CN-122021256-A - Hydraulic tunnel surrounding rock mechanical parameter inversion method based on FGO-CB collaborative optimization algorithm

CN122021256ACN 122021256 ACN122021256 ACN 122021256ACN-122021256-A

Abstract

The invention provides a hydraulic tunnel surrounding rock mechanical parameter inversion method based on FGO-CB collaborative optimization algorithm, which relates to the technical field of hydraulic and hydroelectric engineering, combines FGO global optimization algorithm with CB local agent model, fully exerts the super-strong exploration ability of FGO algorithm in global optimization, meanwhile, the CB model is utilized to have excellent predictive regression analysis capability on the search space of the mycelium colony, the calling times of the refined numerical model in the inversion process are obviously reduced, the convergence condition is rapidly reached, and the inversion precision is further improved.

Inventors

LI YIMING
Peng Chengju
YAN JIUQIU
Yin Yiting
CHEN XIANJIE
SU GUOSHAO
LUO DANNI
GAN BIN
LIN DONGCAI
WEI YU

Assignees

广西壮族自治区水利电力勘测设计研究院有限责任公司

Dates

Publication Date: 20260512
Application Date: 20251229

Claims (6)

1. The hydraulic tunnel surrounding rock mechanical parameter inversion method based on the FGO-CB collaborative optimization algorithm is characterized by comprising the following steps of: the method comprises the steps of S1, establishing a numerical calculation model of hydraulic tunnel surrounding rock, namely finely modeling the hydraulic tunnel and the surrounding rock thereof according to a hydraulic tunnel design drawing, and establishing the numerical calculation model of the hydraulic tunnel surrounding rock by adopting an elastoplastic constitutive model which accords with a Hooke-Brownian destruction criterion; s2, determining an optimization variable, wherein rock mechanical parameters of an elastoplastic constitutive model which accords with a Hooke-Brownian destruction criterion are used as the optimization variable; S3, establishing a displacement error minimum optimization objective function, namely taking a residual error between a displacement calculated value of the numerical calculation model and an engineering actually-measured displacement value as the objective function; S4, performing global optimization on the parameter space of the objective function by adopting an FGO-CB collaborative optimization algorithm, so as to obtain optimal rock mechanical parameters; S5, bringing the optimal rock mechanical parameters into a numerical calculation model of the hydraulic tunnel surrounding rock, and carrying out support optimization and surrounding rock deformation management on the rock mechanical parameters obtained by inversion of the numerical calculation model of the hydraulic tunnel surrounding rock based on the optimal rock mechanical parameters.
2. The method for inverting the mechanical parameters of the hydraulic tunnel surrounding rock based on the FGO-CB collaborative optimization algorithm according to claim 1, wherein in the step S2, the objective function is: ; in the formula, i is the ith displacement measuring point, X is an optimization variable combination obtained through an optimization algorithm, namely the optimal value of the model rock mechanical parameter; Performing a numerical model displacement calculation value of an ith displacement measuring point of surrounding rock of a current tunnel excavation section through an input operator X; The method comprises the steps of excavating an actual measurement displacement value of an ith displacement measuring point of a section surrounding rock for a current tunnel; in the step S3, the objective of global optimization of the parameter space is to enable the objective function f (X) to be minimum, namely the average error between the numerical model displacement calculated value of the surrounding rock displacement measuring point of the current tunnel excavation section and the actually measured displacement value reaches the minimum value.
3. The method for inverting mechanical parameters of hydraulic tunnel surrounding rock based on FGO-CB collaborative optimization algorithm according to claim 1, wherein the optimization variables comprise elastic modulus E, poisson ratio v, empirical parameters m b , s, a of Hoek-Brown criterion and rock uniaxial compressive strength sigma ci , the rock mass experimental parameters m b , s, a are replaced by a rock mass scoring system RMR and a rock mass integrity coefficient K v as optimization variables to be considered, wherein the rock mass scoring system RMR is determined by comprehensively evaluating rock mass quality indexes according to factors of 5 aspects of uniaxial compressive strength of rock mass, RQD value of rock core, joint spacing, joint condition and groundwater effect.
4. The method for inverting the mechanical parameters of the hydraulic tunnel surrounding rock based on the FGO-CB collaborative optimization algorithm according to claim 1, wherein the step S3 comprises the following steps: A1, setting FGO-CB collaborative optimization algorithm parameters, namely determining a seed scale N according to the number of rock parameters to be inverted, setting convergence accuracy epsilon of an algorithm, and solving exploration and attraction weights of an exploration rate E r Maximum optimization times t max , entering local proxy times t CB and training sample number m; step A2, initializing a population, namely randomly initializing N mycelium populations; a3, obtaining nutrition conditions of all hypha positions by solving an objective function And the optimal hypha ; Step A4, controlling a mycelium population by using random numbers r 1 and r 2 , wherein r 1 and r 2 are random numbers of 0-1, if r 1 >r 2 , performing mycelium tip growth behavior on the mycelium population, and otherwise, performing mycelium branching and spore germination behavior; step A5, if the mycelium population performs mycelium tip growth behavior, then: Hyphae in the hypha population differentiated themselves into growth exploration or growth attraction phase by comparing nutritional status P i with exploration rate E r : if the mycelium is P i > E r , the mycelium will perform growth exploration, and the growth position is updated according to the following formula; ; In the formula, The growth position after the i-th hypha update; The current growth position of the ith hypha; E is the growth rate of hyphae; Is the growth direction of hypha; On the contrary, P i ≦E r , the mycelium enters a growth suction stage, and the growth position is updated according to the following formula: ; In the formula, A random number of 0 to 1; to be attracted by the most nutritious hyphae, the growth direction of the hyphae is changed; Is an environmental impact parameter; if the mycelium population performs mycelium branching and spore germination actions, then: Controlling hypha individuals in the hypha population to carry out hypha branching or spore germination according to the generated random number r 8 , and carrying out branching growth on the hypha when r 8 is a random number of 0-1 and r 8 is more than 0.5, and updating the growth position according to the following formula; ; In the formula, To update the mycelium locations at the mycelium branching stage, Is the current position of mycelium in the mycelium branching stage, E L is the branching growth speed, And The growth directions under the influence of individual hyphae of the population and hyphae of the most nutrient areas are respectively, and r 9 is a random number of 0-1; On the contrary, r 8 is less than or equal to 0.5, the hypha germinates spores, and the growth position is updated according to the following formula; ; In the formula, To update the new position of the spores, Is the current position of the spores; Is a random value of-1 or 1, R 10 is a random value of 0-1, E is the growth rate of hypha; Representing the dimension of the problem; Step A6, comparing the nutrition condition of the current position of the mycelium with the growth position, namely comparing the objective function value corresponding to the current position with the growth position, if The mycelium grows to Otherwise, hypha does not grow ; Step A7, if i is not less than N, finishing the growth position updating of all mycelium individuals in the mycelium population, otherwise, i=i+1, and returning to the step A5; Step A8, if the current optimization time t is smaller than the entering local agent time t CB , t=t+1 and returning to step A4, otherwise, t CB =t+ t CB ; step A9, sequentially selecting m nearest The selected m hypha growth positions and corresponding nutrition conditions As training samples; Step A10, setting training parameters of the CB local agent model, and training the CB local agent model by using the training samples to obtain the adaptability At the position of CB approximation optimization objective function distribution in local neighborhood of (2) ; Step A11 obtaining a function Optimum value in distribution Corresponding to , The mycelium position corresponding to the optimal value predicted by the CB local agent model is determined as the function Optimum value in distribution Is superior to Then update = ; Step A12, if < Ε, or t is greater than t max , then the optimization is ended, otherwise, t=t+1 and return to step A4; Step A13, outputting the current global optimal mycelium position Current global optimal hypha position The optimal rock mechanical parameters are obtained.
5. The method for inverting the mechanical parameters of the hydraulic tunnel surrounding rock based on the FGO-CB collaborative optimization algorithm according to claim 4, wherein the ratio of t CB to the maximum optimizing frequency t max is 10-40, namely, local agent is performed 10-40 times in the optimizing process.
6. The method for inverting the mechanical parameters of the hydraulic tunnel surrounding rock based on the FGO-CB collaborative optimization algorithm according to claim 4, wherein the number m of training samples is 50-100.

Description

Hydraulic tunnel surrounding rock mechanical parameter inversion method based on FGO-CB collaborative optimization algorithm Technical Field The invention relates to the technical field of hydraulic and hydroelectric engineering, in particular to a hydraulic tunnel surrounding rock mechanical parameter inversion method based on an FGO-CB collaborative optimization algorithm. Background Along with the system propulsion and continuous investment of the national water network strategy, the hydraulic tunnel construction in the China water conservancy and hydropower engineering enters a high-strength networking stage. The water network is a comprehensive system which integrates the water resource optimization configuration, the river flood control and disaster reduction and the water ecological protection into a whole by taking natural rivers and lakes as the basis, taking the drainage engineering as a channel, taking the storage engineering as a node and taking intelligent regulation as a means. In the system, a large number of hydraulic tunnels for communicating each level of water network and connecting a water source area with a water use area are planned, designed and built, and the hydraulic tunnels become a core infrastructure for guaranteeing the function of the water network to be in a 'skeletonized' ground. Surrounding rock stability of hydraulic tunnels is a core problem for ensuring smooth implementation of water network construction. In recent years, the hydraulic tunnel of the water network engineering is remarkably enlarged in scale, and the common increase of the lengths of the single tunnels leads to more complex geological environments spanned by the engineering, namely, long-distance tunneling needs to pass through various stratum combinations, high-ground stress areas, fracture zones, karst and poor water-rich areas are frequently alternated, and the surrounding rock mechanical properties are remarkably heterogeneous and uncertain, so that the difficulty of analyzing the surrounding rock mechanical behaviors is remarkably raised. Particularly, under the condition of coupling a long large tunnel with complex geology, once surrounding rock is unstable, risks such as collapse, leakage, water and mud bursting and the like are extremely easy to be caused, and thus construction safety, construction period and construction cost are affected. Therefore, the stress mechanism of the surrounding rock is deeply characterized and accurately mastered, and the key premise of ensuring the safety and stability of the cavity group is realized, and the advanced numerical simulation and optimization design method is mainly relied on at present. However, one major challenge faced when applying numerical simulation techniques is the "parameter inaccuracy" of the numerical simulation model. The mechanical parameters of the surrounding rock dynamically change along with the change of the geological conditions of the hydraulic tunnel path. For example, once the construction section of a hydraulic tunnel encounters a significant change in geological conditions, the mechanical parameters of the cavity surrounding rock may change significantly, such that the difference between the displacement results of the numerical simulation using the early mechanical parameters and the monitored deformation of the actual cavity surrounding rock is out of range, thereby resulting in numerical simulation distortion, which cannot effectively simulate the surrounding rock mechanical behavior of the existing cavity section. To solve this problem, an optimized inverse analysis method based on the sidewall displacement monitoring data is verified as an effective countermeasure strategy. During the construction stage, the displacement conditions of the side walls of the cavern are monitored in real time, and the changes of the displacements are closely related to the mechanical parameters. By means of on-site monitoring data, displacement inverse analysis work is rapidly carried out, mechanical parameters of the side wall of the cavity can be accurately measured, and further the estimation accuracy and reliability of the numerical calculation model are improved. The continuous improvement of engineering safety standards puts more stringent requirements on the optimization inverse analysis precision based on a numerical simulation model, and the error tolerance of simulation results in practical engineering application is tightened by orders of magnitude. However, the traditional optimization inversion algorithm (such as particle swarm optimization algorithm and genetic algorithm) has difficulty in meeting the requirement of modern engineering on calculation accuracy. In this context, a new generation of intelligent optimization algorithms has developed, typically representing e.g. bat optimization algorithms (BO algorithms). Such improved algorithms have significant advantages in terms of global optimizing capabilities. However, as the gri