Genetic algorithm-based foundation pit inner support section optimization method

Abstract

The invention discloses a foundation pit inner support section optimization method based on a genetic algorithm, and the method comprises the steps: representing the shape and size of an inner support section through the chromosome of an individual, evaluating the quality degree of the individual through a fitness function, and carrying out the optimization search through the operation of selection, intersection and variation; in each generation, individuals are selected according to a fitness function, then crossover and mutation operations are performed to generate new individuals, and an optimal section shape and size combination is found through continuous iterative optimization, so that the structure has the minimum weight or cost on the premise of meeting constraint conditions. According to the invention, the calculation parameters of the algorithm are adaptively adjusted through the genetic algorithm and the optimization module based on the neural network, so that the logic of the genetic algorithm better fits the actual situation, the efficiency and accuracy of the genetic algorithm are improved, and the calculation structure is better.

Description

Technical Field

[0001] The invention relates to the technical field of internal support cross-section optimization, and in particular to a method for optimizing the internal support cross-section of a foundation pit based on a genetic algorithm. Background Art

[0002] The internal support cross section optimization problem is one of the common optimization problems in the engineering field. Its goal is to adjust the cross-sectional shape and size so that the structure has the minimum weight or cost under the premise of satisfying certain constraints, thereby achieving the optimal structure or cost of the engineering construction. When optimizing the internal support cross section, it usually involves multiple design variables, such as the height, width, thickness of the cross section, etc., as well as multiple different constraints, such as bearing capacity, deflection, etc. Conventional optimization methods are mostly carried out by manual trial calculation, but there are defects such as large calculation limitations, large errors and low efficiency. Therefore, a method of cross section optimization using genetic algorithms has emerged, such as the scheme disclosed in CN106547987A for optimization, but there are still problems of poor algorithm performance and adaptability in actual use, resulting in a large difference between the calculated results and the actual situation, and the reference significance is not great. The calculation results have a strong knowledge of the improvement direction of the construction implementation, so it is necessary to ensure that the calculation results provide a better reference for the actual construction on specific issues. The present invention provides a method for optimizing the cross section of the support in the foundation pit based on a genetic algorithm to solve the above problems. Summary of the invention

[0003] The invention provides a method for optimizing the support section in a foundation pit based on a genetic algorithm. By improving the genetic algorithm, the calculation result obtained by the method is better, and a correct guiding direction is provided for construction improvement.

[0004] The technical solution adopted by the present invention to solve the above technical problems is: A method for optimizing the support section in a foundation pit based on a genetic algorithm comprises the following steps: S1, establish an optimization model; the model includes an input module, a calculation module, an optimization module and an output module; the input module is used to input initial data, the calculation module is used to calculate the fitness of the initial data, the optimization module is used to optimize the performance of the calculation module, and the output module is used to output the calculation results; S2, input initial data and calculation parameters; input the initial data of the internal support section parameters and the calculation parameters through the input module for calculation; S3, model calculation: S31, initializing the population: the calculation module randomly generates the initial population from the initial data, and each individual in the population is an evaluation object; S32, calculating the individual fitness: each individual is evaluated and calculated by the fitness function to obtain the fitness value of each individual; S33, selection operation: selection is performed according to the fitness value. The higher the fitness value, the greater the probability of being selected. S34, crossover operation: perform a crossover operation on the selected individuals to generate new offspring individuals; S35, mutation operation: perform mutation operation on offspring individuals to introduce new genetic information; S36, evaluate the fitness of the mutant individuals in S35 and select them; S37, iterating according to the operations of S33-S36 until a termination condition is met; S4, the output module outputs the calculation results.

[0005] Furthermore, the optimization module is used to optimize the selection rate, crossover rate and mutation rate in the calculation module. The operations of the optimization module for training and optimizing the calculation module are as follows: S11, randomly generate a set of candidate solutions as the initial population, each solution contains parameters of the selection rate, crossover rate and mutation rate of the calculation module; S12, evaluating each solution by a fitness function in a calculation module to obtain a fitness value of each solution; S13, taking the solutions in the population and their corresponding fitness values ​​as training data, and inputting them into the optimization module to train the optimization module, and training the optimization module on the mapping relationship between the solutions and their fitness values; S14, using the preliminarily trained optimization module to predict the fitness value of each solution in the population, and the optimization module automatically adjusts the selection rate, crossover rate and mutation rate of the input calculation module according to the prediction results; S15, after the initial data is input, the optimization module adaptively generates the selection rate, crossover rate and mutation rate and inputs them into the calculation module.

[0006] Furthermore, the calculation parameters include constraints, population size, number of iterations, selection rate, mutation rate and crossover rate. The population size and number of iterations are set manually, and the selection rate, mutation rate and crossover rate are adaptively generated through the optimization module.

[0007] Furthermore, in step S2, after the initial data is input, it is divided into two paths and enters the calculation module and the optimization module respectively. After the initial data enters the optimization module, the optimization module adaptively generates the selection rate, mutation rate and crossover rate according to the initial data, and inputs the selection rate, mutation rate and crossover rate into the calculation module. Then, the calculation module calculates the entered initial data through the above selection rate, mutation rate and crossover rate.

[0008] Furthermore, the cross-sectional parameters of the inner support structure have an optimal weight or cost after model calculation under the premise of satisfying the constraint conditions, and the constraint conditions include strength constraint, stiffness constraint and stability constraint; The requirements for strength constraints are as follows: σ max ≤f y In the formula, σ max Represents the ultimate stress value of the relevant component section, and the tensile and compressive design values ​​are fy=315MPa; The requirements for stiffness constraints are as follows: λ max ≤[λ] In the formula, λ max is the maximum slenderness ratio of different components, [λ] is the allowable slenderness ratio; The requirements for stability constraints are as follows: Overall stability requirements for compression-bending members in plane: In the formula, M x is the maximum bending moment in the compression-bending member; W1x is the resistance moment of the gross section; γ x is the cross-section plastic development coefficient, is the section coefficient in the x-axis direction; N ex is the standard axial force parameter; N is the axial force; A is the cross-sectional area, β x is the moment reduction factor; Calculated using Perry's formula: In the formula, α is the coefficient; Overall stability requirements for compression-bending members out of plane: In the formula, is the overall stability factor of the member under uniform bending moment, is the section coefficient in the y-axis direction, where the biaxially symmetrical steel section is: In the formula, is the slenderness ratio in the y direction; Local stability requirements: Where b1 is the flange width, t3 is the web thickness, and t1 is the flange thickness; Deflection constraint requirements: In the formula, l represents the span and max represents the maximum deflection of the structure; Purlin displacement constraint: U max ≤30mm In the formula, max represents the maximum displacement of the structure.

[0009] Furthermore, the initial data in step S2 and the calculated data in step S36 are normalized when entering the optimization module to form standard data.

[0010] Furthermore, the optimization module is a neural network module.

[0011] The beneficial effects of the present invention are as follows: The genetic algorithm is used to perform multiple iterative optimizations on the internal support cross-section data to obtain the optimal combination of cross-sectional shape and size, so that the structure has the minimum weight or cost while satisfying the constraints. The algorithm's calculation parameters are adaptively adjusted through an optimization module based on a neural network, making the logic of the genetic algorithm more in line with the actual situation, improving the efficiency and accuracy of the genetic algorithm, and thus making the calculation structure better. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 It is a schematic diagram of the structure of the genetic algorithm of the present invention; Figure 2 It is a schematic diagram of the steps of the optimization calculation method of the present invention; Figure 3 It is a schematic diagram of the cross section and formation characteristics of the rectangular cross-section internal support optimization embodiment of the present invention; Figure 4 It is a schematic diagram of the initial three iteration results of the rectangular cross-section internal support optimization embodiment of the present invention; Figure 5 This is a schematic diagram of the final three iteration results of the rectangular cross-section internal support optimization embodiment of the present invention; Figure 6 It is a schematic diagram of the cross section and formation characteristics of an optimized embodiment of the circular ring cross section of the present invention; Figure 7 It is a schematic diagram of the initial three iteration results of the embodiment of the circular ring cross-section internal support optimization of the present invention; Figure 8 This is a schematic diagram of the final three iteration results of the circular ring cross-section internal support optimization embodiment of the present invention; Fig. 9 It is a schematic diagram of the cross section and formation characteristics of an optimized embodiment of the I-shaped cross-section internal support of the present invention; Fig.10 It is a schematic diagram of the initial three iteration results of the I-section internal support optimization embodiment of the present invention; Fig.11 It is a schematic diagram of the final three iteration results of the I-section internal support optimization embodiment of the present invention. DETAILED DESCRIPTION

[0013] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings of the specification. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of 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.

[0014] In the description of the present invention, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship are based on the orientation or position relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0015] Genetic algorithm is a heuristic optimization algorithm that simulates the genetic mechanism and natural selection principle in the process of biological evolution, including three core basic operations: selection, crossover and mutation.

[0016] The selection operation evaluates individuals through the fitness function, and individuals with higher fitness are selected as parents to produce the next generation of individuals. In the selection process, individuals with higher fitness have a greater probability of being selected, thereby retaining the characteristics of excellent individuals.

[0017] The crossover operation simulates the gene exchange process of biological individuals. During the crossover process, the chromosomes of two parent individuals are exchanged to produce new offspring individuals. The crossover operation helps to generate new individuals and increase the diversity of the population. x′i=(xi1,x i2 ,…,x ic ,x j,c+1 ,…,x jn ) x′j=(xj1,x j2 ,…,x jc ,x i,c+1 ,…,x in )

[0018] The mutation operation simulates the process of biological gene mutation.

[0019] The basic principle of the genetic algorithm used in the present invention is as follows: the shape and size of the internal support section are represented by the chromosomes of the individuals, the quality of the individuals is evaluated by the fitness function, and the optimization search is performed using the operations of selection, crossover and mutation; in each generation, the individuals are selected according to the fitness function, and then crossover and mutation operations are performed to generate new individuals, and through continuous iterative optimization, the optimal combination of cross-sectional shape and size is found so that the structure has the minimum weight or cost while satisfying the constraints.

[0020] like Figure 1 , 2 As shown, a method for optimizing support cross-section in a foundation pit based on a genetic algorithm of the present invention specifically includes the following steps.

[0021] S1, establish an optimization model; the model includes an input module, a calculation module, an optimization module and an output module; the input module is used to input initial data, the calculation module is used to calculate the fitness of the initial data, the optimization module is used to optimize the performance of the calculation module, and the output module is used to output the calculation results; The optimization module is a neural network module, which is used to optimize the selection rate, crossover rate and mutation rate in the calculation module. The specific operations of the optimization module for training and optimizing the calculation module are as follows: S11, randomly generate a set of candidate solutions as the initial population, each solution contains parameters of the selection rate, crossover rate and mutation rate of the calculation module; S12, evaluating each solution by a fitness function in a calculation module to obtain a fitness value of each solution; S13, taking the solutions in the population and their corresponding fitness values ​​as training data, and inputting them into the optimization module to train the optimization module, and training the optimization module on the mapping relationship between the solutions and their fitness values; S14, using the preliminarily trained optimization module to predict the fitness value of each solution in the population, and the optimization module automatically adjusts the selection rate, crossover rate and mutation rate of the input calculation module according to the prediction results; S15, after the initial data is input, the optimization module adaptively generates the selection rate, crossover rate and mutation rate and inputs them into the calculation module.

[0022] The optimization module needs to be trained first to learn the mapping relationship of the calculation module. After training, the optimization module can adaptively adjust the selection rate, crossover rate and mutation rate according to the prediction, thereby improving the calculation accuracy of the calculation module and making its calculation process clearer.

[0023] S2, input initial data and calculation parameters; input the initial data of the internal support section parameters and the calculation parameters through the input module for calculation; the calculation parameters include constraint conditions, population size, number of iterations, selection rate, mutation rate and crossover rate; The population size and the number of iterations are set manually. The appropriate population size and the number of iterations are manually selected. Then the fitness function is defined and the constraint conditions of the cross section are calculated. The selection rate, mutation rate, and crossover rate are adaptively generated through the optimization module; The cross-sectional parameters of the internal support structure have the optimal weight or cost after model calculation under the premise of satisfying the constraint conditions, which include strength constraint, stiffness constraint and stability constraint; The requirements for strength constraints are as follows: σ max ≤f y In the formula, σ max Represents the ultimate stress value of the relevant component section, and the tensile and compressive design values ​​are fy=315MPa; The requirements for stiffness constraints are as follows: λ max ≤[λ] In the formula, λ max is the maximum slenderness ratio of different components, [λ] is the allowable slenderness ratio; The requirements for stability constraints are as follows: Overall stability requirements for compression-bending members in plane: In the formula, M x is the maximum bending moment in the compression-bending member; W1x is the resistance moment of the gross section; γ x is the cross-section plastic development coefficient, is the section coefficient in the x-axis direction; N ex is the standard axial force parameter; N is the axial force; A is the cross-sectional area, β x is the moment reduction factor; Calculated using Perry's formula: In the formula, α is the coefficient; Overall stability requirements for compression-bending members out of plane: In the formula, is the overall stability factor of the member under uniform bending moment, is the section coefficient in the y-axis direction, where the biaxially symmetrical steel section is: In the formula, is the slenderness ratio in the y direction; Local stability requirements: Where b1 is the flange width, t3 is the web thickness, and t1 is the flange thickness; Deflection constraint requirements: In the formula, l represents the span and max represents the maximum deflection of the structure; Purlin displacement constraint: U max ≤30mm In the formula, max represents the maximum displacement of the structure.

[0024] After the initial data is input, it is divided into two paths and enters the calculation module and the optimization module respectively. After the initial data enters the optimization module, the optimization module adaptively generates the selection rate, mutation rate and crossover rate according to the initial data, and inputs the selection rate, mutation rate and crossover rate into the calculation module. Then the calculation module calculates the entered initial data through the above selection rate, mutation rate and crossover rate.

[0025] like Figure 2 As shown in S3, the specific process of the model calculation is as follows: S31, initializing the population: the calculation module randomly generates the initial population from the initial data, and each individual in the population is an evaluation object; S32, calculating the individual fitness: each individual is evaluated and calculated by the fitness function to obtain the fitness value of each individual; S33, selection operation: select individuals according to their fitness values. The higher the fitness, the greater the probability that the individual will be selected. S34, crossover operation: perform a crossover operation on the selected individuals to generate new offspring individuals; S35, mutation operation: perform mutation operation on offspring individuals to introduce new genetic information; S36, evaluate the fitness of the mutated individuals in S35, and perform selection, crossover and mutation operations again; S37, iterating according to the operations of S33-S36 until a termination condition is satisfied, where the termination condition is set to reach a maximum number of iterations or the calculation result converges to an optimal solution; S4, stop the calculation after the termination condition is met, the output module outputs the calculation result, and the next step is performed according to the calculation result.

[0026] Further, the initial data in step S2 and the calculated data in S36 are normalized when entering the optimization module to form standard data. Since the shapes of the internal support cross sections are different, their size parameters are also different, and data normalization is required to form standard data. The optimization module of the neural network identifies the standard data, and then adaptively generates the corresponding selection rate, crossover rate and mutation rate according to the standard data and inputs them into the calculation module. If the calculation result obtained in step S36 meets the termination condition, it is directly output through the output module. If the termination condition is not met, the data of the calculation result is input into the optimization module after data normalization. The optimization module adaptively generates the corresponding selection rate, crossover rate and mutation rate for this group of data and inputs them into the calculation module. Then the calculation module calculates this group of data according to the generated selection rate, crossover rate and mutation rate. Each group of calculated data generates the corresponding selection rate, crossover rate and mutation rate through the optimization module. When the calculation module calculates each group of data, the corresponding selection rate, crossover rate and mutation rate are used, thereby improving the accuracy of the calculation. After the accuracy is improved, the number of iterations to calculate the optimal solution will also be reduced, thereby improving the calculation efficiency.

[0027] The shapes of the inner support cross section are divided into rectangular cross section, circular cross section and I-shaped cross section. The present invention performs optimization calculations for the three different shapes of inner supports.

[0028] like Figure 3 , 4 As shown in Figure 5, an application example of optimization of rectangular cross-section internal support based on genetic algorithm is shown. A soft soil foundation pit is adjacent to a river channel, and the plane is approximately trapezoidal. The project consists of two 50-story office buildings and a four-story basement. The excavation depth of the foundation pit is 25.15 meters, covering an area of ​​17,000 m2. On the north side of the excavation, 140 meters away from the edge of the excavation is the river channel. According to geological surveys, the typical strata at the excavation depth are miscellaneous fill, plain fill, silty clay, fine sand and limestone from top to bottom. The excavation depth of the foundation pit is 25m, and the retaining structure adopts the form of underground continuous wall plus internal support support. The thickness of the ground-connected wall is 1200mm and the depth is 50m; a total of 4 layers of C30 reinforced concrete internal support are set, with a cross-sectional size of 1000mm×1000mm. The groundwater level on site fluctuates with weather changes. Before excavation, precipitation wells are used to dewater and control the groundwater level 2-3 meters below the excavation surface of the foundation pit. Based on the above parameters, the genetic algorithm of the present invention is used to set the number of iterations to 10 times. 300 individual values ​​are selected from the population of the initial iteration. Selection, crossover and mutation operations are performed from the iteration conditions. Finally, a total of 8 individuals survive. By comparing the results of the first three iterations and the final three iterations, after 10 iterations, the individual with a width and height of (0.68m, 0.83m) is the best, which is better than the (1m, 1m) used in actual engineering.

[0029] like Figure 6 ,7 As shown in Figure 8, an application example of optimization of internal supports in a circular cross section based on a genetic algorithm is shown in a station foundation pit project, in which the second line uses a circular steel support. Through the genetic algorithm of the present invention, the number of iterations is set to 10, and the population of the initial iteration selects nearly 300 individual values. Selection, crossover, mutation and other operations are performed from the iteration conditions, and finally a total of 6 individuals survive. By comparing the results of the first three iterations and the final three iterations, after 10 iterations, the individuals with the best diameter and thickness (833mm, 24mm) are better than those (1000mm, 25mm) used in actual engineering.

[0030] like Fig. 9 , 10 As shown in 11, an application example of optimization of I-shaped cross-section internal support based on genetic algorithm is an underground garage project with a plane size (inner cavity size) of 8m×15m and a depth of 35.5m. It is located in a large basement foundation pit with a depth of 6m, which is a pit in a pit. The top of the eastern side of the large basement foundation pit is adjacent to a layer of brick-concrete residential houses. The terrain on the east side of the foundation pit is 5.6m higher than that on the west side, and there are two layers of brick-concrete houses on the top of the pit. It is a typical internal support application project under unbalanced earth pressure conditions. Based on the above parameters, the genetic algorithm of the present invention is used to set the number of iterations to 10 times. 300 individual values ​​are selected for the population of the initial iteration, and the default thickness is 25mm. Selection, crossover, mutation and other operations are performed from the iteration conditions. Finally, a total of 5 individuals survive. By comparing the results of the first three iterations and the final three iterations, after 10 iterations, the individual with width and height (0.68m, 1.19m) is the best, which is better than the (0.8m, 1.4m) used in the actual project.

[0031] It can be seen from the above embodiments that the calculation results obtained by the optimized genetic algorithm have a positive effect on the selection and optimization of construction materials, and have higher accuracy at a lower number of iterations, effectively improving the economy of construction.

[0032] It is obvious to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential features of the present invention. Therefore, the embodiments should be regarded as exemplary and non-limiting from any point of view, and the scope of the present invention is defined by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention, and any reference numerals in the claims should not be regarded as limiting the claims involved.

Claims

1. A method for optimizing the support section in a foundation pit based on a genetic algorithm, characterized in that: The following steps are involved: S1, establishing an optimization model; the model includes an input module, a calculation module, an optimization module and an output module; The input module is used to input initial data, the calculation module is used to calculate the fitness of the initial data, the optimization module is used to optimize the performance of the calculation module, and the output module is used to output the calculation results; S2, input initial data and calculation parameters; input the initial data of the internal support section parameters and the calculation parameters through the input module for calculation; S3, model calculation: S31, initializing the population: the calculation module randomly generates an initial population from the initial data, and each individual in the population is an evaluation object; S32, calculating individual fitness: evaluating and calculating each individual through the fitness function to obtain the fitness value of each individual; S33, selection operation: selection is performed according to the fitness value. The higher the fitness value, the greater the probability of being selected. S34, crossover operation: perform a crossover operation on the selected individuals to generate new offspring individuals; S35, mutation operation: perform mutation operation on offspring individuals to introduce new genetic information; S36, evaluate the fitness of the mutant individuals in S35 and select them; S37, iterating according to the operations of S33-S36 until a termination condition is met; S4, the output module outputs the calculation results.

2. The method for optimizing the support cross section in a foundation pit based on a genetic algorithm according to claim 1, characterized in that: The optimization module is used to optimize the selection rate, crossover rate and mutation rate in the calculation module. The operations of the optimization module for training and optimizing the calculation module are as follows: S11, randomly generate a set of candidate solutions as the initial population, each solution contains parameters of the selection rate, crossover rate and mutation rate of the calculation module; S12, evaluating each solution by a fitness function in a calculation module to obtain a fitness value of each solution; S13, taking the solutions in the population and their corresponding fitness values ​​as training data, and inputting them into the optimization module to train the optimization module, and training the optimization module on the mapping relationship between the solutions and their fitness values; S14, using the preliminarily trained optimization module to predict the fitness value of each solution in the population, and the optimization module automatically adjusts the selection rate, crossover rate and mutation rate of the input calculation module according to the prediction results; S15, after the initial data is input, the optimization module adaptively generates the selection rate, crossover rate and mutation rate and inputs them into the calculation module.

3. The method for optimizing the support cross section in a foundation pit based on a genetic algorithm according to claim 1, characterized in that: The calculation parameters include constraints, population size, number of iterations, selection rate, mutation rate and crossover rate. The population size and number of iterations are set manually, and the selection rate, mutation rate and crossover rate are adaptively generated through the optimization module.

4. The method for optimizing the support cross section in a foundation pit based on a genetic algorithm according to claim 1, characterized in that: In step S2, after the initial data is input, it is divided into two paths and enters the calculation module and the optimization module respectively. After the initial data enters the optimization module, the optimization module adaptively generates the selection rate, mutation rate and crossover rate according to the initial data, and inputs the selection rate, mutation rate and crossover rate into the calculation module. Then the calculation module calculates the entered initial data through the above selection rate, mutation rate and crossover rate.

5. The method for optimizing the support cross section in a foundation pit based on a genetic algorithm according to claim 3 is characterized by: The cross-sectional parameters of the inner support structure have the optimal weight or cost after model calculation under the premise of satisfying the constraint conditions, which include strength constraint, stiffness constraint and stability constraint; The requirements for strength constraints are as follows: s maX ≤f y In the formula, σ max Represents the ultimate stress value of the cross section of the relevant component, and the tensile and compressive design values ​​are fy=315MPa; The requirements for stiffness constraints are as follows: l max ≤[λ] In the formula, λ max is the maximum slenderness ratio of different components, [λ] is the allowable slenderness ratio; The requirements for stability constraints are as follows: Overall stability requirements for compression-bending members in the plane: In the formula, M x is the maximum bending moment in the compression-bending member, W1x is the gross section resistance moment, γ x is the cross-section plastic development coefficient, is the section coefficient in the x-axis direction, N ex is the standard axial force parameter, N is the axial force, A is the cross-sectional area, β x is the moment reduction factor; Calculated using Perry's formula: In the formula, α is the coefficient; Overall stability requirements out of plane for compression-bending members: In the formula, is the overall stability factor of the member under uniform bending moment, is the section coefficient in the y-axis direction, where the biaxially symmetrical steel section is: In the formula, is the slenderness ratio in the y direction; Local stability requirements: Where b1 is the flange width, t3 is the web thickness, and t1 is the flange thickness; Deflection constraint requirements: In the formula, l represents the span and max represents the maximum deflection of the structure; Purlin displacement constraint: IN max ≤30mm In the formula, max represents the maximum displacement of the structure.

6. The method for optimizing the support cross section in a foundation pit based on a genetic algorithm according to claim 5, characterized in that: The initial data in step S2 and the calculated data in step S36 are normalized when entering the optimization module to form standard data.

7. The method for optimizing support cross-section in a foundation pit based on a genetic algorithm according to claim 1, characterized in that: The optimization module is a neural network module.

Images