267A (en) | TBM tunnel rock burst grade prediction method and system based on tunneling parameter inversion | |
| CN114117654A (en) | Horizontal well perforation optimization design method and device based on machine learning | |
| CN117328939A (en) | Tunnel construction plan generation method, system, equipment and medium | |
| CN117113644A (en) | Slope temporary slip forecasting method, system and medium based on deep monitoring deformation sequence | |
| CN112906239B (en) | Comprehensive pipe rack safety state evaluation method and device, processor and storage medium | |
| CN110909402B (en) | Advanced small catheter design method based on neural network technology | |
| CN114792209B (en) | Method, equipment and storage medium for engineering construction risk assessment | |
| CN117852402B (en) | Tunnel surrounding rock stability prediction method, device, equipment and storage medium | |
| CN117871293B (en) | Self-correction-based high strain force detection method and detection device |
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 |