Gravity dam deep sliding reliability calculation method based on optimized extreme learning machine

By optimizing the extreme learning machine model and particle swarm optimization algorithm, the problems of improper parameter setting and the combined influence of multiple factors in the anti-slip stability analysis of gravity dams are solved, and a higher accuracy calculation of deep sliding reliability of gravity dams is achieved. More accurate anti-slip stability analysis support.

CN119047016AActive Publication Date: 2024-11-29YANGZHOU UNIV
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202410839972.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-06-26
Publication Date
2024-11-29
Estimated Expiration
2044-06-26

AI Technical Summary

Technical Problem

The existing anti-slip stability analysis method of gravity dam based on machine learning algorithms has problems such as improper parameter setting, neglecting the comprehensive influence of multiple factors, and lacking the ability to describe the sliding surface, resulting in deviations from the actual situation.

Method used

Using the Optimization Extreme Learning Machine (ELM) model, the particle swarm optimization algorithm SHPSO optimizes parameters, combined with the multi-strategy fusion improvement algorithm, a model that can more accurately describe the deep sliding of gravity dams is constructed, and a comprehensive consideration of multiple geological conditions and load conditions is considered. Influence.

Benefits of technology

The accuracy of the calculation of deep sliding reliability of gravity dams is improved, and more accurate anti-slip stability analysis data is provided, making up for the shortcomings of existing methods.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119047016A_ABST
    Figure CN119047016A_ABST
Patent Text Reader

Abstract

The invention discloses a gravity dam deep sliding reliability calculation method based on an optimized extreme learning machine, and relates to the technical field of hydraulic structure engineering, and the method comprises the steps: obtaining the basic information of a gravity dam; optimizing the extreme learning machine through a multi-strategy fusion improved particle swarm optimization (SHPSO) algorithm, and constructing an ELM model of the optimized extreme learning machine; and according to the optimal parameters of the optimized extreme learning machine ELM model, constructing the optimal ELM model, and calculating the deep sliding reliability of the gravity dam. A foundation is laid for calculation by obtaining the basic information of the gravity dam; an extreme learning machine (ELM) parameter is used as an optimization target, the ELM parameter is optimized by using a multi-strategy fusion improved particle swarm optimization (SHPSO) algorithm, and the prediction precision and generalization ability of an ELM model are improved; and an optimal model is constructed according to optimal parameters of the optimized ELM model, random samples are extracted and input for calculation, the deep sliding failure rate and reliability of the gravity dam can be accurately obtained, accurate data support is provided for anti-sliding stability analysis of the dam body, and the defects of an existing method are overcome.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of hydraulic structure engineering, in particular to a method for calculating the reliability of deep sliding of a gravity dam based on an optimized extreme learning machine. Background Art

[0002] In recent years, with the continuous expansion of engineering projects and the increasing complexity of engineering conditions, the anti-sliding stability analysis of gravity dams has become a key link in the construction of major water conservancy and hydropower projects. Traditional gravity dam anti-sliding stability analysis methods are mostly based on empirical formulas or simplified assumptions, which are difficult to accurately reflect the actual engineering conditions and therefore cannot meet the needs of modern engineering design. To solve this problem, in recent years, scholars have begun to apply machine learning algorithms to analyze and predict the anti-sliding stability of dam bodies.

[0003] Although the existing anti-sliding stability analysis methods of gravity dams based on machine learning algorithms have improved the calculation accuracy to a certain extent, there are still some shortcomings. First, the machine learning algorithm parameters used in most methods are improperly set, and they cannot effectively mine the deep nonlinear mapping relationship, thus affecting the prediction accuracy. Secondly, the existing methods often only consider a single influencing factor, ignoring the comprehensive influence of various geological conditions and load conditions on the stability of the dam body, resulting in a deviation between the calculation results and the actual situation. In addition, the deep sliding mechanism of gravity dams is complex, and many methods lack the ability to accurately describe the position and morphology of the sliding surface. Therefore, a deep sliding reliability calculation method for gravity dams based on an optimized extreme learning machine is proposed. Summary of the invention

[0004] In view of the problems of improper parameter setting, neglect of the comprehensive influence of multiple factors and lack of ability to describe the sliding surface in the existing gravity dam anti-sliding stability analysis method based on machine learning algorithm, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is how to optimize the parameters of the extreme learning machine and comprehensively consider multiple influencing factors, so as to improve the accuracy of the calculation of the deep sliding reliability of the gravity dam.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides a method for calculating the deep sliding reliability of a gravity dam based on an optimized extreme learning machine, which includes: the process of constructing an optimized extreme learning machine ELM model includes the following steps: a set of input weights and biases of the extreme learning machine ELM are converted into the SHPSO particle position x i , the mean square error MSE obtained through training is used as the fitness function of the SHPSO particle, that is, the position of the SHPSO particle x i The relevant formula is as follows:

[0008] x i =[w 11 ,w 12 ,...,w 1n ,...,w k1 ,w k2 ,...,w kn ,b1,b2,...,b k ]

[0009] Among them, w kn is the weight between the kth hidden layer neural node and the nth input layer node, b k is the bias of the kth hidden layer node.

[0010] Randomly select a part of the given sample set as the training set and the rest as the test set; determine the parameters, set the population size, maximum number of iterations, search space range, search space dimension, number of ELM hidden layer nodes and activation function of the SHPSO algorithm; initialize the particle position x i and particle velocity v i ; Call the extreme learning machine ELM to calculate the fitness of each particle and initialize the individual optimal value; According to the fitness value of each particle, the particles are divided into elite particles, ordinary particles and inferior particles; Perform hierarchical learning on the divided particles and update the particle position x i , particle speed v i and particle fitness value; determine whether the individual optimal value of the elite particle is improved and update the individual optimal value; at the same time, determine whether the number of fitness evaluations has reached the maximum number of evaluations, and obtain the input layer weights and biases based on the optimal individuals to build an optimized extreme learning machine ELM model; use multi-strategy fusion to improve the particle swarm optimization algorithm SHPSO to find the optimal parameters of the extreme learning machine ELM; use the test set to verify the generalization ability of the optimized extreme learning machine ELM model.

[0011] As a preferred solution of the method for calculating the deep sliding reliability of the gravity dam by optimizing the extreme learning machine of the present invention, the fitness evaluation times include: when the fitness evaluation times reaches the maximum evaluation times, the personal optimal value is output, that is, the relevant formula for updating the personal optimal values ​​of all particles is as follows:

[0012]

[0013] Among them, p besti is the personal optimal value of particle i, is the fitness value of the t+1th iteration, is the fitness value at the tth iteration, f(p besti ) is the fitness value of the optimal value of particle i.

[0014] When the number of fitness evaluations does not reach the maximum number of evaluations, the particles are divided again.

[0015] As an optimal solution for the method for calculating the deep sliding reliability of gravity dams using the optimized extreme learning machine of the present invention, the optimized extreme learning machine ELM model includes a data preprocessing module, a parameter optimization module and a prediction module; the data preprocessing module is used for data normalization to eliminate the influence of numerical magnitude; the parameter optimization module searches for the optimal parameters of the extreme learning machine ELM by improving the particle swarm optimization algorithm SHPSO through multi-strategy fusion; the prediction module is used to input the required data into the optimized extreme learning machine ELM model and output predicted values.

[0016] As a preferred solution of the method for calculating the deep sliding reliability of the gravity dam by optimizing the extreme learning machine of the present invention, the multi-strategy fusion improved particle swarm optimization algorithm SHPSO includes the following steps: using the inertia weight adaptive change strategy AW to update the particle speed and position, the update formula is as follows:

[0017]

[0018] Among them, u(i) is the inertia weight direction change coefficient, rand is a random number, is the fitness value of the t+1th iteration, is the fitness value at the tth iteration.

[0019] According to the fitness value of the particles after each iteration, the direction of the inertia weight is adaptively adjusted to change the direction of particle movement; the particles are divided into dominant layer, common layer and inferior layer according to the particle fitness ranking, and an adaptive hierarchical learning strategy is adopted for the characteristics of particles in each layer; differential mutation is performed on the particle group type that is the dominant layer.

[0020] As a preferred solution of the method for calculating the deep sliding reliability of the gravity dam by optimizing the extreme learning machine of the present invention, the differential variation includes: if the fitness value of the t+1th iteration is greater than the fitness value under the tth iteration, it is judged that the fitness value of the elite particle i under the t+1th iteration has not been improved, and the flight direction of the elite particle deviates from the actual personal optimal value, and the inertia weight direction is changed by making the inertia weight direction coefficient a negative random number, so that the elite particle flies in the opposite direction of the current movement in the next iteration; if the fitness value of the t+1th iteration is Less than or equal to the fitness value at the tth iteration Then the inertia weight directional coefficient is set to 1, and the inertia weight adaptive change strategy AW is used to encourage the elite particles to continue flying in the current speed direction.

[0021] As a preferred solution of the method for calculating the deep sliding reliability of the gravity dam by optimizing the extreme learning machine of the present invention, the construction process of the optimal ELM model includes the following steps: obtaining a sample set and normalizing it; inputting the normalized sample set into the extreme learning machine ELM, and taking the mean square error MSE as the fitness evaluation function of the extreme learning machine ELM, that is, the relevant formula of the fitness evaluation function is as follows:

[0022]

[0023] Where N is the number of samples, t i is the actual value corresponding to the i-th sample, is the simulated predicted value of the i-th sample.

[0024] Based on the optimal parameters, the particle swarm optimization algorithm is improved by multi-strategy fusion to find the optimal weight and bias of ELM and build the optimal ELM model; extract n s Set up random variable X = (x1, x2, ..., x6) samples, according to the variance, mean and distribution type of the random variable, use MATLAB distribution type functions to extract n s Group samples; input random variable samples into the optimal ELM model and predict n s The group function value; the prediction result of the statistical optimal ELM model; the prediction result includes failure rate and reliability, that is, the relevant specific formula is as follows:

[0025]

[0026] Among them, P f is the failure rate, n f is the total number of predicted results less than zero, n f is the total number of predicted results greater than zero, P r For reliability.

[0027] In the second aspect, an embodiment of the present invention provides a deep sliding reliability calculation system for a gravity dam based on an optimized extreme learning machine, which includes: an acquisition module for acquiring basic information of the gravity dam; a construction module for optimizing the extreme learning machine by improving the particle swarm optimization algorithm SHPSO through multi-strategy fusion, and constructing an optimized extreme learning machine ELM model; a calculation module for constructing an optimal ELM model based on the optimized extreme learning machine ELM model, and calculating the deep sliding reliability of the gravity dam.

[0028] In a third aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the method for calculating the deep sliding reliability of a gravity dam by optimizing the extreme learning machine as described in the first aspect of the present invention are implemented.

[0029] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of the method for calculating the deep sliding reliability of a gravity dam by optimizing the extreme learning machine as described in the first aspect of the present invention are implemented.

[0030] The beneficial effects of the present invention are as follows: the present invention lays the foundation for calculation by acquiring basic information of the gravity dam; takes the extreme learning machine ELM parameters as the optimization target, and utilizes the particle swarm optimization algorithm SHPSO improved by multi-strategy fusion to optimize the ELM parameters, thereby improving the prediction accuracy and generalization ability of the ELM model; constructs an optimal model according to the optimal parameters of the optimized ELM model, extracts random samples for input calculation, and can accurately obtain the deep sliding failure rate and reliability of the gravity dam, thereby providing accurate data support for the anti-sliding stability analysis of the dam body, and making up for the shortcomings of the existing methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0032] Figure 1 This is a flow chart of the optimized extreme learning machine ELM model of the method for calculating the deep sliding reliability of a gravity dam using the optimized extreme learning machine in Example 1.

[0033] Figure 2 A cross-sectional view of a gravity dam that optimizes the extreme learning machine method for calculating the deep sliding reliability of a gravity dam in Example 1.

[0034] Figure 3 This is the comparison result between the extreme learning machine ELM model for deep anti-sliding of gravity dams that optimizes the extreme learning machine in Example 1 and the optimal ELM model.

[0035] Figure 4 This is a true value error diagram of the optimized extreme learning machine ELM model before the optimization of the method for calculating the deep sliding reliability of a gravity dam using the optimized extreme learning machine in Example 1.

[0036] Figure 5 Example 1 is a prediction value error diagram of the optimized extreme learning machine ELM model of the optimized extreme learning machine for calculating the deep sliding reliability of a gravity dam. DETAILED DESCRIPTION

[0037] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0038] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0039] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0040] Example 1

[0041] Reference Figure 1-2 , which is the first embodiment of the present invention, and provides a method for calculating the reliability of deep sliding of a gravity dam based on an optimized extreme learning machine, comprising:

[0042] S1: Obtain basic information of gravity dam.

[0043] Specifically, a concrete gravity dam is a secondary structure. The dam section is as follows: Figure 2 As shown, the dam crest height is 70m, the dam crest width is 7m, the water depth in front of the dam is 65m, the width of the dam bottom foundation surface is 45.5m, the downstream dam slope ratio is 1:0.8, the curtain is 10m away from the dam heel, and the structural planes AB and BC are two weak structural planes in the bedrock, with a width of 40cm. The angles between the two weak structural planes and the horizontal plane are and respectively. With the dam heel as the origin, the coordinates of point A are (-5,0), the coordinates of point B are (48.25,-14.27), and the coordinates of point C are (94.93,0).

[0044] S2: The particle swarm optimization algorithm SHPSO is improved by multi-strategy fusion to optimize the extreme learning machine and build an optimized extreme learning machine ELM model.

[0045] Specifically, constructing an optimized extreme learning machine ELM model includes the following steps: transforming a set of input weights and biases of the ELM into the SHPSO particle position x i , the mean square error MSE obtained through training is used as the fitness function of the SHPSO particle; the position of the SHPSO particle x i The relevant formula is as follows:

[0046] x i =[w 11 ,w 12,...,w 1n ,...,w k1 ,w k2 ,...,w kn ,b1,b2,...,b k ]

[0047] Among them, w kn is the weight between the kth hidden layer neural node and the nth input layer node, b k is the bias of the kth hidden layer node.

[0048] Furthermore, a part of the given sample set is randomly selected as the training set, and the rest is used as the test set; the parameters are determined, and the population size, maximum number of iterations, search space range, search space dimension, number of ELM hidden layer nodes and activation function of the SHPSO algorithm are set; the particle position x is initialized i and particle velocity v i ; Call the extreme learning machine ELM to calculate the fitness of each particle and initialize the individual optimal value; According to the fitness value of each particle, the particles are divided into elite particles, ordinary particles and inferior particles; Perform hierarchical learning on the divided particles and update the particle position x i , particle speed v i And particle fitness value. Determine whether the individual optimal value of the elite particle has been improved and update the individual optimal value; at the same time, determine whether the number of fitness evaluations has reached the maximum number of evaluations, and obtain the input layer weights and biases based on the optimal individuals to build an optimized extreme learning machine ELM model; use multi-strategy fusion to improve the particle swarm optimization algorithm SHPSO to find the optimal parameters of the extreme learning machine ELM; use the test set to verify the generalization ability of the optimized extreme learning machine ELM model.

[0049] Specifically, the fitness evaluation times include: when the fitness evaluation times reaches the maximum evaluation times, the personal optimal value is output, that is, the relevant formula for updating the personal optimal values ​​of all particles is as follows:

[0050]

[0051] Among them, p besti is the personal optimal value of particle i, is the fitness value of the t+1th iteration, is the fitness value at the tth iteration, f(p besti ) is the fitness value of the optimal value of particle i.

[0052] Furthermore, when the number of fitness evaluations does not reach the maximum number of evaluations, the particles are re-divided.

[0053] Specifically, the bias of the hidden layer nodes is shown in Table 1.

[0054] Table 1 Weight table of hidden layer nodes

[0055]

[0056] Furthermore, the weights from hidden layer nodes to output layer nodes are shown in Table 2.

[0057] Table 2 Bias table of hidden layer nodes

[0058] Hidden layer nodes Bias Hidden layer nodes Bias 1 0.687 6 0.487 2 0.128 7 0.026 3 0.001 8 0.999 4 1.000 9 0.334 5 0.149 10 0.305

[0059] Furthermore, the multi-strategy fusion improved particle swarm optimization algorithm SHPSO includes the following steps: using the inertia weight adaptive change strategy AW to update the particle speed and position, the update formula is as follows:

[0060]

[0061] Among them, u(i) is the inertia weight direction change coefficient, rand is a random number, is the fitness value of the t+1th iteration, is the fitness value at the tth iteration.

[0062] Specifically, according to the fitness value of the particles after each iteration, the direction of the inertia weight is adaptively adjusted to change the direction of particle movement; the particles are divided into dominant layer, common layer and inferior layer according to the particle fitness ranking, and an adaptive hierarchical learning strategy is adopted for the characteristics of particles in each layer; differential mutation is performed on the particle group type that is the dominant layer.

[0063] It should be noted that the random variables and functional functions are determined; six random factors X = (x1, ..., x6) that affect the deep anti-sliding of the gravity dam are considered, namely: bedrock I friction coefficient, bedrock I cohesion, structural surface AB friction coefficient, structural surface AB cohesion, structural surface BC friction coefficient and structural surface BC cohesion, the gravity dam profile is shown in Figure 2 The distribution characteristics of each factor’s value are shown in Table 1.

[0064] Table 3 Distribution characteristics of each factor value

[0065] Random Factors Mean Standard Deviation Distribution Type Bedrock I friction coefficient 1.1 0.22 normal distribution BedrockⅠcohesion / MPa 1.1 0.33 normal distribution Friction coefficient of structural surface AB 0.35 0.07 normal distribution Structural surface AB cohesion / MPa 0.05 0.015 Lognormal Structural surface BC friction coefficient 0.45 0.09 normal distribution Structural surface BC cohesion / MPa 0.1 0.03 Lognormal

[0066] Furthermore, the differential mutation includes: if the fitness value of the t+1th iteration is greater than the fitness value of the tth iteration, then it is judged that the fitness value of the elite particle i in the t+1th iteration has not improved, and the flight direction of the elite particle deviates from the actual personal optimal value. By changing the inertia weight direction coefficient to a negative random number, the elite particle flies in the opposite direction of the current movement in the next iteration; if the fitness value of the t+1th iteration is Less than or equal to the fitness value at the tth iteration Then the inertia weight directional coefficient is set to 1, and the inertia weight adaptive change strategy AW is used to encourage the elite particles to continue flying in the current speed direction.

[0067] Furthermore, the optimized extreme learning machine ELM model includes a data preprocessing module, a parameter optimization module and a prediction module; the data preprocessing module is used for data normalization to eliminate the influence of numerical magnitude; the parameter optimization module finds the optimal parameters of the extreme learning machine ELM by improving the particle swarm optimization algorithm SHPSO through multi-strategy fusion; the prediction module is used to input the required data into the optimized extreme learning machine ELM model and output the predicted value.

[0068] S3: Based on the optimized extreme learning machine ELM model, the optimal ELM model is constructed to calculate the deep sliding reliability of the gravity dam.

[0069] Specifically, the optimal ELM model includes the following steps: obtaining a sample set and normalizing it; inputting the normalized sample set into the extreme learning machine ELM, and taking the mean square error MSE as the fitness evaluation function of the extreme learning machine ELM, that is, the relevant formula of the fitness evaluation function is as follows:

[0070]

[0071] Where N is the number of samples, t i is the actual value corresponding to the i-th sample, is the simulated predicted value of the i-th sample.

[0072] Furthermore, based on the optimal parameters, the particle swarm optimization algorithm is improved by multi-strategy fusion to find the optimal weight and bias of ELM and construct the optimal ELM model; extract n s Set up random variable X = (x1, x2, ..., x6) samples, according to the variance, mean and distribution type of the random variable, use MATLAB distribution type functions to extract n s Group samples; input random variable samples into the optimized extreme learning machine ELM model to predict n s Group performance function value; Statistics SHPSO-ELM prediction results; The prediction results include failure rate and reliability, the specific formula is as follows:

[0073]

[0074] Among them, P f is the failure rate, n f is the total number of predicted results less than zero, n f is the total number of predicted results greater than zero, P r For reliability.

[0075] Furthermore, this embodiment also provides a deep sliding reliability calculation system for a gravity dam based on an optimized extreme learning machine, including: an acquisition module for acquiring basic information of the gravity dam; a construction module for optimizing the extreme learning machine by improving the particle swarm optimization algorithm SHPSO through multi-strategy fusion, and constructing an optimized extreme learning machine ELM model; a calculation module for constructing an optimal ELM model based on the optimized extreme learning machine ELM model, and calculating the deep sliding reliability of the gravity dam.

[0076] This embodiment also provides a computer device, which is suitable for optimizing the method for calculating the deep sliding reliability of a gravity dam using an extreme learning machine, and includes a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions to implement the method for calculating the deep sliding reliability of a gravity dam using an extreme learning machine as proposed in the above embodiment.

[0077] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0078] This embodiment also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the following steps: obtaining basic information of a gravity dam; optimizing an extreme learning machine by improving the particle swarm optimization algorithm SHPSO through multi-strategy fusion, and constructing an optimized extreme learning machine ELM model; constructing an optimal ELM model based on the optimal parameters of the optimized extreme learning machine ELM model, and calculating the deep sliding reliability of the gravity dam.

[0079] In summary, the present invention lays the foundation for calculation by acquiring the basic information of the gravity dam; takes the ELM parameters of the extreme learning machine as the optimization target, and uses the particle swarm optimization algorithm SHPSO improved by multi-strategy fusion to optimize the ELM parameters, so as to improve the prediction accuracy and generalization ability of the ELM model; constructs the optimal model according to the optimal parameters of the optimized ELM model, extracts random samples for input calculation, and can accurately obtain the deep sliding failure rate and reliability of the gravity dam, provide accurate data support for the anti-sliding stability analysis of the dam body, and make up for the shortcomings of the existing methods.

[0080] Example 2

[0081] Reference Figure 3 to Figure 5 , which is the second embodiment of the present invention, provides a method for calculating the reliability of deep sliding of a gravity dam based on an optimized extreme learning machine. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0082] Specifically, to verify the effectiveness of the present invention, the results of the optimized extreme learning machine ELM model before optimization and the optimal ELM model after optimization are compared. 200 sets of data are substituted into the optimized extreme learning machine ELM model for training. The input layer, hidden layer and output layer of the optimized extreme learning machine ELM model are all one layer, wherein the input layer has 6 neurons, the optimal number of hidden layer neurons is 10, and the output layer has 1 neuron. The sample set is divided into a training set and a test set, each accounting for 50% of the number of samples in the data set. The population number of the SHPSO algorithm is 40, the mutation probability is 0.2, the number of iterations is 1000, and the activation function is the Sigmoid function.

[0083] Furthermore, as shown in Table 4, the results comparison table of the optimal ELM model and the optimized extreme learning machine ELM model shows that the regression coefficient of the optimized extreme learning machine ELM model test set is larger, indicating that the optimized extreme learning machine ELM model has better generalization effect. It can be seen that the MSE and MAPE of the optimized extreme learning machine ELM model training set and test set are smaller than those of the ELM model.

[0084] Table 4 Results comparison table

[0085] Model MSE_train MAPE_train MSE_test MAPE_test R_test Optimizing the Extreme Learning Machine (ELM) model 6.84E-05 2.90E-03 2.30E-03 1.00-02 0.98352 Optimal ELM model 2.40E-03 1.80E-02 4.70E-03 2.44E-02 0.96567

[0086] Furthermore, if Figure 3 As shown in the figure, the error comparison between the predicted value and the true value of the ELM model test set before and after optimization is shown in the figure. Figure 4 Optimized extreme learning machine ELM model and the relationship between the true value and predicted value of the optimal ELM model test set. Figure 5 It can also be directly seen that the data point distribution of the optimized extreme learning machine ELM model is more concentrated on the Y=T straight line.

[0087] First determine the random variables and functional functions, according to Figure 4 It can be seen that the mean, standard deviation and distribution type of each factor are sampled, and 2 million groups of samples are extracted through MATLAB. Then, the 2 million groups of samples are substituted into the constructed optimized extreme learning machine ELM model, and 2 million groups of functional function values ​​are predicted. The predicted value less than zero means that the structure fails once. The number of structural failures is counted, and the failure probability is calculated to be 0.00072%, and the reliability index is 4.320. The failure probability obtained by the Monte Carlo method of the known LHS-BP proxy model is 0.00065%, and the reliability index is 4.360. The absolute error of the calculation results of the reliability index of the two methods is 0.040, and the relative error is 0.925%, indicating that the calculation results of the two methods are not much different. If the calculation result of the Monte Carlo method of the LHS-BP proxy model is taken as the exact solution, it means that the Monte Carlo method of the optimized extreme learning machine ELM model has high accuracy while improving the calculation efficiency. Deep sliding instability of gravity dams belongs to the second type of damage in the specification. The concrete gravity dam in the embodiment of the present invention is a secondary building, and the corresponding target reliability index is 3.7. The calculated reliability index is greater than the standard value, indicating that the gravity dam meets the deep anti-sliding stability requirements.

[0088] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for calculating the reliability of deep sliding of gravity dams based on an optimized extreme learning machine, characterized in that: include, Get basic information about gravity dams; Improve the particle swarm optimization algorithm SHPSO by integrating multiple strategies to optimize the extreme learning machine and build an optimized extreme learning machine ELM model; According to the optimal parameters of the ELM model of the extreme learning machine, the optimal ELM model is constructed to calculate the reliability of deep sliding of the gravity dam.

2. The method for calculating the deep sliding reliability of a gravity dam by optimizing an extreme learning machine according to claim 1, characterized in that: The process of constructing an optimized extreme learning machine ELM model includes the following steps: The set of input weights and biases of the extreme learning machine ELM is converted into the SHPSO particle position x i , the mean square error MSE obtained through training is used as the fitness function of the SHPSO particle, that is, the position of the SHPSO particle x i The relevant formula is as follows: x i =[w 11 ,w 12 ,...,w 1n ,...,w k1 ,w k2 ,...,w kn ,b1,b2,...,b k ] Among them, w kn is the weight between the kth hidden layer neural node and the nth input layer node, b k is the bias of the kth hidden layer node; Randomly select a part of the given sample set as the training set, and the rest as the test set; Determine the parameters, set the population size, maximum number of iterations, search space range, search space dimension, number of ELM hidden layer nodes and activation function of the SHPSO algorithm; Initialize particle position x i and particle velocity v i ; Call the extreme learning machine ELM to calculate the fitness of each particle and initialize the individual optimal value; According to the fitness value of each particle, the particles are divided into elite particles, ordinary particles and inferior particles; Perform hierarchical learning on the divided particles and update the particle position x i , particle speed v i and particle fitness values; Determine whether the personal optimal value of the elite particle has been improved and update the personal optimal value; At the same time, it is determined whether the number of fitness evaluations has reached the maximum number of evaluations, and the input layer weights and biases are obtained based on the optimal individual to construct an optimized extreme learning machine ELM model; The particle swarm optimization algorithm SHPSO is improved by multi-strategy fusion to find the optimal parameters of the extreme learning machine ELM; The test set is used to verify the generalization ability of the optimized extreme learning machine ELM model.

3. The method for calculating the deep sliding reliability of a gravity dam by optimizing an extreme learning machine according to claim 2, characterized in that: The fitness evaluation times include: When the number of fitness evaluations reaches the maximum number of evaluations, the personal optimal value is output, that is, the relevant formula for updating the personal optimal values ​​of all particles is as follows: Among them, p besti is the personal optimal value of particle i, is the fitness value of the t+1th iteration, is the fitness value at the tth iteration, f(p besti ) is the fitness value of the optimal value of particle i; When the number of fitness evaluations does not reach the maximum number of evaluations, the particles are divided again.

4. The method for calculating the deep sliding reliability of a gravity dam by optimizing an extreme learning machine according to claim 2, characterized in that: The optimized extreme learning machine ELM model includes a data preprocessing module, a parameter optimization module and a prediction module; the data preprocessing module is used for data normalization to eliminate the influence of numerical magnitude; the parameter optimization module finds the optimal parameters of the extreme learning machine ELM by improving the particle swarm optimization algorithm SHPSO through multi-strategy fusion; the prediction module is used to input the required data into the optimized extreme learning machine ELM model and output the predicted value.

5. The method for calculating the deep sliding reliability of a gravity dam by optimizing an extreme learning machine according to claim 4, characterized in that: The multi-strategy fusion improved particle swarm optimization algorithm SHPSO comprises the following steps: The inertia weight adaptive change strategy AW is used to update the particle speed and position. The update formula is as follows: Among them, u(i) is the inertia weight direction change coefficient, rand is a random number, is the fitness value of the t+1th iteration, is the fitness value at the tth iteration; According to the fitness value of the particle after each iteration, the direction of the inertia weight is adaptively adjusted to change the direction of particle movement; According to the particle fitness ranking, the particles are divided into the dominant layer, the common layer and the inferior layer, and an adaptive hierarchical learning strategy is adopted for the characteristics of particles in each layer; The particle group type is the dominant layer and differential mutation is performed.

6. The method for calculating the deep sliding reliability of a gravity dam by optimizing an extreme learning machine according to claim 5, characterized in that: The differential variation includes: If the fitness value of the t+1th iteration is greater than the fitness value of the tth iteration, it is judged that the fitness value of the elite particle i in the t+1th iteration has not improved, and the flight direction of the elite particle deviates from the actual personal optimal value. The inertia weight direction coefficient is changed by making it a negative random number, so that the elite particle flies in the opposite direction of the current movement in the next iteration; If the fitness value of the t+1th iteration Less than or equal to the fitness value at the tth iteration Then the inertia weight directional coefficient is set to 1, and the inertia weight adaptive change strategy AW is used to encourage the elite particles to continue flying in the current speed direction.

7. The method for calculating the deep sliding reliability of a gravity dam by optimizing an extreme learning machine according to claim 1, characterized in that: The construction process of the optimal ELM model includes the following steps: Get the sample set and normalize it; The normalized sample set is input into the extreme learning machine ELM, and the mean square error MSE is used as the fitness evaluation function of the extreme learning machine ELM. The relevant formula of the fitness evaluation function is as follows: Where N is the number of samples, t i is the actual value corresponding to the i-th sample, is the simulated predicted value of the i-th sample; Based on the optimal parameters, the particle swarm optimization algorithm is improved by multi-strategy fusion to find the optimal weight and bias of ELM and build the optimal ELM model; Extract n s Set up random variable X = (x1, x2, ..., x6) samples, according to the variance, mean and distribution type of the random variable, use MATLAB distribution type functions to extract n s Group sample; Input the random variable sample into the optimal ELM model and predict n s Group function value; Statistically calculate the prediction results of the optimal ELM model; The prediction results include failure rate and reliability, that is, the relevant specific formula is as follows: Among them, P f is the failure rate, n f is the total number of predicted results less than zero, n f is the total number of predicted results greater than zero, P r For reliability.

8. A gravity dam deep sliding reliability calculation system based on an optimized extreme learning machine, and a gravity dam deep sliding reliability calculation method based on an optimized extreme learning machine according to any one of claims 1 to 7, characterized in that: include, Acquisition module, used to obtain basic information of gravity dam; A construction module is used to improve the particle swarm optimization algorithm SHPSO by multi-strategy fusion to optimize the extreme learning machine and build an optimized extreme learning machine ELM model; The calculation module is used to construct the optimal ELM model according to the optimized extreme learning machine ELM model and calculate the deep sliding reliability of the gravity dam.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for calculating the deep sliding reliability of a gravity dam using an optimized extreme learning machine as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for calculating the deep sliding reliability of a gravity dam by optimizing an extreme learning machine according to any one of claims 1 to 7 are implemented.

Citation Information

Patent Citations

  • Concrete dam deformation combined forecasting model construction method based on ARIMA and PSO-ELM

    CN112100711A

  • Gravity dam shallow anti-sliding stability influence factor sensitivity analysis method

    CN114492127A

  • Mining subsidence-oriented rock mass mechanical parameter inversion method

    CN116680990A