A hydraulic control optimization method for a tunnel intelligent steel reinforcement trolley

Abstract

The application discloses a hydraulic control optimization method for a tunnel intelligent steel reinforcement trolley, and belongs to the technical field of control optimization, which comprises the following steps: constructing a tunnel intelligent steel reinforcement trolley hydraulic proportional control valve PID control system, introducing an improved freshwater snail optimization algorithm, adjusting the hydraulic proportional control valve PID control parameters, mapping the obtained three optimal control parameters to the tunnel intelligent steel reinforcement trolley hydraulic proportional control valve PID controller, and optimizing the hydraulic proportional control valve adjustment control quality. Based on the improved freshwater snail optimization algorithm, the dynamic response quality of the hydraulic proportional control system is relatively optimized, the system overshoot can be relatively effectively inhibited while the adjustment time is greatly shortened, the steady-state margin and the closed-loop robustness of the system are enhanced, and relatively reliable control guarantee is provided for improving the positioning accuracy, operation stability and construction safety of the tunnel intelligent steel reinforcement trolley.

Description

Technical Field

[0001] This invention belongs to the technical field of hydraulic control optimization, and particularly relates to a hydraulic control optimization method for a smart rebar trolley used in tunnels. Background Technology

[0002] As a key piece of mechanical equipment in underground engineering construction, the intelligent rebar trolley in tunnels undertakes tasks such as grabbing, transporting, positioning, and auxiliary installation. The hydraulic proportional control valve, as the core control element of the equipment's hydraulic system, enables precise drive of actuators such as the robotic arm, lifting mechanism, and telescopic cylinder. The performance of the hydraulic proportional control system directly affects the trolley's operational stability, positioning accuracy, and safety under complex tunnel conditions, serving as a crucial foundation for ensuring tunnel construction quality. However, in practical engineering applications, the control schemes for hydraulic proportional control valves currently mostly employ traditional proportional-integral-derivative (PID) control strategies, using methods such as the critical proportionality method and empirical trial-and-error to tune the controller parameters. However, tunnel construction environments face complex conditions such as sudden load changes, temperature rises, and oil contamination. Hydraulic systems themselves exhibit severe nonlinearity, time-varying characteristics, and hysteresis, generally resulting in limitations such as cumbersome debugging processes and limited control accuracy, potentially leading to problems like mechanical oscillations, large overshoot, and slow response speeds. Therefore, exploring more efficient parameter optimization strategies to achieve better control of the hydraulic proportional valve of the intelligent rebar trolley in tunnels is the main direction for current technological improvement, addressing the shortcomings of existing control technologies. Summary of the Invention

[0003] To overcome the technical problems described in the background section, this invention provides a hydraulic control optimization method for a tunnel intelligent rebar trolley. Based on an improved freshwater snail optimization algorithm, the dynamic response quality of the hydraulic proportional control system is relatively optimized. This method can significantly shorten the adjustment time while effectively suppressing system overshoot, enhancing the system's steady-state margin and closed-loop robustness, and providing a relatively reliable control guarantee for improving the positioning accuracy, operational stability, and construction safety of the tunnel intelligent rebar trolley.

[0004] The technical solution of this invention is: a hydraulic control optimization method for an intelligent steel reinforcement trolley in a tunnel, comprising the following steps:

[0005] S1. Construct a PID control system for the hydraulic proportional control valve of the intelligent steel bar trolley in the tunnel.

[0006] S2. An improved freshwater snail optimization algorithm is introduced. The specific improvement strategy is as follows:

[0007] S21. The chaotic flow field niche distribution and reverse landscape observation enhancement strategy introduced in the initialization stage enhances the distribution uniformity and search space coverage of the initial population.

[0008] S22. A buoyancy adjustment exploration strategy based on heavy-tailed random migration is introduced in the location update stage of the explorer subgroup. By simulating the strong jump characteristics and dynamic buoyancy disturbance of Levy's flight, the algorithm's ability to jump out of the local optimal habitat is enhanced.

[0009] S23. An adaptive gene mutation and information crossover development strategy is introduced in the evolutionary cycle of the developer subgroup. By dynamically adjusting the mutation factor and crossover probability, the convergence speed and control parameter refinement of the algorithm in the local search phase are improved.

[0010] S24. A collision-triggered freestep displacement correction strategy is introduced in the conflict avoidance phase after the individual position is updated to compensate for the adaptive displacement of individuals in dense areas, preventing the population from over-clustering and causing the search to stagnate.

[0011] S25. The golden ratio-driven precision foraging strategy introduced in the later stages of algorithm evolution refines the local search space and improves the final steady-state accuracy of PID parameter tuning.

[0012] S26. The deep survival stagnation monitoring and chaotic perturbation restart strategy introduced in the stagnation determination phase of the main loop enhances the algorithm's global exploration potential at the end of the convergence period.

[0013] S3. The improved freshwater snail optimization algorithm is used to optimize the PID control parameters of the hydraulic proportional control valve of the rebar trolley.

[0014] S4. Set the three optimal control parameters obtained from the optimization as the parameters of the PID controller of the hydraulic proportional control valve of the rebar trolley.

[0015] Furthermore, in the PID control system of the intelligent steel bar trolley hydraulic proportional control valve constructed in step S1, the improved freshwater snail optimization algorithm module is used to optimize the internal parameters of the hydraulic proportional control valve PID controller module. , , Offline optimization is performed to obtain the optimal combination of control parameters, which is then mapped to the PID controller module of the hydraulic proportional control valve. The hydraulic proportional control valve monitoring module collects the actual measured values ​​of the hydraulic proportional control valve in real time and transmits them to the hydraulic proportional control valve error calculation module. The hydraulic proportional control valve error calculation module receives a preset target value and compares the target value with the actual measured value to calculate the error signal. Then the error signal The input is sent to the PID controller module of the hydraulic proportional control valve. After parameter optimization, the PID controller module of the hydraulic proportional control valve adjusts the input based on the error signal. Real-time calculation of control quantity and control quantity The signal is sent to the hydraulic proportional control valve adjustment module to drive the hydraulic proportional control valve to complete the action.

[0016] Furthermore, the strategy for enhancing the ecological niche distribution of chaotic flow fields and reverse landscape observation in step S21 includes the following steps:

[0017] Step S211: Generate a uniformly distributed sequence of chaotic variables using the Logistic chaotic mapping iterative formula;

[0018] ,

[0019] In the formula This represents the next generation of chaotic variables after iteration. Represents the chaotic variables of the current generation. This represents the control parameter for the chaotic mapping and its value range is [0,4].

[0020] Step S212: Use the mapping formula to map the chaotic variable sequence to the search space of the PID parameters of the hydraulic proportional control valve to generate the initial freshwater snail individual position;

[0021] ,

[0022] In the formula This represents the initial position vector of the generated freshwater snail individuals. Represents the lower bound vector of the search space. Represents the chaotic variables of the current generation. This represents the upper bound vector of the search space for PID parameters in a hydraulic proportional control valve.

[0023] Step S213: Perform reverse landscape observation, generate the corresponding reverse landscape location vector, and select the first position from the initial position and reverse position in ascending order according to the fitness function value. The initial evolutionary population is composed of these positions;

[0024] ,

[0025] In the formula This represents the generated reverse landscape location vector. Represents the lower bound vector of the search space. Represents the upper bound vector of the search space. This represents the initial position vector of a freshwater snail individual.

[0026] Furthermore, the buoyancy regulation exploration strategy based on heavy-tailed random migration in step S22 includes the following steps:

[0027] Step S221: Utilize the Levy Flight Index and random vectors that follow a standard normal distribution Calculate the step size of heavy-tailed random migration ;

[0028] ,

[0029] In the formula This represents the Lévy step size for heavy-tailed random migration. Let represent a random vector that follows a standard normal distribution. Let represent a random vector that follows a standard normal distribution. Indicates the Levi Flight Index;

[0030] Step S222: Based on the current evolutionary cycle number With the maximum number of evolutionary cycles Calculate the dynamic buoyancy coefficient and combined with random numbers Branch decision logic calculates sine and cosine flow fluctuation factors ;

[0031] ,

[0032] In the formula Indicates the buoyancy coefficient in the current evolutionary cycle. This represents a random number within the closed interval [0,1]. This represents an exponential function with the natural logarithm as its base. This represents the buoyancy attenuation coefficient. Indicates the current evolutionary cycle number. Indicates the maximum total number of evolutionary cycles;

[0033] ,

[0034] In the formula Represents the sine and cosine flow fluctuation factor. Represents the sine function. Represents the cosine function. Represents pi (π). This represents the global optimal habitat location vector. This represents the vector representing the historical best position of an individual freshwater snail. This represents the L2 norm distance operation. This represents the function for calculating the vector mean. Represents the upper bound vector of the search space. Represents the lower bound vector of the search space;

[0035] Step S223: Use the position update formula to realize the global position evolution of the explorer subgroup;

[0036] ,

[0037] In the formula This represents the updated freshwater snail individual position vector corresponding to the explorer. This represents the current global optimal habitat location vector. Indicates the buoyancy coefficient in the current evolutionary cycle. This represents the Lévy step size for heavy-tailed random migration. This represents the explorer's position vector before the update. Indicates the water flow fluctuation factor. This represents the historical best position vector of a freshwater snail.

[0038] Furthermore, the adaptive gene variation and information cross-development strategy in step S23 includes the following steps:

[0039] Step S231: Monitor random numbers When it meets the preset update probability threshold, a mutation factor is generated within the preset closed interval. With cross probability ;

[0040] Step S232: Generate a mutation vector guided by the global optimal position using the formula. ;

[0041] ,

[0042] In the formula This represents the mutation vector of the developer subgroup. This represents the current global optimal habitat location vector. Indicates the variable factor. This represents the position vector of the first random individual within the developer subgroup. This represents the position vector of the second random individual within the developer subgroup;

[0043] Step S233: Perform information crossover operation, based on the random number and crossover probability. The proportional relationship determines the individual position of the developer. The source of the components is used to achieve feature recombination.

[0044] Furthermore, the collision-triggered freestep displacement correction strategy in step S24 includes the following steps:

[0045] Step S241: Calculate the Euclidean distance between individual freshwater snails. When the distance is less than the collision detection threshold... At that time, the freestep mechanism is triggered;

[0046] Step S242: Calculate the free step factor using distance deviation. ;

[0047] ,

[0048] In the formula Indicates the generated free step factor. This represents the function for calculating the vector mean. This represents the global optimal habitat location vector. This represents the L2 distance between the current location of the freshwater snail and the location of the globally optimal habitat. Indicates the size of the freshwater snail population. This represents the updated initial position vector of the freshwater snail;

[0049] Step S243: Execute the displacement correction formula to achieve the second-order discretization of the population distribution;

[0050] ,

[0051] In the formula This represents the updated position vector of the freshwater snail individual. This represents the generated freestep factor.

[0052] Furthermore, the golden ratio-driven precision foraging strategy in step S25 includes the following steps:

[0053] Step S251: Based on the golden ratio Calculate feature sampling points and ;

[0054] , ;

[0055] In the formula Indicates the first sampling point. Indicates the second sampling point. Represents pi (π). Represents the golden ratio;

[0056] Step S252: Use the golden sine formula to find the global optimal solution. Nearby refining locations ;

[0057] ,

[0058] In the formula This represents the generated refined position vector. This represents the global optimal habitat location vector. This represents the absolute value operation. Represents the sine function. Represents a random number within the interval [0, 2π]. Represents a random number within the interval [0, π]. As the first sampling point, This is the second sampling point. This represents the position vector of a randomly selected freshwater snail individual in the population.

[0059] Step S253: Execute the logic of selecting the smaller fitness function value, and update the refined position with the smaller fitness function value as the new global optimal position.

[0060] Furthermore, the deep survival stagnation monitoring and chaotic perturbation restart strategy in step S26 includes the following steps:

[0061] Step S261: Monitor the fitness function value of the global optimal position. If it is continuous... If no improvement occurs within a certain number of cycles, it is considered a survival stagnation.

[0062] Step S262: Sort the snails in ascending order according to their fitness function values ​​and lock the freshwater snail individuals at the bottom of the population ranking by a preset proportion.

[0063] Step S263: Use the restart formula to perform chaotic perturbation reset on the target individual's position;

[0064] ,

[0065] In the formula This represents the reset position vector of the freshwater snail. This represents the current global optimal habitat location vector. This represents the preset disturbance amplitude coefficient. It represents a random number within the closed interval [0,1].

[0066] Furthermore, a circular ecological boundary mapping step is included between steps S24 and S25, specifically including:

[0067] The positions of freshwater snails that are outside the search space [L,U] are corrected using modulo operations;

[0068] ,

[0069] In the formula This represents the updated position vector of the freshwater snail individual. This indicates a modulo operation performed on each component. Represents the lower bound vector of the search space. This represents the upper bound vector of the search space for PID parameters in a hydraulic proportional control valve.

[0070] Furthermore, the execution steps of the improved freshwater snail optimization algorithm in step S3 include:

[0071] Step A, Initialization Phase: Based on the logical mapping sequence, the initial population position of freshwater snails is generated in the PID parameter search space, and the corresponding reverse population is generated using the reverse landscape observation strategy. The initial evolutionary population is constructed by selecting the smaller fitness function value as the best.

[0072] Step B, Dynamic Grouping Phase: Based on the current evolutionary cycle number... Dynamically calculate the proportion of explorers The population is divided into an explorer subgroup that performs global searches and a developer subgroup that performs local development.

[0073] Step C, Explorer Evolution Stage: By combining Levy's flight step length with dynamic buoyancy coefficient and sine and cosine water flow fluctuation factors, the individual positions of the explorer subgroup are updated to enhance the global diffusion capability of the algorithm.

[0074] Step D, Developer Evolution Stage: Adaptive mutation and information crossover operations are performed on the developer subgroup to determine the recombination method of component features based on random probability, thereby improving the local search accuracy of the algorithm.

[0075] Step E, the correction and refinement stage, after the individual position is updated, the collision-triggered free step displacement correction, the ring ecological boundary mapping, and the precise foraging operation based on the golden ratio are executed in sequence to perform local refinement sampling of the current global optimal position.

[0076] Step F, Stagnation Monitoring and Restart Phase: Monitor the change in the fitness function value of the global optimal solution. If the survival stagnation criterion is met, perform chaotic perturbation reset on some low-concentration individuals until the maximum evolution period is reached, and then output the optimal PID parameter combination.

[0077] The beneficial effects of this invention due to the adoption of the above-mentioned technologies are as follows: This invention proposes a PID parameter tuning method for hydraulic proportional valves based on an improved freshwater snail optimization algorithm for intelligent steel reinforcement trolleys used in tunnels. Building upon the freshwater snail optimization algorithm, by introducing a chaotic flow field niche distribution and reverse landscape observation enhancement strategy during the initialization phase, the distribution uniformity of the initial population in the parameter search space is significantly optimized, laying a high-quality starting point for global optimization. During the algorithm evolution process, a buoyancy adjustment exploration strategy based on heavy-tailed random migration utilizes the strong jump characteristics of Levy flight, effectively enhancing the system's ability to escape local optimum traps. Combined with adaptive gene mutation and information cross-development strategies, the search step size and feature recombination probability are dynamically adjusted, greatly improving the convergence efficiency and parameter refinement during the local search phase. To address the search stagnation problem caused by over-aggregation of the population, a collision-triggered free-step displacement correction mechanism is introduced to achieve adaptive displacement compensation for individuals. This is combined with a precise foraging strategy driven by the golden ratio, which refines the sampling of the optimal solution neighborhood in the later stages of evolution, ensuring the final steady-state accuracy of the PID parameter tuning. Furthermore, the coupled application of deep survival stagnation monitoring and a chaotic perturbation restart strategy further enhances the algorithm's global exploration resilience at the end of convergence. The improved freshwater snail optimization algorithm relatively optimizes the dynamic response quality of the hydraulic proportional control system, significantly shortening the adjustment time while effectively suppressing system overshoot, enhancing the system's steady-state margin and closed-loop robustness, and providing a relatively reliable control guarantee for improving the positioning accuracy, operational stability, and construction safety of the intelligent steel reinforcement trolley in the tunnel. Attached Figure Description

[0078] Figure 1 This is a flowchart illustrating the present invention.

[0079] Figure 2 This is a comparison chart of the fitness function value curves of the freshwater snail optimization algorithm and the improved freshwater snail optimization algorithm in this invention.

[0080] Figure 3 This is a comparison chart of the step response curves of the freshwater snail optimization algorithm and the improved freshwater snail optimization algorithm in this invention.

[0081] Figure 4 These are the PID optimization parameters of the freshwater snail optimization algorithm and the improved freshwater snail optimization algorithm in this invention. A comparison chart of the change curves.

[0082] Figure 5 These are the PID optimization parameters of the freshwater snail optimization algorithm and the improved freshwater snail optimization algorithm in this invention. A comparison chart of the change curves.

[0083] Figure 6These are the PID optimization parameters of the freshwater snail optimization algorithm and the improved freshwater snail optimization algorithm in this invention. A comparison chart of the change curves. Detailed Implementation

[0084] Example 1: As Figure 1 As shown, the present invention provides a hydraulic control optimization method for a tunnel intelligent rebar trolley, comprising the following steps:

[0085] S1. Construct a PID control system for the hydraulic proportional control valve of the tunnel intelligent rebar trolley, including a hydraulic proportional control valve error calculation module, a hydraulic proportional control valve PID controller module, an improved freshwater snail optimization algorithm module, a hydraulic proportional control valve adjustment module, and a hydraulic proportional control valve monitoring module. Specifically, the improved freshwater snail optimization algorithm module is used to optimize the internal parameters of the hydraulic proportional control valve PID controller module. , , Offline optimization is performed to obtain the optimal combination of control parameters, which is then mapped to the PID controller module of the hydraulic proportional control valve. The hydraulic proportional control valve monitoring module collects the actual measured values ​​of the hydraulic proportional control valve in real time and transmits them to the hydraulic proportional control valve error calculation module. The hydraulic proportional control valve error calculation module receives a preset target value and compares the target value with the actual measured value to calculate the error signal. Then the error signal The input is sent to the PID controller module of the hydraulic proportional control valve. After parameter optimization, the PID controller module of the hydraulic proportional control valve adjusts the input based on the error signal. Real-time calculation of control quantity and control quantity Send to the hydraulic proportional control valve adjustment module to drive the hydraulic proportional control valve to complete the action;

[0086] S2. An improved freshwater snail optimization algorithm is introduced. The specific improvement strategy is as follows:

[0087] S21. The chaotic flow field niche distribution and reverse landscape observation enhancement strategy introduced in the algorithm initialization stage improves the starting quality of the algorithm's global optimization by enhancing the distribution uniformity of the initial population and the search space coverage.

[0088] S22. A buoyancy adjustment exploration strategy based on heavy-tailed random migration is introduced in the location update stage of the explorer subgroup. By simulating the strong jump characteristics and dynamic buoyancy disturbance of Levi's flight, the algorithm's ability to jump out of the local optimal habitat is enhanced.

[0089] S23. An adaptive gene mutation and information crossover development strategy is introduced in the evolutionary cycle of the developer subgroup. By dynamically adjusting the mutation factor and crossover probability, the convergence speed and control parameter refinement of the algorithm in the local search phase are improved.

[0090] S24. The collision-triggered freestep displacement correction strategy introduced in the conflict avoidance phase after individual position update prevents search stagnation caused by excessive population aggregation by adaptive displacement compensation for individuals in dense areas.

[0091] S25. The golden ratio-driven precision foraging strategy introduced in the later stage of algorithm evolution uses the golden sine formula to refine the local search space and improve the final steady-state accuracy of PID parameter tuning.

[0092] S26. The deep survival stagnation monitoring and chaotic perturbation restart strategy introduced in the stagnation determination phase of the main loop enhances the algorithm's global exploration potential at the end of convergence by performing chaotic reset on the bottom individuals with large fitness function values ​​at the end of the ascending sort.

[0093] S3. Using an improved freshwater snail optimization algorithm, the PID control parameters of the hydraulic proportional control valve in the PID control system of the intelligent rebar trolley in the tunnel are optimized and tuned to obtain the optimal control parameters. , , ;

[0094] S4. The three optimal control parameters obtained by using the improved freshwater snail optimization algorithm are set as the parameters of the PID controller of the hydraulic proportional control valve in the PID control system of the intelligent steel bar trolley hydraulic proportional control valve, thereby optimizing the regulation and control quality of the hydraulic proportional control valve.

[0095] The hydraulic proportional control valve regulating module is used to convert the control commands into proportional drive current to adjust the valve port diameter of the hydraulic proportional control valve. Its main function is to receive control commands from the hydraulic proportional control valve PID controller module. The built-in proportional amplifier circuit converts the weak level signal into a power current with pulse width modulation characteristics, drives the proportional electromagnet to generate controlled electromagnetic force, and then overcomes the resistance of the valve core reset spring to realize the continuous adjustment of the hydraulic valve opening, thus completing the physical conversion from weak electrical control signal to high-power hydraulic energy.

[0096] The hydraulic proportional control valve monitoring module is used to acquire the execution status data of the hydraulic proportional control valve in real time and feed the data back to the error calculation module to form a closed loop. It mainly includes a high-precision displacement sensor and a pressure transmitter, which captures the valve core displacement, inlet and outlet pressure and actuator stroke of the hydraulic proportional valve in real time. Through the signal conditioning circuit, filtering, linearization compensation and analog-to-digital conversion are performed to map the physical domain status information into a digital feedback signal synchronized with the control system, providing accurate real-time parameter support for the deviation calculation module.

[0097] Among them, the transfer function model of the hydraulic proportional control valve of the intelligent steel bar trolley in the tunnel is established. The corresponding mathematical expression is:

[0098] ,

[0099] In the formula This represents the transfer function of a hydraulic proportional control valve. Indicates the gain coefficient. The base of the natural logarithm. Indicates the lag time. Represents the Laplace operator. Representing the time constant, after system identification, The value is 1.2. The value is 0.15s. The value is 0.6s;

[0100] Correspondingly, the PID control law with filtering component... The mathematical expression is:

[0101] ,

[0102] In the formula This represents the controller transfer function. Represents the proportionality coefficient. Represents the integral coefficient. Represents the differential coefficient. Represents the Laplace operator. This represents the filter coefficient of the differential element and has a value of 0.02.

[0103] Simultaneously, a fitness function is constructed to measure the quality of PID control parameters. A smaller fitness function value indicates better control quality. The corresponding mathematical expression is:

[0104] ,

[0105] In the formula This represents the fitness function value, also known as fitness. The smaller the value, the better. The best fitness is called the optimal fitness. This represents the time integral index value of the absolute value of the error. This represents the weight of the time integral index of the absolute value of the error, and its value is 1.0. and These represent the constraint penalty weights for the overshoot constraint penalty term and the settling time constraint penalty term, respectively, and both have a value of 10.0. This represents the system overshoot deviation value. This indicates the adjustment time deviation value. This indicates that the energy weight is controlled and has a value of 0.5. This indicates the term for controlling energy loss;

[0106] Among them, the time integral index value of absolute error The mathematical expression is:

[0107] ,

[0108] In the formula This represents the time integral index value of the absolute value of the error. This represents the total simulation duration and is set to 10 seconds. Indicates the simulation time. This represents the real-time error between the given signal and the output signal.

[0109] Among them, the energy loss control item The mathematical expression for evaluating the smoothness of the control output is:

[0110] ,

[0111] In the formula This indicates the term for controlling energy loss. Represents the proportionality coefficient. Represents the integral coefficient. This represents the differential coefficient.

[0112] Furthermore, the steps for optimizing the hydraulic proportional control valve of the intelligent rebar trolley hydraulic control system using the improved freshwater snail optimization algorithm include:

[0113] Step 1, Initialization, includes the following steps:

[0114] Step 11: Set the basic parameters of the algorithm, including the freshwater snail population size. And the value is 30, dimension And the value is 3, the maximum evolutionary period And the value is 100, buoyancy attenuation coefficient And the value is 7, the collision detection threshold. And the value is 0.5, the golden ratio. And the value is 0.618, the stagnation limit. Furthermore, the value is 20 and the chaotic mapping parameter is 4;

[0115] Step 12: Perform chaotic flow field niche distribution initialization and generate chaotic variable sequences using Logistic chaotic mapping. Its iterative calculation formula is:

[0116] ,

[0117] In the formula Represents the chaotic variable after iteration. represents the current chaotic variable with an initial value of a random number between 0 and 1, and 4 represents the chaotic mapping control parameter;

[0118] Step 13: Use the mapping formula to convert the chaotic variable sequence Mapping to the search space to generate initial freshwater snail individual locations The mathematical expression is:

[0119] ,

[0120] In the formula This represents the initial position vector of a freshwater snail individual. This represents the lower bound vector of the search space and takes the value [0,0,0]. Represents the generated chaotic variables. This represents the upper bound vector of the search space and takes the value [6.41, 5.12, 0.96].

[0121] Step 14: Perform reverse landscape observation enhancement based on the initial freshwater snail individual locations. Calculate the reverse landscape location The corresponding mathematical expression is:

[0122] ,

[0123] In the formula This represents the generated reverse landscape location vector. Represents the lower bound vector of the search space. Represents the upper bound vector of the search space. This represents the initial position vector of a freshwater snail individual;

[0124] Step 15: Locate the individual freshwater snails. and reverse landscape location PID control parameter evaluation is performed, which involves calculating the fitness function value and selecting the top control parameters with relatively smaller fitness function values. These locations form an evolutionary population and determine the globally optimal habitat location. The initial position of each freshwater snail is assigned to its historical best position. ;

[0125] Step 2, the evolutionary cycle of multi-strategy coupling, specifically includes the following steps:

[0126] Step 21: Calculate the current evolutionary period The proportion of explorers below The corresponding mathematical expression is:

[0127] ,

[0128] In the formula This indicates the proportion of explorers in the current evolutionary cycle. This represents the initial exploration ratio and has a value of 0.5. This represents the final exploration percentage and has a value of 0.1. Indicates the current evolutionary cycle. Indicates the maximum evolutionary period;

[0129] Step 22: Sort the population in ascending order based on the fitness function values, and round up the real numbers before sorting. One freshwater snail individual was divided into the Explorer subgroup, and the remaining freshwater snail individuals were divided into the Developer subgroup.

[0130] Step 23: Perform heavy-tailed random migration of the explorer subgroup, using the Lévy Flight Index. Generate heavy-tailed random migration Levy step size The corresponding mathematical expression is:

[0131] ,

[0132] In the formula This represents the Lévy step size for heavy-tailed random migration. Let represent a random vector that follows a standard normal distribution. Let represent a random vector that follows a standard normal distribution. This indicates the Levi Flight Index, with a value of 1.5.

[0133] Step 24: Calculate the buoyancy coefficient for the current evolutionary cycle. The corresponding mathematical expression is:

[0134] ,

[0135] In the formula Indicates the buoyancy coefficient in the current evolutionary cycle. Represents a random number between 0 and 1. This represents the natural exponential function. This represents the buoyancy attenuation coefficient. Indicates the current evolutionary cycle. Indicates the maximum evolutionary period;

[0136] Step 25: When a random number is generated When less than or equal to 0.5, the water flow fluctuation factor The calculation formula used is:

[0137] ,

[0138] In the formula Indicates the water flow fluctuation factor. Represents the sine function. Represents pi (π). This represents the global optimal habitat location vector. This represents the vector representing the historical best position of an individual freshwater snail. This represents the L2 norm distance operation. This represents the mean calculation function. Represents the upper bound vector of the search space. Represents the lower bound vector of the search space;

[0139] Step 26: When a random number is generated When it is greater than 0.5, the water flow fluctuation factor The calculation formula used is:

[0140] ,

[0141] In the formula Indicates the water flow fluctuation factor. Represents the cosine function. Represents pi (π). This represents the global optimal habitat location vector. This represents the vector representing the historical best position of an individual freshwater snail. This represents the L2 norm distance operation. This represents the mean calculation function. Represents the upper bound vector of the search space. Represents the lower bound vector of the search space;

[0142] Step 27: Generate the updated location of the freshwater snail corresponding to the explorer. At this time, the position index The corresponding mathematical expression is:

[0143] ,

[0144] In the formula This represents the updated freshwater snail individual position vector corresponding to the explorer. This represents the global optimal habitat location vector. Indicates the buoyancy coefficient. This represents the Lévy step size for heavy-tailed random migration. This represents the explorer's position vector before the update. Indicates the water flow fluctuation factor. This represents the historical best position vector of a freshwater snail.

[0145] Step 28: Execute developer subgroup adaptive gene mutation, when random numbers... Parameter update is performed when the value is less than or equal to 0.2, and the mutation factor is... Randomly generated within a closed interval of 0.1 to 0.9, with crossover probability... The mutation vector is randomly generated within a closed interval of 0 to 1; otherwise, it retains the value from the previous evolutionary cycle. The mathematical expression is:

[0146] ,

[0147] In the formula This represents the mutation vector of the developer subgroup. This represents the global optimal habitat location vector. Indicates the adaptive variation factor. This represents the position vector of the first random individual within the developer subgroup. This represents the position vector of the second random individual within the developer subgroup. A random number between 0 and 1;

[0148] Step 29: Perform information cross-operation to determine the individual freshwater snail location vector corresponding to the developer. At this time, the position index Its selection rule is based on random numbers. Judgment and Following a uniform distribution between 0 and 1, when random numbers Less than the crossover probability When the developer is located, the mutation vector is taken. The weight, otherwise the developer's location will be the best historical location of the freshwater snail. Quantity;

[0149] Step 3, Position Correction and Refinement Output, specifically includes the following steps:

[0150] Step 31: Perform collision-triggered freestep displacement correction when the distance between individual freshwater snails is less than the threshold. Calculate the free step factor at time The corresponding mathematical expression is:

[0151] ,

[0152] In the formula Represents the free step factor. This represents the mean calculation function. This represents the global optimal habitat location vector. This represents the L2 distance between the current location of the freshwater snail and the location of the globally optimal habitat. Indicates population size, This represents the updated freshwater snail position vector;

[0153] Step 32: Utilize the free step factor After completing the displacement correction, the corresponding mathematical expression is:

[0154] ,

[0155] In the formula This represents the corrected position vector of a freshwater snail. Represents the free step factor. This represents the position vector of the freshwater snail before correction, where the position index is... ;

[0156] Step 33: Perform circular ecological boundary mapping, using modulo operations to correct the positions of individuals exceeding the boundary. The corresponding mathematical expression is:

[0157] ,

[0158] In the formula This represents the mapped and corrected position vector of a freshwater snail. This represents the modulo operation. Represents the lower bound vector of the search space. Represents the upper bound vector of the search space;

[0159] Step 34: Execute the greedy selection and global optimal update logic, which involves calculating the current position of the freshwater snail. The corresponding fitness function value is compared with the individual's historical best position. The corresponding historical best fitness function values ​​are compared numerically, and the one with the smaller fitness function value is selected as the new historical best position. After updating the positions of all individuals, the location with the smallest fitness function value is selected from the historical best locations of the entire population and defined as the new globally optimal habitat location. ;

[0160] Step 35: Perform precise foraging in the later stages of evolution and calculate the location of the first sampling point. Location of the second sampling point The corresponding mathematical expression is:

[0161] ,

[0162] In the formula Indicates the location of the first sampling point. Represents pi (π). This represents the golden ratio and has a value of 0.618.

[0163] ,

[0164] In the formula Indicates the location of the second sampling point. Represents pi (π). Represents the golden ratio;

[0165] Step 36: Generate refining positions based on the golden sine formula. The corresponding mathematical expression is:

[0166] ,

[0167] In the formula This represents the generated refined position vector. This represents the global optimal habitat location vector. This represents the absolute value operation. Represents the sine function. Represents 0 to Random numbers between Represents 0 to Random numbers between Indicates the location of the first sampling point. Indicates the location of the second sampling point. This represents the location vector of a freshwater snail individual randomly selected from the current population and different from the globally optimal habitat location;

[0168] Step 37: Perform stagnation monitoring and chaos restart, which means monitoring the global optimal fitness function value in real time during the evolution process. When the global optimal fitness function value is continuously... If the fitness function value does not decrease further within an evolutionary cycle, the algorithm is considered to have entered a search stagnation state. At this point, the individuals in the current population are sorted in ascending order according to their fitness function value, and the bottom 30% of individuals with relatively high fitness function values ​​are selected for a reset operation. This introduces diversity and encourages the algorithm to escape the local optimum habitat. The reset positions of the freshwater snail individuals are then determined. The mathematical expression is:

[0169] ,

[0170] In the formula This represents the reset position vector of the freshwater snail. This represents the global optimal habitat location vector. Represents a random number between 0 and 1;

[0171] Step 38: Determine the current evolutionary cycle Has the maximum evolutionary cycle been reached? If the current evolutionary cycle The maximum evolutionary period has not been reached. season And return to execute step 2 to enter the next evolutionary cycle, if the current evolutionary cycle Reaching the maximum evolutionary cycle When the time comes, the evolution stops and the process enters the result output stage;

[0172] Step 4: Output the results, which includes the following steps:

[0173] Step 41: Output the globally optimal habitat location As the final optimization result, the corresponding mathematical expression is:

[0174] ,

[0175] In the formula This represents the global optimal habitat location vector. This represents the optimization component corresponding to the scaling factor. This represents the optimization component corresponding to the integral coefficient. This represents the optimization component corresponding to the differential coefficient;

[0176] Step 42: Determine the global optimal habitat location The component mapping is used to obtain the optimal parameters of the PID controller for the hydraulic proportional control valve of the tunnel intelligent rebar trolley, and the corresponding mathematical expression is:

[0177] ,

[0178] In the formula Represents the proportionality coefficient. Represents the integral coefficient. Represents the differential coefficient. , , These represent the optimization components corresponding to the proportional coefficient, the integral coefficient, and the derivative coefficient, respectively.

[0179] Step 43: Use the optimal parameter combination obtained through optimization to control the hydraulic proportional control valve in real time, and complete the performance optimization and tuning of the intelligent steel bar trolley hydraulic control system for tunnels.

[0180] The hydraulic system of the intelligent rebar trolley in the tunnel controls the displacement of the piston in the hydraulic cylinder for rebar bending and forming by adjusting the opening of the proportional control valve. In a positioning task where the target displacement changes from 0mm to 100mm in a step, simulations were performed in Matlab to optimize and tune the PID control parameters of the hydraulic proportional control valve of the intelligent rebar trolley's hydraulic control system using both the freshwater snail optimization algorithm and an improved freshwater snail optimization algorithm. Figure 2 As shown, in the convergence curve of the optimal fitness function value, the improved freshwater snail optimization algorithm exhibits significantly better evolutionary pressure than the original freshwater snail optimization algorithm. Specifically, it achieves a step decrease in the fitness function value in the early stages of the first 10 iterations, rapidly converging from the initial state to a fitness function value of approximately 0.03, and finally reaching its optimal value after 100 iterations. =0.0279, while the freshwater snail optimization algorithm, in contrast, encountered significant search stagnation in the early stages of iterations (around generations 0-15), exhibiting premature convergence, and only achieving secondary convergence around generation 20, with the final fitness function value settling at 0.0455. Therefore, the improved freshwater snail optimization algorithm, through its optimized search mechanism, enhances global optimization capabilities, effectively avoiding local optima traps, and obtains a more accurate optimal solution in a shorter time; simultaneously, as Figure 3 As shown, in terms of dynamic characteristics, the improved freshwater snail optimization algorithm drives the system to rapidly approach the target value at approximately 0.5s and fully enter steady state at approximately 1s, while the original algorithm has a slower response speed and only stabilizes after 1.5s. The overshoot of the optimized system response is also smaller than that of the original freshwater snail optimization algorithm, exhibiting better damping characteristics. Furthermore, the curve corresponding to the original freshwater snail optimization algorithm shows obvious local fluctuations in the rising segment and slight oscillations before reaching steady state, indicating that its steady-state margin is relatively low. Therefore, it can be said that the system corresponding to the improved freshwater snail optimization algorithm has a faster response speed and stronger stability; while... Figures 4-6 As shown, observe , , The search trajectories of the three parameters over 100 iterations reveal that the parameter vector of the improved freshwater snail optimization algorithm essentially locks into the optimal range around generation 15, exhibiting extremely high numerical stability in subsequent iterations. The parameters of the freshwater snail optimization algorithm, however, show drastic fluctuations before generation 20, especially in... and The prolonged zero-value stagnation in the dimension reflects the loss of population diversity at a specific stage, and the improved freshwater snail optimization algorithm was ultimately determined to be... =4.0311 is relatively high, and it is matched with a smaller value. =0.1451, which greatly optimizes the system's damping ratio while ensuring zero steady-state error; In summary, the improved freshwater snail optimization algorithm is superior to the freshwater snail optimization algorithm in terms of optimization accuracy, convergence speed, and dynamic quality of the control system. It not only shortens the optimization calculation time but also significantly improves the robustness and response speed of the closed-loop system. The final optimized PID control parameter vector is: =[ , , =[2.6409,4.0311,0.1451], which can relatively improve the PID control quality of the hydraulic proportional control valve of the intelligent steel bar trolley hydraulic control system in the tunnel.

Claims

1. A hydraulic control optimization method for an intelligent steel reinforcement trolley used in tunnels, characterized in that: Includes the following steps: S1. Construct a PID control system for the hydraulic proportional control valve of the intelligent steel bar trolley in the tunnel. S2. An improved freshwater snail optimization algorithm is introduced. The specific improvement strategy is as follows: S21. The chaotic flow field niche distribution and reverse landscape observation enhancement strategy introduced in the initialization stage enhances the distribution uniformity and search space coverage of the initial population. S22. A buoyancy adjustment exploration strategy based on heavy-tailed random migration is introduced in the location update stage of the explorer subgroup. By simulating the strong jump characteristics and dynamic buoyancy disturbance of Levy's flight, the algorithm's ability to jump out of the local optimal habitat is enhanced. S23. An adaptive gene mutation and information crossover development strategy is introduced in the evolutionary cycle of the developer subgroup. By dynamically adjusting the mutation factor and crossover probability, the convergence speed and control parameter refinement of the algorithm in the local search phase are improved. S24. The collision-triggered freestep displacement correction strategy introduced in the conflict avoidance phase after individual position update compensates for the adaptive displacement of individuals in dense areas, preventing excessive population aggregation and search stagnation. S25. The golden ratio-driven precision foraging strategy introduced in the later stages of algorithm evolution refines the local search space and improves the final steady-state accuracy of PID parameter tuning. S26. The deep survival stagnation monitoring and chaotic perturbation restart strategy introduced in the stagnation determination phase of the main loop enhances the algorithm's global exploration potential at the end of the convergence period. S3. The improved freshwater snail optimization algorithm is used to optimize the PID control parameters of the hydraulic proportional control valve of the rebar trolley. S4. Set the three optimal control parameters obtained from the optimization as the parameters of the PID controller of the hydraulic proportional control valve of the rebar trolley.

2. The hydraulic control optimization method for a tunnel intelligent rebar trolley according to claim 1, characterized in that: In the PID control system of the intelligent rebar trolley hydraulic proportional control valve constructed in step S1, the improved freshwater snail optimization algorithm module is used to optimize the internal parameters of the PID controller module of the hydraulic proportional control valve. , , Offline optimization is performed to obtain the optimal combination of control parameters, which is then mapped to the PID controller module of the hydraulic proportional control valve. The hydraulic proportional control valve monitoring module collects the actual measured values ​​of the hydraulic proportional control valve in real time and transmits them to the hydraulic proportional control valve error calculation module. The hydraulic proportional control valve error calculation module receives a preset target value and compares the target value with the actual measured value to calculate the error signal. Then the error signal The input is sent to the PID controller module of the hydraulic proportional control valve. After parameter optimization, the PID controller module of the hydraulic proportional control valve adjusts the input based on the error signal. Real-time calculation of control quantity and control quantity The signal is sent to the hydraulic proportional control valve adjustment module to drive the hydraulic proportional control valve to complete the action.

3. The hydraulic control optimization method for a tunnel intelligent rebar trolley according to claim 1, characterized in that, The strategy for enhancing the ecological niche distribution of chaotic flow fields and reverse landscape observation in step S21 includes the following steps: Step S211: Generate a uniformly distributed sequence of chaotic variables using the Logistic chaotic mapping iterative formula; , In the formula Represents the next generation of chaotic variables after iteration. Represents the chaotic variables of the current generation. This represents the control parameter for the chaotic mapping and its value range is [0,4]. Step S212: Use the mapping formula to map the chaotic variable sequence to the search space of the PID parameters of the hydraulic proportional control valve to generate the initial freshwater snail individual position; , In the formula This represents the initial position vector of the generated freshwater snail individuals. Represents the lower bound vector of the search space. Represents the chaotic variables of the current generation. This represents the upper bound vector of the search space for PID parameters in a hydraulic proportional control valve. Step S213: Perform reverse landscape observation, generate the corresponding reverse landscape location vector, and select the first position from the initial position and reverse position in ascending order according to the fitness function value. The initial evolutionary population is composed of these positions; ， In the formula This represents the generated reverse landscape location vector. Represents the lower bound vector of the search space. Represents the upper bound vector of the search space. This represents the initial position vector of a freshwater snail individual.

4. The hydraulic control optimization method for an intelligent steel reinforcement trolley in a tunnel according to claim 1, characterized in that, The buoyancy regulation exploration strategy based on heavy-tailed random migration in step S22 includes the following steps: Step S221: Utilize the Levy Flight Index and random vectors that follow a standard normal distribution Calculate the step size of heavy-tailed random migration ; ， In the formula This represents the Lévy step size for heavy-tailed random migration. Let represent a random vector that follows a standard normal distribution. Let represent a random vector that follows a standard normal distribution. Indicates the Levi Flight Index; Step S222: Based on the current evolutionary cycle number With the maximum number of evolutionary cycles Calculate the dynamic buoyancy coefficient and combined with random numbers Branch decision logic calculates sine and cosine flow fluctuation factors ; ， In the formula Indicates the buoyancy coefficient in the current evolutionary cycle. This represents a random number within the closed interval [0,1]. This represents an exponential function with the natural logarithm as its base. This represents the buoyancy attenuation coefficient. Indicates the current evolutionary cycle number. Indicates the maximum total number of evolutionary cycles; ， In the formula Represents the sine and cosine flow fluctuation factor. Represents the sine function. Represents the cosine function. Represents pi (π). This represents the vector of the globally optimal habitat location. This represents the vector representing the historical best position of an individual freshwater snail. This represents the L2 norm distance operation. This represents the function for calculating the vector mean. Represents the upper bound vector of the search space. Represents the lower bound vector of the search space; Step S223: Use the position update formula to realize the global position evolution of the explorer subgroup; ， In the formula This represents the updated freshwater snail individual position vector corresponding to the explorer. This represents the current global optimal habitat location vector. Indicates the buoyancy coefficient in the current evolutionary cycle. This represents the Lévy step size for heavy-tailed random migration. This represents the explorer's position vector before the update. Indicates the water flow fluctuation factor. This represents the historical best position vector of a freshwater snail.

5. The hydraulic control optimization method for an intelligent steel reinforcement trolley in a tunnel according to claim 1, characterized in that, The adaptive gene variation and information cross-exploitation strategy in step S23 includes the following steps: Step S231: Monitor random numbers When it meets the preset update probability threshold, a mutation factor is generated within the preset closed interval. With cross probability ; Step S232: Generate a mutation vector guided by the global optimal position using the formula. ; ， In the formula This represents the mutation vector of the developer subgroup. This represents the current global optimal habitat location vector. Indicates the variable factor. This represents the position vector of the first random individual within the developer subgroup. This represents the position vector of the second random individual within the developer subgroup; Step S233: Perform information crossover operation, based on the random number and crossover probability. The proportional relationship determines the individual position of the developer. The source of the components is used to achieve feature recombination.

6. The hydraulic control optimization method for a tunnel intelligent rebar trolley according to claim 1, characterized in that, The collision-triggered freestep displacement correction strategy in step S24 includes the following steps: Step S241: Calculate the Euclidean distance between individual freshwater snails. When the distance is less than the collision detection threshold... At that time, the freestep mechanism is triggered; Step S242: Calculate the free step factor using distance deviation. ; ， In the formula Indicates the generated free step factor. This represents the function for calculating the vector mean. This represents the vector of the globally optimal habitat location. This represents the L2 distance between the current location of the freshwater snail and the location of the globally optimal habitat. Indicates the size of the freshwater snail population. This represents the updated initial position vector of the freshwater snail; Step S243: Execute the displacement correction formula to achieve the second-order discretization of the population distribution; ， In the formula This represents the updated position vector of the freshwater snail individual. This represents the generated freestep factor.

7. The hydraulic control optimization method for an intelligent steel reinforcement trolley in a tunnel according to claim 1, characterized in that, The golden ratio-driven precision foraging strategy in step S25 includes the following steps: Step S251: Based on the golden ratio Calculate feature sampling points and ; ， ； In the formula Indicates the first sampling point. Indicates the second sampling point. Represents pi (π). Represents the golden ratio; Step S252: Use the golden sine formula to find the global optimal solution. Nearby refining locations ; ， In the formula This represents the generated refined position vector. This represents the vector of the globally optimal habitat location. This represents the absolute value operation. Represents the sine function. Represents a random number within the interval [0, 2π]. Represents a random number within the interval [0, π]. As the first sampling point, This is the second sampling point. This represents the position vector of a randomly selected freshwater snail individual in the population. Step S253: Execute the logic of selecting the smaller fitness function value, and update the refined position with the smaller fitness function value as the new global optimal position.

8. The hydraulic control optimization method for a tunnel intelligent rebar trolley according to claim 1, characterized in that, The deep survival stagnation monitoring and chaotic perturbation restart strategy in step S26 includes the following steps: Step S261: Monitor the fitness function value of the global optimal position. If it is continuous... If no improvement occurs within a certain number of cycles, it is considered a survival stagnation. Step S262: Sort the snails in ascending order according to their fitness function values ​​and lock the freshwater snail individuals at the bottom of the population ranking by a preset proportion. Step S263: Use the restart formula to perform chaotic perturbation reset on the target individual's position; ， In the formula This represents the reset position vector of the freshwater snail. This represents the current global optimal habitat location vector. This represents the preset disturbance amplitude coefficient. It represents a random number within the closed interval [0,1].

9. The hydraulic control optimization method for a tunnel intelligent rebar trolley according to claim 1, characterized in that, Between steps S24 and S25, there is also a ring ecological boundary mapping step, which specifically includes: The positions of freshwater snails that are outside the search space [L,U] are corrected using modulo operations; ， In the formula This represents the updated position vector of the freshwater snail individual. This indicates a modulo operation performed on each component. Represents the lower bound vector of the search space. This represents the upper bound vector of the search space for PID parameters in a hydraulic proportional control valve.

10. The hydraulic control optimization method for an intelligent steel reinforcement trolley in a tunnel according to claim 1, characterized in that, The execution steps of the improved freshwater snail optimization algorithm in step S3 include: Step A, Initialization Phase: Based on the logical mapping sequence, the initial population position of freshwater snails is generated in the PID parameter search space, and the corresponding reverse population is generated using the reverse landscape observation strategy. The initial evolutionary population is constructed by selecting the smaller fitness function value as the best. Step B, Dynamic Grouping Phase: Based on the current evolutionary cycle number Dynamically calculate the proportion of explorers The population is divided into an explorer subgroup that performs global searches and a developer subgroup that performs local development. Step C, Explorer Evolution Stage: By combining Levy's flight step length with dynamic buoyancy coefficient and sine and cosine water flow fluctuation factors, the individual positions of the explorer subgroup are updated to enhance the global diffusion capability of the algorithm. Step D, Developer Evolution Stage: Adaptive mutation and information crossover operations are performed on the developer subgroup to determine the recombination method of component features based on random probability, thereby improving the local search accuracy of the algorithm. Step E, the correction and refinement stage, after the individual position is updated, the collision-triggered free step displacement correction, the ring ecological boundary mapping, and the precise foraging operation based on the golden ratio are executed in sequence to perform local refinement sampling of the current global optimal position. Step F, Stagnation Monitoring and Restart Phase: Monitor the change in the fitness function value of the global optimal solution. If the survival stagnation criterion is met, perform chaotic perturbation reset on some low-concentration individuals until the maximum evolution period is reached, and then output the optimal PID parameter combination.

Images