Tunnel surrounding rock mechanical parameter inversion method based on parrot optimization algorithm

Abstract

The invention relates to a method for inverting tunnel surrounding rock mechanical parameters based on a parrot optimization algorithm. Comprising the following steps: S1, determining a constitutive model of tunnel surrounding rock according to a geological survey report or an indoor and outdoor rock mechanics test result, and establishing an ABAQUS numerical calculation model of the tunnel surrounding rock; s2, converting a parameter inversion problem into an optimization problem, and establishing an objective function of the inversion optimization problem; s3, solving the target function by adopting a parrot optimization algorithm; s4, subsequent excavation of the tunnel is guided according to the surrounding rock mechanics obtained through inversion; the method effectively solves the problem that the tunnel surrounding rock mechanical parameters are difficult to rapidly and reasonably valued.

Description

Technical Field

[0001] This invention patent relates to a method for inverting the mechanical parameters of tunnel surrounding rock based on the parrot optimization algorithm, belonging to the field of geotechnical engineering technology. Background Technology

[0002] With the advancement of infrastructure construction, tunnel engineering in my country is gradually developing towards deep burial, high water pressure, and complex geological conditions. This trend is evidenced by numerous practical engineering cases. For example, the Chongqing Yangtze River Tunnel is my country's first high-speed railway tunnel crossing a river, with a total length of 11.942 kilometers; the Qinling Zhongnanshan Tunnel is my country's longest operational highway tunnel, with a length of 18.02 kilometers, constructed using the drill-and-blast method. Tunnel deformation during the excavation process is difficult to predict and monitor. This is because rock mass is a product of complex geological processes, and rock mechanical parameters are irregularly distributed within the rock mass, making traditional mathematical analysis methods ineffective. With the development of computer technology, numerical analysis methods have been widely used in predicting rock tunnel deformation; however, numerical analysis methods are highly dependent on the selection of model parameters. How to select model parameters is crucial for the predictive power of numerical analysis methods.

[0003] In tunnel design and construction, the significant scale effect of rock, coupled with the diversity and discreteness of geological conditions, makes it difficult to accurately obtain the true mechanical parameters of the surrounding rock through traditional laboratory tests, often leading to biased parameter estimation. Inaccurate parameter estimation can result in misjudgments of surrounding rock stability, further affecting the design of support structures and the selection of excavation schemes. Currently, numerical simulation of tunnels faces challenges such as long computation times and difficulty in accurately determining the mechanical parameters of the surrounding rock. The parrot optimization (PO) algorithm-based method for inverting surrounding rock mechanical parameters offers an effective solution. By utilizing field monitoring information, it determines the mechanical parameters of the surrounding rock, thereby improving the reliability and accuracy of numerical simulation. The parameter inversion problem is transformed into an unconstrained mathematical optimization problem, using the difference between calculated and monitored displacement values ​​as the objective function, and the surrounding rock mechanical parameters as optimization variables. The parrot optimization algorithm is then used to solve the problem, ultimately yielding the optimal solution.

[0004] Invention Patent Content

[0005] To overcome the shortcomings of existing technologies, this invention provides a method for inverting tunnel surrounding rock mechanical parameters based on the Parrot Optimization Algorithm.

[0006] To achieve the above inversion method, this invention patent is implemented through the following technical solution:

[0007] A method for inverting tunnel surrounding rock mechanical parameters based on the parrot optimization algorithm includes the following steps:

[0008] Step S1: Determine the constitutive model of the tunnel surrounding rock based on the geological survey report or indoor and outdoor rock mechanics test results, and establish the ABAQUS numerical calculation model of the tunnel surrounding rock;

[0009] Step S2: Transform the parameter inversion problem into an optimization problem, and establish the objective function for the inversion optimization problem: Where x is a set of surrounding rock mechanical parameters, d i (x) represents the measured displacement d at the i-th displacement monitoring point. i (x), Let ε be the calculated displacement of the i-th displacement monitoring point, and let ε be the convergence accuracy of the algorithm. The smaller the objective function, the closer the calculated displacement is to the measured displacement, and the higher the reliability of the calculation results of the corresponding ABAQUS numerical model.

[0010] Step S3: Solve the objective function using the Parrot optimization algorithm. The optimization steps are as follows:

[0011] Step S3-1: Algorithm parameter settings: Determine the population size N based on the number of rock mass parameters to be inverted, and set the algorithm's convergence accuracy ε and the maximum allowed number of iterations T. max Initial parameters such as foraging strategy coefficient α and individual location boundaries;

[0012] Step S3-2: Randomly generate the initial position vector x of individual parrots. i Each individual represents a possible solution, and the fitness value f(x) is calculated. i ), where i = 1, 2, ..., N;

[0013] Step S3-3: For the experimental population x i Fitness evaluation is performed to obtain the objective function value f(x) for all individuals. i ), current x i The corresponding objective function value f(x) i The current global optimal solution is t, where the iteration count is t = 1.

[0014] Step S3-4: When f(x) i )<ε,t <T max If yes, then proceed with the following steps; otherwise, output the inverted parameter values.

[0015] Steps S3-5: After the above foraging, communication, staying, and moving away from strangers, perform the following steps on the currently updated individual x. i t+1 Use ABAQUS to perform the simulation and obtain the simulation displacement results;

[0016] Step S3-6: Calculate fitness. Based on the displacement results obtained from ABAQUS, calculate the fitness value f(x) of the individual. it +1 ( ), used to evaluate the merits and demerits of individuals.

[0017] Step S3-7: Determine if the convergence condition is met. If the global optimal fitness value f(x) is satisfied... i If ) < ε, or the current iteration number t ≥ T max If convergence is achieved, the iteration stops, and the current global best individual is output as the optimal solution for the soil and rock parameters. If the convergence condition is not met, return to step S3-3 and continue the process of foraging, communicating, staying, and moving away from strangers.

[0018] Step S4: Based on the inversion, the predicted values ​​of the mechanical parameters of the surrounding rock of the tunnel are used to guide the subsequent excavation of the tunnel.

[0019] Furthermore, the method for solving the individual fitness value is as follows: A joint inversion is performed using Python software and the commercial numerical calculation software ABAQUS. In the Python environment, the individual position coordinates from the Parrot optimization algorithm are saved to data interface file A. A custom script is used to start the ABAQUS numerical calculation software and enter working mode. The built-in user subroutine of ABAQUS is then called to read a set of surrounding rock mechanical parameters from data interface file A, which are substituted into the established ABAQUS numerical model to obtain the calculated displacement, thereby obtaining the objective function value. Then, the objective function value is saved in data file B, and the Python program reads the objective function value from file B to obtain the fitness value of the individual. The beneficial effects of this invention are:

[0020] (1) This invention patent is a non-embedded technical architecture that integrates numerical calculation software as a separate module into the PO algorithm. With the assistance of the efficient kernel solver of the numerical simulation software, it can significantly accelerate the tunnel inversion analysis, providing users with great convenience and strong practicality.

[0021] (2) Compared with traditional inversion methods such as particle swarm optimization, gray wolf algorithm and zebra algorithm, this invention can obtain reasonable values ​​of surrounding rock mechanical parameters in tunnel simulation calculation more efficiently and quickly, thereby effectively solving the problem of excessive calculation time in the current inversion of surrounding rock mechanical parameters of tunnel excavation.

[0022] (3) This invention effectively solves the problem of the difficulty in quickly and reasonably determining the mechanical parameters of tunnel surrounding rock. This method transforms the parameter inversion problem into an unconstrained mathematical optimization problem. The objective function is the sum of squared errors between the finite element calculations and monitoring values ​​of the surrounding rock convergence displacement. The surrounding rock mechanical parameters are used as optimization variables. Combined with a finite element model for tunnel surrounding rock displacement calculation based on ABAQUS finite element numerical software, the biomimetic global optimization algorithm, the Parrot Algorithm, is used to quickly search for the global optimal solution. Inputting the global optimal solution, i.e., the surrounding rock mechanical parameter values, into the finite element model yields displacement calculation results that approximate the monitoring values ​​of the surrounding rock convergence displacement. Attached Figure Description

[0023] Figure 1 This invention patent discloses a flowchart of a method for inverting tunnel surrounding rock mechanical parameters based on the parrot optimization algorithm;

[0024] Figure 2 Finite element model diagram of the tunnel of this invention;

[0025] Figure 3 The layout diagram of the measuring points in the tunnel of this invention;

[0026] Figure 4 Error diagram of tunnel monitoring displacement and calculated displacement according to the present invention;

[0027] Figure 5 A comparison chart of the average error and computation time of the Parrot Algorithm with other algorithms in Embodiment 1 of this invention; Detailed Implementation

[0028] The technical solution of this invention patent will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0029] See Figure 1 , Figure 2 , Figure 3 , Figure 4 and Figure 5 As shown, this invention provides a method for inverting tunnel surrounding rock mechanical parameters based on the Parrot optimization algorithm, comprising the following steps:

[0030] Step S1: Determine the constitutive model of the tunnel surrounding rock based on the geological survey report or indoor and outdoor rock mechanics test results, and establish the ABAQUS numerical calculation model of the tunnel surrounding rock.

[0031] Step S2: Transform the parameter inversion problem into an optimization problem, and establish the objective function for the inversion optimization problem: Where x is a set of surrounding rock mechanical parameters, d i (x) represents the measured displacement d at the i-th displacement monitoring point. i (x), Let ε be the calculated displacement of the i-th displacement monitoring point, and let ε be the convergence accuracy of the algorithm. The smaller the objective function, the closer the calculated displacement is to the measured displacement, and the higher the reliability of the calculation results of the corresponding ABAQUS numerical model.

[0032] Step S3: Solve the objective function using the Parrot optimization algorithm. The optimization steps are as follows:

[0033] Step S3-1: Algorithm parameter settings: Determine the scale N based on the number of rock mass parameters to be inverted, set the algorithm's convergence accuracy ε, and the maximum allowed number of iterations T. max Initial parameters such as foraging strategy coefficient α and individual location boundaries;

[0034] Reset population N:

[0035] x i =rand(ub-lb)+lb (1)

[0036] In the formula: ub is the upper bound of the search space; lb is the lower bound of the search space; rand is a random number in the interval [0,1]; x i Let be the initial position of the i-th parrot.

[0037] During foraging behavior, parrots primarily estimate the approximate location of food by observing its location or considering the location of their owner. Their positional movements follow the formula below:

[0038]

[0039] In the formula: x i t+1 To update the position; x mean The average position within the current population; Levy(d) describes the parrot's flight behavior; x best This is the best location found in the current search.

[0040] Stop location:

[0041]

[0042] In the formula: x i t+1 To update the position; x best The current best position found; ones(1,d) is a row vector of all 1s in dimension d.

[0043] Parrots are naturally social animals, characterized by close communication within their groups. This communication includes both flying towards the flock and not flying towards it. In PO (Positioning of the Group), it is assumed that these two behaviors occur with equal probability, and the average position of the current population is used to symbolize the center of the group. This process can be represented as:

[0044]

[0045] The behavior of distancing oneself from unfamiliar individuals and seeking a safer environment can be represented by the following formula:

[0046]

[0047] Step S3-2: Randomly generate the initial position vector x of individual parrots. i Each individual represents a possible solution, and the fitness value f(x) is calculated. i ), where i = 1, 2, ..., N;

[0048] Step S3-3: For the experimental population x i Fitness evaluation is performed to obtain the objective function value f(x) for all individuals. i ), current x i The corresponding objective function value f(x) i The current global optimal solution is t, where the iteration count is t = 1.

[0049] Step S3-4: When f(x) i )<ε,t <T max If yes, then proceed with the following steps; otherwise, output the inverted parameter values.

[0050] Steps S3-5: After the above foraging, communication, staying, and avoiding strangers, analyze the currently updated individual x. i t+1 Use ABAQUS to perform the simulation and obtain the simulation displacement results;

[0051] Step S3-6: Calculate fitness. Based on the displacement results obtained from ABAQUS, calculate the fitness value f(x) of the individual. i t +1 ( ), used to evaluate the merits and demerits of individuals.

[0052] Step S3-7: Determine if the convergence condition is met. If the global optimal fitness value f(x) is satisfied... i If ) < ε, or the current iteration number t ≥ T max If convergence is achieved, the iteration stops, and the current global best individual is output as the optimal solution for the soil and rock parameters. If the convergence condition is not met, the process returns to step S3-3 and continues the process of foraging, communicating, staying, and moving away from strangers.

[0053] Step S4 uses the surrounding rock mechanics obtained from the inversion to guide the subsequent excavation of the tunnel.

[0054] As a preferred embodiment, the method for solving the individual fitness value is as follows: The open-source software Python and the commercial numerical calculation software ABAQUS are used for joint inversion. In the Python environment, the individual position coordinates (a set of surrounding rock mechanical parameters) of the Parrot Algorithm are saved to data interface file A. A custom script is used to call commands to start the ABAQUS numerical calculation software and enter working state. The built-in user subroutine of ABAQUS is called to read a set of surrounding rock mechanical parameters from data interface file A, which are then substituted into the established ABAQUS numerical model to obtain the calculated displacement, and thus obtain the objective function value. The objective function value is then saved in data file B, and the Python program reads the objective function value from file B to obtain the fitness value of the individual.

[0055] Example 1:

[0056] The tunnel in a certain project has a diameter of 5m, with mean lithology and initial horizontal and vertical in-situ stresses of 15MPa. Assume the constitutive model is a Mohr-Coulomb ideal elastoplastic model, and its constitutive model obeys the Mohr-Coulomb criterion formula:

[0057]

[0058] Where, τ f σ represents the shear strength of the rock, c represents the cohesion of the rock (i.e., the shear strength when the normal stress is zero), and σ represents the shear strength of the rock. n This represents the normal component of the normal stress in the rock. This indicates the internal friction angle of the rock.

[0059] The surrounding rock mechanical parameters to be inverted are elastic modulus E, cohesion c, Poisson's ratio v, and internal friction angle. Five monitoring points were selected at 0°, 30°, 45°, 60°, and 90° along the tunnel circumferential direction (e.g., ...). Figure 4 To verify the feasibility of the algorithm, the actual mechanical parameters of the surrounding rock are assumed to be E = 2.1 GPa and c = 1.25 MPa. Substituting v = 0.25 into the ABAQUS program to obtain the displacement is taken as the "measured displacement". An objective function for inversion optimization is then established. To calculate the displacement, d i (x) represents the monitored displacement; the assumed true mechanical parameters of the surrounding rock are E = 2.0 GPa, c = 1.25 MPa, The displacement obtained by substituting v = 0.25 into the ABAQUS program for forward calculation is used as the "monitoring displacement" d. i (x); The calculated displacement results are shown in Table 1.

[0060] Table 1 Measured displacement values ​​at each monitoring point

[0061]

[0062]

[0063] The search range for the inversion parameters is set as follows: 2.0 GPa ≤ E ≤ 2.5 GPa, 3 MPa ≤ c ≤ 5 MPa. 0.2 ≤ v ≤ 0.3. The convergence criterion is always f(X) < 0.005. The Python software is used to solve the objective function using the PO algorithm, with an initial population N = 100 and a maximum number of iterations T. max =1000, foraging behavior probability p f =0.7, probability of staying behavior p s =0.2. The inverted displacement is shown in Table 2, and the calculation results are shown in Table 3. The average error is calculated for the elastic modulus E, cohesion c, and internal friction angle. The average of the relative errors and Poisson's ratio v.

[0064] Table 2. Inverted displacement values ​​for each monitoring point

[0065]

[0066] Table 3 Inversion Calculation Results

[0067]

[0068] Following the same steps, with an initial population N = 100 and a maximum number of iterations T max =1000. The Particle Swarm Optimization (PSO) algorithm, Gray Wolf (GWO) algorithm, and Zebra (ZO) algorithm were used in conjunction with ABAQUS for inversion. The inversion results are shown in Table 4, where the average error is calculated for elastic modulus E, cohesion c, and internal friction angle. The average of the relative errors and Poisson's ratio v.

[0069] Table 4 shows a comparison of the calculation results with each algorithm.

[0070]

[0071] As can be seen from Table 4, the PO algorithm is more efficient and faster in obtaining the values ​​of surrounding rock mechanical parameters in tunnel simulation calculations compared with traditional optimization algorithms such as PSO, GWO and ZO algorithms, further demonstrating the superiority of the PO optimization algorithm.

[0072] It should be noted that the above description is merely a preferred embodiment of the present invention and does not limit the scope of the invention. For those skilled in the art, various modifications and variations can be made to this application. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention should be considered within the scope of protection of the present invention. The scope of protection claimed by the present invention is determined by the scope defined in the claims.

Claims

1. A method for inverting tunnel surrounding rock mechanical parameters based on the parrot optimization algorithm, characterized in that, Includes the following steps: Step S1: Determine the constitutive model of the tunnel surrounding rock based on the geological survey report or indoor and outdoor rock mechanics test results, and establish the ABAQUS numerical calculation model of the tunnel surrounding rock; Step S2: Transform the parameter inversion problem into an optimization problem, and establish the objective function for the inversion optimization problem: Where x is a set of surrounding rock mechanical parameters, d i (x) represents the measured displacement d at the i-th displacement monitoring point. i (x), Let ε be the calculated displacement of the i-th displacement monitoring point, and let ε be the convergence accuracy of the algorithm. The smaller the objective function, the closer the calculated displacement is to the measured displacement, and the higher the reliability of the calculation results of the corresponding ABAQUS numerical model. Step S3: Solve the objective function using the Parrot optimization algorithm. The optimization steps are as follows: Step S3-1: Algorithm parameter settings: Determine the population size N based on the number of rock mass parameters to be inverted, and set the algorithm's convergence accuracy ε and the maximum allowed number of iterations T. max Initial parameters such as foraging strategy coefficient α and individual location boundaries; Step S3-2: Randomly generate the initial position vector x of individual parrots. i Each individual represents a possible solution, and the fitness value f(x) is calculated. i ), where i = 1, 2, ..., N; Step S3-3: For the experimental population x i Fitness evaluation is performed to obtain the objective function value f(x) for all individuals. i ), current x i The corresponding objective function value f(x) i The current global optimal solution is t, where the iteration count is t = 1. Step S3-4: When f(x) i )<ε,t <T max Then proceed with the following steps; Otherwise, output the inversion parameter values; Steps S3-5: After the above foraging, communication, staying, and moving away from strangers, perform the following steps on the currently updated individual x. i t+1 Use ABAQUS to perform the simulation and obtain the simulation displacement results; Step S3-6: Calculate fitness. Based on the displacement results obtained from ABAQUS, calculate the fitness value f(x) of the individual. i t+1 ( ), used to evaluate the merits and demerits of an individual. Step S3-7: Determine if the convergence condition is met. If the global optimal fitness value f(x) is satisfied... i If ) < ε, or the current iteration number t ≥ T max If convergence is achieved, the iteration stops, and the current global best individual is output as the optimal solution for the soil and rock parameters. If the convergence condition is not met, return to step S3-3 and continue the process of foraging, communicating, staying, and moving away from strangers. Step S4: Based on the inversion, the predicted values ​​of the mechanical parameters of the surrounding rock of the tunnel are used to guide the subsequent excavation of the tunnel.

2. The method for inverting tunnel surrounding rock mechanical parameters based on the parrot optimization algorithm according to claim 1, characterized in that, The method for solving the individual fitness value is as follows: Python software and the commercial numerical calculation software ABAQUS are used for joint inversion. In the Python environment, the individual position coordinates of the Parrot optimization algorithm are saved to data interface file A. A custom script is used to call the command to start the ABAQUS numerical calculation software and enter working state. The built-in user subroutine of ABAQUS is called to read a set of surrounding rock mechanical parameters from data interface file A, which are then substituted into the established ABAQUS numerical model to obtain the calculated displacement, and thus obtain the objective function value. The objective function value is then saved in data file B, and the Python program reads the objective function value from file B to obtain the fitness value of the individual.

Images