A tunnel construction parameter optimization method, device, equipment and readable storage medium
Through the combination of mutation theory and PSO-MSVR model, the tunnel construction parameters are optimized, and the problems of surrounding rock stability assessment and construction parameters are solved, and the construction progress and material use are accelerated while ensuring surrounding rock stability.
Patent Information
- Application Number
- CN202210495193.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-05-07
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2042-05-07
AI Technical Summary
In tunnel engineering, there are difficulties in assessing and predicting the stability of surrounding rocks, and how to obtain the optimal economical construction plan while ensuring safety is a challenge.
Using mutation theory combined with plastic yield zone volume, a constrained multi-objective planning function is constructed, and the tunnel construction parameters are optimized through particle swarm optimization algorithm and support vector machine regression model (PSO-MSVR) to ensure surrounding rock stability and materials Use optimization.
An intelligent system is achieved that while ensuring the stability of surrounding rock, speeding up construction progress and reducing construction materials required for construction, effectively solving the problems of surrounding rock stability assessment and construction parameter optimization.
Smart Images

Figure CN114969900B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of highway tunnels, and in particular to a method, device, equipment and readable storage medium for optimizing tunnel construction parameters. Background Art
[0002] my country is a country with a large number of tunnel construction projects. With the development of the times, the scale and difficulty of tunnel projects have gradually increased, and the most important problem of surrounding rock stability in tunnel construction has become increasingly prominent. The instability of surrounding rock in tunnel projects often leads to significant losses of life and property, so it is extremely important to conduct effective risk assessment and prediction of tunnel surrounding rock stability issues; how to obtain the most economical solution while ensuring safety during design is the current difficulty.
[0003] At present, the relevant theories and testing technologies of rock mechanics are developing rapidly. Electronic computers and finite element theory have also been applied to the study of the instability of surrounding rock in underground engineering, and many good results have been achieved. Using the mutation theory, it is believed that there is a continuous exchange of matter and energy between the surrounding rock of underground engineering and the external environment, and it is an open system in constant change. Based on this understanding, the mutation theory is introduced into the judgment of the stability of the surrounding rock. In order to judge the overall surrounding rock, the present invention introduces the volume of the plastic yield zone into the mutation theory to form a criterion for the stability of the surrounding rock. Summary of the invention
[0004] The purpose of the present invention is to provide a tunnel construction parameter optimization method, device, equipment and readable storage medium to improve the above problems. In order to achieve the above purpose, the technical solution adopted by the present invention is as follows:
[0005] In a first aspect, the present application provides a method for optimizing tunnel construction parameters, comprising:
[0006] Acquire first data, wherein the first data includes a volume of a plastic yield zone of a tunnel surrounding rock obtained by sampling a tunnel scene based on a catastrophe theory method;
[0007] Constructing a constrained multi-objective programming function according to the first data;
[0008] According to the constrained multi-objective programming function, a MSVR model corresponding to the constrained multi-objective programming function is calculated;
[0009] Optimizing the parameters of the MSVR model by using a particle swarm optimization algorithm to obtain an optimal MSVR model;
[0010] The particle swarm optimization algorithm is used to generate tunnel construction parameters, and the optimal MSVR model is used for calculation, and finally the optimal tunnel construction parameters are output.
[0011] In a second aspect, the present application also provides a tunnel construction parameter optimization device, including an acquisition module, a construction module, a calculation module, a processing module and a generation module, wherein:
[0012] An acquisition module: used to acquire first data, wherein the first data includes a volume of a plastic yield zone of a tunnel surrounding rock obtained by sampling a tunnel scene based on a mutation theory method;
[0013] A construction module: used for constructing a constrained multi-objective planning function according to the first data;
[0014] A calculation module: used for calculating, according to the constrained multi-objective programming function, an MSVR model corresponding to the constrained multi-objective programming function;
[0015] Processing module: used for performing parameter optimization processing on the MSVR model through a particle swarm optimization algorithm to obtain an optimal MSVR model;
[0016] Generation module: used to generate tunnel construction parameters through the particle swarm optimization algorithm, use the optimal MSVR model for calculation, and finally output the optimal tunnel construction parameters.
[0017] In a third aspect, the present application also provides a tunnel construction parameter optimization device, including:
[0018] Memory for storing computer programs;
[0019] A processor is used to implement the steps of the tunnel construction parameter optimization method when executing the computer program.
[0020] In a fourth aspect, the present application further provides a readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the above-mentioned method based on tunnel construction parameter optimization are implemented.
[0021] The beneficial effects of the present invention are as follows: by combining the SVR algorithm with the particle swarm optimization algorithm, which is a bionic optimization algorithm, the PSO-MSVR model can be trained with a small number of samples, so that effective parameter optimization prediction can be performed; and through a small amount of sample learning, an intelligent system combining the mutation model and PSO-MSVR can be obtained to obtain the optimal tunnel construction parameters, so that the construction can be accelerated to the greatest extent while ensuring the stability of the surrounding rock and reducing the building materials required for construction.
[0022] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by implementing the embodiments of the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.
[0024] Figure 1 It is a schematic diagram of the flow of the tunnel construction parameter optimization method described in an embodiment of the present invention;
[0025] Figure 2 It is a schematic diagram of the structure of the tunnel construction parameter optimization device described in an embodiment of the present invention;
[0026] Figure 3 It is a schematic diagram of the structure of the tunnel construction parameter optimization equipment described in an embodiment of the present invention.
[0027] In the figure: 701, acquisition module; 7011, first calculation unit; 7012, first establishment unit; 7013, second calculation unit; 7014, transformation unit; 7015, judgment unit; 702, construction module; 7021, second establishment unit; 7022, third calculation unit; 7023, fourth calculation unit; 703, calculation module; 704, processing module; 7041, processing unit; 7042, prediction unit; 7043, evaluation unit; 7044, update unit; 70441, optimization unit; 70442, prediction sample unit; 70443, identification value unit; 70444, comparison unit; 7045, loop unit; 705, generation module; 7051, generation particle unit; 7052, fifth calculation unit; 7053, setting unit; 7054, sixth calculation unit; 7055, update particle unit; 7056, acquisition unit; 800, tunnel construction parameter optimization equipment; 801, processor; 802, memory; 803, multimedia component; 804, input / output (I / O) interface; 805, communication component. DETAILED DESCRIPTION
[0028] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.
[0029] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.
[0030] Embodiment 1:
[0031] The support vector machine algorithm originates from statistical learning theory. It is a small sample learning machine based on the principle of structural risk minimization. It has the advantages of good generalization performance and global optimization, so it can be applied to geotechnical engineering. Combining the SVR algorithm with the bionic optimization algorithm of particle swarm optimization (PSO) algorithm can realize the training of PSO-MSVR model with a small number of samples, so that it can perform effective parameter optimization prediction.
[0032] This embodiment provides a method for optimizing tunnel construction parameters.
[0033] See also Figure 1 , the figure shows that the method includes step S100, step S200, step S300, step S400 and step S500.
[0034] S100. Acquire first data, where the first data includes a volume of a plastic yield zone of a tunnel surrounding rock obtained by sampling a tunnel scene based on a mutation theory method.
[0035] It can be understood that step S100 includes steps S101, S102, S103, S104, and S105, wherein:
[0036] S101, calculating the mutation characteristic value by calculating the potential function of the cusp mutation model;
[0037] S102, establishing a volume increment function of the plastic zone of the tunnel according to the mutation characteristic value;
[0038] S103, expanding the volume increment function to the fourth power according to Taylor's formula to obtain a replaced volume increment function;
[0039] S104, according to the Chirnhausen transformation, obtaining a volume increment function of the plastic zone of the tunnel after the transformation which is consistent with the potential function form;
[0040] S105. Calculate the mutation characteristic value according to the transformed volume increment function of the tunnel plastic zone to determine whether the tunnel surrounding rock is stable.
[0041] It should be noted that the establishment of the surrounding rock stability criterion based on the cusp catastrophe theory refers to:
[0042] (1) First, the calculation formula of the mutation characteristic value is derived through the potential function expression of the cusp mutation model.
[0043] (2) Establish a volume increment function of the plastic zone, perform Taylor expansion on it to the fourth power, and finally perform Tschirnhaus transformation to make the volume increment function of the plastic zone have the same form as the potential function of the cusp catastrophe model.
[0044] (3) After obtaining the volume increment of the plastic zone at each step in the construction, it can be fitted into a quartic function, and then the mutation characteristic value △ is calculated by formulas (1) and (2). When △>0, the surrounding rock is judged to be stable; when △=0, the surrounding rock is judged to be in a critical equilibrium state; when △<0, the surrounding rock is judged to be unstable and damaged. The specific formula is derived as follows:
[0045] The criterion for surrounding rock stability in the present invention uses a cusp catastrophe model, and its potential function expression is as shown in formula (1):
[0046] V(x)=x 4 +ux 2 +vx (1)
[0047] In the formula, u and v are control variables, and x is the state variable of the potential function. Taking the first-order and second-order derivatives of x in the above formula, we can get the result as shown in formula (2):
[0048] 8u 3 +27v 2 =0 (2)
[0049] △=8u 3 +27v 2 It is the mutation characteristic value.
[0050] In the combination of the plastic zone volume increment and the cusp mutation model, the volume growth x of the surrounding rock plastic zone at each excavation step during the construction process is defined as: when the excavation step reaches x, the volume increment f(x) of the plastic zone is:
[0051]
[0052] In the formula, a i is the independent variable x i The coefficient of f(x) is a continuous function used to represent the change of volume increment in the plastic zone. The volume function f(x) is expanded by Taylor series and raised to the fourth power to obtain:
[0053] f(x)=a0+a1x+a2x 2 +a3x 3 +a4x 4 (4)
[0054] For equation (4), through Tschirnhaus transformation, let get
[0055] P(t)=b0+b1t+b2t 2 +b4t 4 (5)
[0056] In the formula (5), t is the variable after substitution, b0 is a constant, which can be ignored as an intercept in three-dimensional space, and the standard form can be further obtained:
[0057] Q=t 4 +ut 2 +vt (6)
[0058] Where:
[0059]
[0060] S200. Construct a constrained multi-objective programming function according to the first data.
[0061] It can be understood that step S200 includes step S201, S202 and S203, wherein:
[0062] S201, establishing a tunnel support optimization model according to the collected concrete volume and steel bar usage of the tunnel;
[0063] S202, calculating the tunnel support optimization model according to the Lagrange multiplier method to obtain the objective function;
[0064] S203, performing weighted summation of concrete volume and steel bar usage based on the objective function, and calculating a constrained multi-objective programming function with the goal of minimizing concrete volume and steel bar usage.
[0065] It should be noted that the optimal construction of the tunnel needs to satisfy both the stability of the surrounding rock and the minimum use of materials. Therefore, the present invention uses the mutation characteristic value △>0 as a constraint condition to ensure the stability of the surrounding rock, and constructs a constrained multi-objective programming function with the goal of minimizing the amount of steel and concrete. The function uses a comprehensive material value (the weighted sum of the amount of steel and concrete) to represent the amount of consumed materials. The specific construction process is as follows:
[0066] The support optimization mathematical model of the chamber can be expressed as:
[0067] f=Min(v1,v2) (8)
[0068] stv3=△>0 (9)
[0069] Where Min means the objective is minimization; v1, v2, and v3 represent the volume of concrete, the amount of steel bars, and the characteristic value of the volume entropy mutation in the plastic zone of the surrounding rock system, respectively; st represents the constraint condition, and △ represents the characteristic value of the entropy mutation.
[0070] Using the Lagrange multiplier method, combining equations (8) and (9), the objective function can be specifically formulated as:
[0071] f=Min[(v1,v2)+λ△]=Min[(v1,v2)+C] (10)
[0072] In the formula, λ is the Lagrange multiplier, and since Δ is a numerical value, it is a constant. Therefore, λ△ in formula (10) can be expressed as a constant C - penalty term.
[0073] v1 and v2 can be calculated from multiple objectives to a single objective through weighted summation, as follows:
[0074] f=Min[(w1·v1+w2·v2)+λ△] (11)
[0075] In the formula, w1 and w2 are the weight coefficients of v1 and v2 respectively, and:
[0076] w1+w2=1 (12)
[0077] S300 , according to the constrained multi-objective programming function, calculate and obtain an MSVR model corresponding to the constrained multi-objective programming function.
[0078] It is understandable that in this step, there are several steps as follows:
[0079] (1) The input parameter X of the learning sample is used as the training input of the network, and the sample value y1 of the first output parameter is used as the training output of the network. After machine learning, the input variable X′ of the test sample is predicted to obtain the first output parameter y1′ of the test sample, where j=1;
[0080] (2) Determine whether j reaches the number of regression output parameters n. If so, output the test sample y1′. Otherwise, the calculation goes to step (3), and j=j+1.
[0081] (3) Using X, y1 as the training input of the network and the sample value y2 of the second output parameter as the training output of the network, after machine learning, the input variables X′, y1′ of the test sample are predicted to obtain the second output parameter y2′ of the test sample;
[0082] (4) X, y1 and y2 are used as the training input of the network, and the sample value y3 of the third output parameter is used as the training output of the network. After machine learning, the input variables X′, y1′ and y2′ of the test sample are predicted to obtain the third output parameter y3′ of the test sample;
[0083] Repeat the above steps in the same way until the condition of step (2) is met;
[0084] The SVR algorithm is used in the selection of the SVR algorithm, and the kernel function uses the radial basis function kernel function. The specific contents are as follows:
[0085] Suppose the dimension of a sample space is n, and there are k samples in a certain area of this space, namely (x1,y1),(x2,y2),…,(x k ,y k )∈R n ×R, the linear function is
[0086] f(x)=w*x+b (13)
[0087] Where b is the random error and w is the weight vector.
[0088] Minimizes to:
[0089]
[0090] In the formula, ξ i and is the relaxation factor, and C is the compromise coefficient.
[0091] The constraints are
[0092]
[0093] Formula, where ε is the insensitive loss coefficient. For nonlinear regression, the optimized dual function is:
[0094]
[0095] In the formula, α, α * is the Lagrange multiplier, K(x i ,x j ) is the kernel function, and the radial basis function kernel function is used, which is in the following form:
[0096]
[0097] Where x is a point in space, y is a certain center, and σ is the width parameter of the function.
[0098] S400, optimizing the parameters of the MSVR model by using a particle swarm optimization algorithm to obtain an optimal MSVR model.
[0099] It can be understood that step S400 includes S401, S402, S403, S404 and S405, wherein:
[0100] S401, initializing tunnel parameters according to a particle swarm optimization algorithm;
[0101] S402, based on the result of initializing the tunnel parameters, learning the tunnel sample data in the MSVR model and performing tunnel sample prediction, the initialization result includes: determining the particle swarm size, the number of iterations and the particle random solution, each particle is an MSVR model, and an MSVR model includes an insensitive loss coefficient, a penalty coefficient and a width coefficient of the RBF kernel;
[0102] S403, evaluating the fitness of particles in the particle swarm optimization algorithm according to the tunnel sample prediction results;
[0103] S404, according to the calculation result of the fitness of the particle, the speed and position of the particle are updated, and a new MSVR model is formed;
[0104] S405. According to the iterative algorithm, the MSVR model is judged in a loop until the prediction error of the MSVR model is less than a specific value or reaches the maximum iteration set in the particle swarm optimization algorithm. Then, the iteration of the MSVR model is completed, the calculation is stopped, and the MSVR parameters at this time are output to obtain the optimal MSVR model.
[0105] It should be noted that S404 includes S4041, S4042, S4043 and S4044, among which:
[0106] S4041. Introducing a particle swarm optimization algorithm on the preset MSVR model to optimize the penalty factor c and the kernel parameter g, and optimizing to obtain a PSO-MSVR model;
[0107] S4042, using the MSVR model corresponding to each particle to predict the tunnel data sample, and calculating the current position value and prediction error of each particle;
[0108] S4043, taking the current position value and prediction error of each particle as the fitness value of the particle;
[0109] S4044. Compare the fitness value of the current position of the particle with its own fitness value, determine the optimal value as the current optimal position of the particle, and continuously update the latest position and speed of the particle.
[0110] It should be noted that the PSO algorithm is introduced to obtain the optimal parameters of MSVR. In simple terms, the PSO algorithm first generates a group of particles, which are random, and then continuously updates (iterates) these particles to finally find the optimal solution.
[0111] In each iteration, the particle will "evolve" itself by choosing one of two data, namely its own optimal solution and the global optimal solution, denoted as pbesti and gbesti; at this time, the particle will update its position and speed according to the latest solution obtained.
[0112] Based on this principle, the steps to use PSO to obtain the optimal MSVR model are:
[0113] (1) Initialization of PSO network parameters, including particle swarm size, number of iterations, particle random solution - particle initial velocity and particle initial position. Each particle vector represents an MSVR model, which contains three parameters of MSVR - insensitive loss coefficient ε, penalty coefficient C and RBF kernel width coefficient σ.
[0114] (2) After the PSO algorithm generates the initial MSVR model, it starts to learn the samples and make sample predictions. The fitness of the PSO particles can be evaluated based on the prediction results.
[0115] (3) Based on the calculated results of fitness, the particles in the PSO algorithm will update their speed and position to form a new MSVR model. In this way, the PSO algorithm completes a generation of “evolution”.
[0116] (4) Repeat steps (1) to (3). When the prediction error of the MSVR network is less than a certain value or reaches the maximum iteration step set by the PSO algorithm, the optimal MSVR network can be output. This network can predict the stability of tunnel construction and the comprehensive material value.
[0117] The specific expression is as follows:
[0118] The particle updates its velocity according to equation (18) and its position according to equation (19) based on the latest solution obtained.
[0119]
[0120]
[0121] In the formula, v is the velocity of the particle; t1, t2 are learning factors, usually t1=t2=2; rand() is a random number between (0,1); pbest is the volume extreme value; gbest is the global extreme value; P is the current position of the particle.
[0122] The form of the fitness function is shown in formula (20):
[0123]
[0124] In the formula, u ij Represents the sample value of the jth component (plastic zone volume entropy mutation characteristic value △ or comprehensive material) in the i-th test sample during the training process; u′ ij Represents the MSVR network prediction value of the jth component (plastic zone volume entropy mutation characteristic value △ or comprehensive material) in the ith test sample.
[0125] S500, generating tunnel construction parameters through particle swarm optimization algorithm, using the optimal MSVR model for calculation, and finally outputting the optimal tunnel construction parameters.
[0126] It can be understood that, in step S500, it includes S501, S502, S503, S504, S505 and S506, wherein:
[0127] S501, randomly generating a particle swarm according to a particle swarm optimization algorithm, each particle including a construction parameter of tunnel construction;
[0128] S502, calculating according to the generated particle swarm and the characteristic values of the preset MSVR model to construct a fitness function;
[0129] S503, according to the fitness function, set the population size, maximum genetic generation and algorithm termination condition in the function;
[0130] S504, input the MSVR prediction model into the set fitness function, and use the particle swarm optimization algorithm to generate construction parameters for training, and calculate individual extreme values and global optimal solutions;
[0131] S505, updating the speed and position of each particle according to the individual extreme value and the global optimal solution;
[0132] S506. Taking the maximum genetic generation as the termination condition of whether the requirement is met, it is determined whether the maximum genetic generation satisfies the condition. If not, the algorithm continues to execute until the algorithm termination condition is met, and the optimal mutation characteristic value parameters and comprehensive material value parameters are obtained.
[0133] It should be noted that after obtaining the optimal MSVR network, the optimal construction parameters are calculated in the following steps:
[0134] (1) The PSO network parameters randomly generate a particle swarm, and each particle contains the construction parameters of the tunnel construction.
[0135] (2) The particles generated by the PSO algorithm are calculated through the optimal MSVR model to obtain the corresponding mutation characteristic value Δ and comprehensive material value.
[0136] (3) The calculation results of the optimal MSVR model can be used to calculate the fitness. The particles in the PSO algorithm will update the speed and position to form new construction parameter data. In this way, the PSO algorithm completes a generation of "evolution".
[0137] (4) Repeat steps (1) to (3). When the construction parameters predicted by the optimal MSVR network ensure that Δ>0 and the comprehensive material value is the smallest, the optimal construction parameters can be output.
[0138] The constraint condition at this time uses the mutation characteristic value Δ>0, and the fitness function is:
[0139] f′=f+C (21)
[0140] Where f and f' represent the PSO fitness function values before and after penalty respectively; C is the penalty parameter.
[0141] Embodiment 2:
[0142] like Figure 2 As shown, this embodiment provides a tunnel construction parameter optimization device, see Figure 2 The device includes an acquisition module 701, a construction module 702, a calculation module 703, a processing module 704 and a generation module 705, wherein:
[0143] Acquisition module 701: used to acquire first data, the first data including the volume of the plastic yield zone of the tunnel surrounding rock obtained by sampling the tunnel scene based on the catastrophe theory method;
[0144] Construction module 702: used to construct a constrained multi-objective programming function according to the first data;
[0145] Calculation module 703: used to calculate and obtain the MSVR model corresponding to the constrained multi-objective programming function according to the constrained multi-objective programming function;
[0146] Processing module 704: used to optimize the parameters of the MSVR model by using a particle swarm optimization algorithm to obtain an optimal MSVR model;
[0147] The generation module 705 is used to generate tunnel construction parameters by using a particle swarm optimization algorithm, use the optimal MSVR model for calculation, and finally output the optimal tunnel construction parameters.
[0148] Preferably, the acquisition module 701 includes a first calculation unit 7011, a first establishment unit 7012, a second calculation unit 7013, a transformation unit 7014 and a judgment unit 7015, wherein:
[0149] The first calculation unit 7011 is used to calculate the mutation characteristic value by calculating the potential function of the cusp mutation model;
[0150] The first establishing unit 7012 is used to establish a volume increment function of the plastic zone of the tunnel according to the mutation characteristic value;
[0151] The second calculation unit 7013 is used to expand the volume increment function to the fourth power according to Taylor's formula to obtain a replaced volume increment function;
[0152] Transformation unit 7014: used for obtaining a volume increment function of the plastic zone of the tunnel after transformation consistent with the potential function form according to the Kilnhausen transformation;
[0153] The judgment unit 7015 is used to calculate the mutation characteristic value according to the volume increment function of the transformed tunnel plastic zone to judge whether the tunnel surrounding rock is stable.
[0154] Preferably, the construction module 702 includes a second establishment unit 7021, a third calculation unit 7022 and a fourth calculation unit 7023, wherein:
[0155] The second establishing unit 7021 is used to establish a tunnel support optimization model according to the collected concrete volume and steel bar usage of the tunnel;
[0156] The third calculation unit 7022 is used to calculate the tunnel support optimization model according to the Lagrange multiplier method to obtain the objective function;
[0157] The fourth calculation unit 7023 is used to perform weighted summation of the concrete volume and the steel bar usage based on the objective function, and calculate a constrained multi-objective programming function with the goal of minimizing the concrete volume and the steel bar usage.
[0158] Preferably, the processing module 704 includes a processing unit 7041, a prediction unit 7042, an evaluation unit 7043, an update unit 7044 and a loop unit 7045, wherein:
[0159] Processing unit 7041: used to initialize tunnel parameters according to a particle swarm optimization algorithm;
[0160] Prediction unit 7042: used to learn the tunnel sample data in the MSVR model and perform tunnel sample prediction based on the result of initializing the tunnel parameters. The initialization result includes: determining the particle swarm size, the number of iterations and the particle random solution. Each particle is an MSVR model. An MSVR model includes an insensitive loss coefficient, a penalty coefficient and a width coefficient of the RBF kernel.
[0161] Evaluation unit 7043: used to evaluate the fitness of particles in the particle swarm optimization algorithm according to the tunnel sample prediction results;
[0162] Update unit 7044: used to update the speed and position of the particle according to the calculation result of the fitness of the particle, and form a new MSVR model;
[0163] Loop unit 7045: used to perform a judgment loop on the MSVR model according to the iterative algorithm until the prediction error of the MSVR model is less than a specific value or reaches the maximum iteration set in the particle swarm optimization algorithm, completes the iteration of the MSVR model, stops the calculation, outputs the MSVR parameters at this time, and obtains the optimal MSVR model.
[0164] Preferably, the updating unit 7044 includes an optimizing unit 70441, a predicting sample unit 70442, an identifying value unit 70443 and a comparing unit 70444, wherein:
[0165] Optimization unit 70441: used to introduce a particle swarm optimization algorithm on the preset MSVR model to optimize the penalty factor c and the kernel parameter g, and optimize to obtain a PSO-MSVR model;
[0166] Prediction sample unit 70442: used to predict the tunnel data sample using the MSVR model corresponding to each particle, and calculate the current position value and prediction error of each particle;
[0167] Identification value unit 70443: used to use the current position value and prediction error of each particle as the fitness value of the particle;
[0168] Comparison unit 70444: used to compare the fitness value of the current position of the particle with its own fitness value, determine the optimal value as the current optimal position of the particle, and continuously update the latest position and speed of the particle.
[0169] Preferably, the generation module 705 comprises: a generation particle unit 7051, a fifth calculation unit 7052, a setting unit 7053, a sixth calculation unit 7054, an update particle unit 7055 and an acquisition unit 7056, wherein:
[0170] Generate particle unit 7051: used to randomly generate particle swarms according to the particle swarm optimization algorithm, and each particle contains the construction parameters of the tunnel construction;
[0171] The fifth calculation unit 7052 is used to perform calculations according to the generated particle swarm and the characteristic values of the preset MSVR model to construct a fitness function;
[0172] Setting unit 7053: according to the fitness function, setting the population size, maximum genetic generation and algorithm termination condition in the function;
[0173] The sixth calculation unit 7054 is used to input the MSVR prediction model into the set fitness function, and use the particle swarm optimization algorithm to generate construction parameters for training, and calculate individual extreme values and global optimal solutions;
[0174] Update particle unit 7055: update the speed and position of each particle according to the individual extreme value and the global optimal solution;
[0175] Acquisition unit 7056: used to use the maximum genetic generation as the termination condition of whether the requirement is met, and judge whether the maximum genetic generation meets the condition. If not, the algorithm continues to execute until the algorithm termination condition is met, and the optimal mutation characteristic value parameters and comprehensive material value parameters are obtained.
[0176] It should be noted that, regarding the device in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment of the method, and will not be elaborated here.
[0177] Embodiment 3:
[0178] Corresponding to the above method embodiment, this embodiment also provides a tunnel construction parameter optimization device. The tunnel construction parameter optimization device described below and the tunnel construction parameter optimization method described above can refer to each other.
[0179] Figure 3 FIG. 8 is a block diagram of a tunnel construction parameter optimization device 800 according to an exemplary embodiment. Figure 3 As shown, the tunnel construction parameter optimization device 800 may include: a processor 801 and a memory 802. The tunnel construction parameter optimization device 800 may also include one or more of a multimedia component 803, an I / O interface 804, and a communication component 805.
[0180] The processor 801 is used to control the overall operation of the tunnel construction parameter optimization device 800 to complete all or part of the steps in the above-mentioned tunnel construction parameter optimization method. The memory 802 is used to store various types of data to support the operation of the tunnel construction parameter optimization device 800, and these data may include, for example, instructions for any application or method operating on the tunnel construction parameter optimization device 800, and application-related data, such as contact data, sent and received messages, pictures, audio, video, etc. The memory 802 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk. The multimedia component 803 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may be further stored in the memory 802 or sent via the communication component 805. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 804 provides an interface between the processor 801 and other interface modules, and the above-mentioned other interface modules can be keyboards, mice, buttons, etc. These buttons can be virtual buttons or physical buttons. The communication component 805 is used for wired or wireless communication between the tunnel construction parameter optimization device 800 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 805 can include: Wi-Fi module, Bluetooth module, NFC module.
[0181] In an exemplary embodiment, the tunnel construction parameter optimization device 800 can be implemented by one or more application specific integrated circuits (Application Specific Integrated Circuit, referred to as ASIC), digital signal processors (Digital Signal Processor, referred to as DSP), digital signal processing devices (Digital Signal Processing Device, referred to as DSPD), programmable logic devices (Programmable Logic Device, referred to as PLD), field programmable gate arrays (Field Programmable Gate Array, referred to as FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned tunnel construction parameter optimization method.
[0182] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned tunnel construction parameter optimization method are implemented. For example, the computer-readable storage medium can be the above-mentioned memory 802 including program instructions, and the above-mentioned program instructions can be executed by the processor 801 of the tunnel construction parameter optimization device 800 to complete the above-mentioned tunnel construction parameter optimization method.
[0183] Embodiment 4:
[0184] Corresponding to the above method embodiment, a readable storage medium is also provided in this embodiment. The readable storage medium described below and the tunnel construction parameter optimization method described above can refer to each other.
[0185] A readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the tunnel construction parameter optimization method of the above method embodiment.
[0186] The readable storage medium may specifically be a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or other readable storage medium that can store program codes.
[0187] In summary, the present invention discloses a tunnel construction parameter optimization method combining a catastrophe model with PSO and MSVR algorithms, and the method includes the following features:
[0188] 1. Through the cusp mutation theory, the mutation characteristic value is calculated using the volume of the plastic yield zone of the surrounding rock during cavern construction as a judgment criterion.
[0189] 2. A constrained multi-objective planning mathematical model for chamber construction parameter optimization was constructed, with the aim of obtaining the optimal construction parameters while ensuring the stability of the surrounding rock.
[0190] 3. Combine the PSO algorithm and the MSVR algorithm to obtain the optimal MSVR model first, and then use it to calculate the optimal construction parameters of the tunnel. The MSVR algorithm can solve the support vector regression problem of multi-dimensional output variables, and solves the limitation that the output variable of the standard SVR algorithm can only be a one-dimensional variable.
[0191] The present invention can obtain an intelligent system that combines a mutation model and PSO-MSVR to obtain optimal tunnel construction parameters through a small amount of sample learning, so that the construction progress can be accelerated to the greatest extent while ensuring the stability of the surrounding rock and reducing the building materials required for construction.
[0192] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
[0193] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
Claims
1. A method for optimizing tunnel construction parameters, characterized in that: include: Acquire first data, wherein the first data includes a volume of a plastic yield zone of a tunnel surrounding rock obtained by sampling a tunnel scene based on a catastrophe theory method; Constructing a constrained multi-objective programming function according to the first data; According to the constrained multi-objective programming function, a MSVR model corresponding to the constrained multi-objective programming function is calculated; Optimizing the parameters of the MSVR model by using a particle swarm optimization algorithm to obtain an optimal MSVR model; Generate tunnel construction parameters by using the particle swarm optimization algorithm, use the optimal MSVR model for calculation, and finally output the optimal tunnel construction parameters; The first data is obtained, and the first data includes the volume of the plastic yield zone of the tunnel surrounding rock obtained by sampling the tunnel scene based on the mutation theory method, including: The mutation characteristic value is obtained by calculating the potential function of the cusp mutation model; according to the mutation characteristic value, the volume increment function of the tunnel plastic zone is established; Expand the volume increment function to the fourth power according to Taylor's formula to obtain a replaced volume increment function; According to the Chirnhausen transformation, the volume increment function of the transformed tunnel plastic zone consistent with the potential function form is obtained; Calculate the mutation characteristic value according to the volume increment function of the transformed tunnel plastic zone to determine whether the tunnel surrounding rock is stable; The step of constructing a constrained multi-objective programming function according to the first data includes: According to the collected concrete volume and steel bar usage of the tunnel, a tunnel support optimization model is established; According to the Lagrange multiplier method, the tunnel support optimization model is calculated to obtain the objective function; Based on the objective function, a weighted sum is taken for the concrete volume and the steel bar usage, and a constrained multi-objective programming function is calculated with the goal of minimizing the concrete volume and the steel bar usage; The method of performing parameter optimization processing on the MSVR model by using a particle swarm optimization algorithm includes: Initializing tunnel parameters according to the particle swarm optimization algorithm; Based on the result of initializing the tunnel parameters, the tunnel sample data in the MSVR model is learned and tunnel sample prediction is performed, wherein the result of initialization includes: determining the particle swarm size, the number of iterations and the particle random solution, each particle is one of the MSVR models, and one of the MSVR models includes an insensitive loss coefficient, a penalty coefficient and a width coefficient of the RBF kernel; According to the prediction results of the tunnel samples, the fitness of the particles in the particle swarm optimization algorithm is evaluated; According to the calculation result of the fitness of the particle, the speed and position of the particle are updated, and a new MSVR model is formed; According to the iterative algorithm, the MSVR model is judged in a loop until the prediction error of the MSVR model is less than a specific value or reaches the maximum iteration set in the particle swarm optimization algorithm, the iteration of the MSVR model is completed, the calculation is stopped, the MSVR parameters at this time are output, and the optimal MSVR model is obtained.
2. A tunnel construction parameter optimization device, characterized in that: include: Acquisition module: used to acquire first data, wherein the first data includes a tunnel scene based on a mutation theory method. The volume of the plastic yield zone of the tunnel surrounding rock obtained by sampling; A construction module: used for constructing a constrained multi-objective planning function according to the first data; A calculation module: used for calculating, according to the constrained multi-objective programming function, an MSVR model corresponding to the constrained multi-objective programming function; Processing module: used for performing parameter optimization processing on the MSVR model through a particle swarm optimization algorithm to obtain an optimal MSVR model; Generation module: used to generate tunnel construction parameters through the particle swarm optimization algorithm, use the optimal MSVR model for calculation, and finally output the optimal tunnel construction parameters; Wherein, the acquisition module includes: The first calculation unit is used to calculate the mutation characteristic value by calculating the potential function of the cusp mutation model; Establishment unit: used to establish the volume increment function of the tunnel plastic zone according to the mutation characteristic value; A second calculation unit is used for performing a fourth-power expansion on the volume increment function according to Taylor's formula to obtain a replaced volume increment function; Transformation unit: used for obtaining the volume increment function of the transformed tunnel plastic zone consistent with the potential function form according to the Kilnhausen transformation; A judgment unit: used to calculate a mutation characteristic value according to the volume increment function of the transformed tunnel plastic zone to judge whether the tunnel surrounding rock is stable; Wherein, the building blocks include: Establishment unit: used to establish a tunnel support optimization model based on the collected concrete volume and steel bar usage of the tunnel; The third calculation unit is used to calculate the tunnel support optimization model according to the Lagrange multiplier method to obtain the objective function; A fourth calculation unit is used for performing a weighted summation of the concrete volume and the steel bar usage based on the objective function, and calculating a constrained multi-objective programming function with the objective of minimizing the concrete volume and the steel bar usage; Wherein, the processing module includes: Processing unit: used for initializing tunnel parameters according to the particle swarm optimization algorithm; Prediction unit: used for learning the tunnel sample data in the MSVR model and performing tunnel sample prediction based on the result of initializing the tunnel parameters, wherein the result of initialization includes: determining the particle swarm size, the number of iterations and the particle random solution, each particle is one MSVR model, and one MSVR model includes an insensitive loss coefficient, a penalty coefficient and a width coefficient of the RBF kernel; Evaluation unit: used to evaluate the fitness of the particles in the particle swarm optimization algorithm according to the prediction results of the tunnel samples; An updating unit: used for updating the speed and position of the particle according to the calculation result of the fitness of the particle, and forming a new MSVR model; Loop unit: used to perform judgment loop on the MSVR model according to the iterative algorithm until the prediction error of the MSVR model is less than a specific value or reaches the maximum iteration set in the particle swarm optimization algorithm, completes the iteration of the MSVR model, stops the calculation, outputs the MSVR parameters at this time, and obtains the optimal MSVR model.
3. A tunnel construction parameter optimization device, characterized in that: include: Memory for storing computer programs; A processor is used to implement the steps of the tunnel construction parameter optimization method as claimed in claim 1 when executing the computer program.
4. A readable storage medium, characterized in that: The readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the tunnel construction parameter optimization method as claimed in claim 1 are implemented.
