Design method of intelligent self-adaptive slope reinforcing anchor rod and anchor rod

Abstract

The invention discloses a design method of an intelligent self-adaptive side slope reinforcing anchor rod and the anchor rod, and relates to the field of side slope reinforcing engineering.The design method comprises the steps that the structure of the reinforcing anchor rod is defined based on the side slope reinforcing requirement, and the optimal control parameters of the reinforcing anchor rod structure are determined through the self-adaptive adjusting technology; establishing a label system based on the slope reinforcement environment information data, and establishing a geological parameter database according to the label system and the one-hot code; the operation state of the reinforcing anchor rod is evaluated on the basis of the geological parameter database and in combination with the neural network technology, and a reinforcing anchor rod stress controller is generated based on the operation state and the PID control technology; and determining a final design result of the reinforced anchor rod based on the reinforced anchor rod stress controller and the optimal control parameters. According to the anchor rod designed by the invention, real-time monitoring, dynamic adjustment and remote management of the anchor rod state can be realized, and the adaptability and the real-time monitoring capability of the anchor rod are remarkably improved.

Description

Technical Field

[0001] The invention relates to the field of slope reinforcement engineering, and in particular to a design method for an intelligent self-adaptive slope reinforcement anchor rod and an anchor rod. Background Art

[0002] Slope reinforcement is a vital engineering measure, whose purpose is to improve the stability of the slope, prevent the occurrence of geological disasters such as landslides, and thus protect people's lives and property. Anchors are common support structures in slope reinforcement projects. Their role is to enhance the stability of the slope. Slope reinforcement projects face complex geological conditions, such as the diversity of rock types, frequent changes in humidity, and temperature fluctuations. These factors will affect the effectiveness and stability of the anchors. In addition, slope reinforcement must also take into account the safety, economy, and environmental protection of construction. Therefore, higher requirements are placed on the anchor system. However, traditional anchor systems have some inherent limitations in design and application, which limits their ability to perform at their best under complex geological conditions.

[0003] Specifically, the design of traditional anchor rods is usually based on static geological parameters, lacks a response mechanism to dynamic changes in the geological environment, and cannot make timely adjustments according to changes in actual geological conditions. The collection and analysis of geological parameters are relatively independent, and there is a lack of effective data integration and sharing mechanisms, which makes it difficult for designers to fully grasp the changes in the geological environment. At the same time, parameters such as the length and thread density of the anchor rod usually rely on manual measurement and adjustment. This process is time-consuming, labor-intensive, and prone to errors. Existing anchor rod systems usually lack effective remote communication means and cannot achieve real-time data transmission and remote monitoring.

[0004] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0005] In view of the problems in the related art, the present invention proposes a design method and an anchor rod for intelligent adaptive slope reinforcement to overcome the above-mentioned technical problems existing in the existing related art.

[0006] To this end, the specific technical solution adopted by the present invention is as follows:

[0007] In a first aspect, the present invention provides a design method for an intelligent adaptive slope reinforcement anchor rod, the design method comprising:

[0008] The structure of the reinforcement anchor is defined based on the slope reinforcement requirements, and the optimal control parameters of the reinforcement anchor structure are determined using adaptive adjustment technology;

[0009] A labeling system is established based on slope reinforcement environmental information data, and a geological parameter database is established based on the labeling system and one-hot coding;

[0010] Based on the geological parameter database and combined with neural network technology, the operating status of the reinforcement anchor is evaluated, and the reinforcement anchor stress controller is generated based on the operating status and PID control technology;

[0011] The final design result of the reinforcement anchor is determined based on the reinforcement anchor stress controller and the optimal control parameters.

[0012] Preferably, the structure of the reinforcement anchor rod is defined based on the slope reinforcement requirement, and the optimal control parameters of the reinforcement anchor rod structure are determined by using the adaptive adjustment technology, including:

[0013] Based on the slope reinforcement requirements and modular processing technology, the outer protective shell of the reinforcement anchor is defined, and the positions of the core processor and the telescopic section are determined inside the outer protective shell;

[0014] A sensor is placed inside the outer protective shell to determine the structure of the reinforcement anchor rod, and the sensor is used to collect working information of the reinforcement anchor rod and environmental information within a preset working range to obtain sensor data;

[0015] The state of the reinforcement anchor is judged based on sensor data and fusion technology, and the optimal control parameters of the reinforcement anchor structure are generated based on the state of the reinforcement anchor and adaptive adjustment.

[0016] Preferably, judging the state of the reinforcement anchor rod based on sensor data and fusion technology, and generating and determining the optimal control parameters of the reinforcement anchor rod structure based on the state of the reinforcement anchor rod and adaptive adjustment include:

[0017] Set the initial weight of the sensor according to the accuracy and reliability information of the sensor, and define the variables for measuring data quality based on preset rules;

[0018] Update sensor weights based on variable definition results, perform sensor data fusion operations according to sensor weights and data fusion formulas, and transmit fused sensor data to a microprocessor;

[0019] The microprocessor receives the sensor data and adjusts the length and thread density of the telescopic section according to the sensor data and the control parameters of the reinforcement anchor;

[0020] A microprocessor and optimization algorithm are used to generate a control parameter population, and the tournament selection technology is used to select the optimal solution to adjust the control parameters of the telescopic section and optimize the functional state of the reinforced anchor structure.

[0021] Preferably, a control parameter population is generated by using a microprocessor and an optimization algorithm, and the control parameters of the telescopic section are selected with the optimal solution in combination with a tournament selection technique, and the functional state of the reinforced anchor structure is optimized, including:

[0022] Generate an initial population based on a random number generator, and integrate the control parameters of the reinforcement anchor into the initial population as a parameter individual structure to generate a control parameter population;

[0023] Define the fitness function and tournament parameters, and randomly select several groups of parameter individuals from the control parameter population to form a parameter individual set;

[0024] Calculate the fitness value of each parameter individual in the parameter individual set according to the fitness function, and select the parameter individual with the highest fitness value as the winner of the championship;

[0025] Repeatedly generate the control parameter population, select the tournament winner, and randomly select two groups of individuals from the winners to perform crossover operations according to the crossover probability to generate new parameter offspring individuals;

[0026] Perform mutation operation on parameter offspring individuals, introduce adaptive mutation probability to dynamically adjust the mutation probability, and mix parameter offspring individuals with parameter individuals in the control parameter population according to the mutation results;

[0027] The mixed results are combined with the adaptive termination conditions to generate the final control parameter population, and the parameter individual with the highest fitness in the final control parameter population is output as the optimal control parameter of the telescopic section. The optimal control parameter is used to optimize the functional state of the reinforced anchor structure.

[0028] Preferably, establishing a label system based on the slope reinforcement environmental information data, and establishing a geological parameter database according to the label system and the one-hot encoding includes:

[0029] Use geo-radar, borehole sampling and physical detection to collect underground structural environmental data at the slope reinforcement location to obtain geological parameters, and clean and standardize the geological parameters;

[0030] Based on the preset geological parameter classification rules, the geological parameters are divided into different geological types, the geological types are used as initial labels, and the initial labels are converted into numerical vectors using one-hot encoding to generate corresponding one-hot encodings;

[0031] After normalizing the numerical vector, the one-hot encoding and the numerical value are combined into a completed vector parameter to obtain a composite label, and the consistency of the composite label is verified;

[0032] Based on the verification results, it is determined whether there is a conflict in the geological parameters in the composite label, and the composite label is adjusted according to the judgment result, and the composite label that meets the requirements is stored in the database to generate a geological parameter database.

[0033] Preferably, the operation status of the reinforcement anchor rod is evaluated based on the geological parameter database in combination with the neural network technology, and the stress controller of the reinforcement anchor rod is generated based on the operation status and the PID control technology, including:

[0034] A configuration parameter acquisition model is constructed based on the geological parameter database and convolutional neural network technology, and the optimal configuration parameters of the expansion section in the reinforcement anchor are output according to the configuration parameter acquisition model;

[0035] Sensors are used to collect stress, displacement and vibration data of reinforcement anchors during the slope reinforcement process, and an evaluation model is established based on the collected data to evaluate the operating status of the reinforcement anchors;

[0036] An intelligent control strategy is designed based on geological parameters and the evaluation results of the operating status of the reinforcement anchor, and the intelligent control strategy is combined with PID control technology to generate a reinforcement anchor stress controller.

[0037] Preferably, using sensors to collect stress, displacement and vibration data of the reinforcement anchor rods during the slope reinforcement process, and establishing an evaluation model based on the collected data to evaluate the operating status of the reinforcement anchor rods includes:

[0038] The stress value and displacement change data of the reinforcement anchor rod at different positions during the slope reinforcement process are measured by sensors, and the vibration data of the reinforcement anchor rod is recorded by vibration measurement equipment;

[0039] The evaluation model is constructed by taking stress values, displacement change data and vibration data as input and stress distribution, deformation status and stability score of the reinforcement anchor as output;

[0040] The generalization ability and robustness of the evaluation model are analyzed based on the gradient descent method and cross-validation technology, and the evaluation model parameters are optimized according to the analysis results, and the evaluation function is defined based on the optimized model parameters;

[0041] The optimal hyperparameter combination is found according to the evaluation function and grid search method to determine the final evaluation model, and the evaluation model is used to evaluate the operating status of the reinforcement anchor during the slope reinforcement process.

[0042] Preferably, the expression of the evaluation model is:

[0043]

[0044] In the formula, S s Represents the stress distribution score, S d Indicates the deformation score, S t represents the stability score, σ i represents the stress at each point of the anchor, σ max Indicates the maximum stress, △L i represents the displacement of each point of the anchor rod, L max represents the maximum displacement, S v represents the vibration influence, and α represents the weight coefficient.

[0045] Preferably, designing an intelligent control strategy based on geological parameters and the evaluation results of the operation status of the reinforcement anchor rod, and combining the intelligent control strategy with the PID control technology to generate a reinforcement anchor rod stress controller includes:

[0046] The target prestress of the reinforcement anchor is obtained based on the geological parameters, and the actual prestress of the reinforcement anchor is analyzed according to the operating status of the reinforcement anchor, and the deviation between the target prestress and the actual prestress is determined;

[0047] The deviation results are used as the state variables of the prestressing of the reinforced anchor rods, and the PID parameters are randomly generated in combination with the PID control theory. The PID parameters, the optimal configuration parameters of the expansion section of the reinforced anchor rods and the geological parameters are used as individuals to generate the initial population.

[0048] Based on the tournament selection technique, the convergence state of the initial population is analyzed by calculating the average fitness, maximum fitness and minimum fitness of the initial population;

[0049] Adjust the adaptive crossover rate and adaptive mutation rate according to the convergence state, perform crossover operation based on the adaptive crossover rate and adaptive mutation rate to generate a new population and output the optimal individual;

[0050] The prestress control parameters of the reinforcement anchor rods during slope reinforcement are determined based on the optimal individual and a reinforcement anchor rod stress controller is generated.

[0051] In the second aspect, the present invention also provides an intelligent adaptive slope reinforcement anchor rod, which includes an external protective shell, a telescopic section is arranged at the bottom end of the external protective shell, a support rod is arranged inside the external protective shell, a sensor is arranged outside the support rod and inside the external protective shell, a closed cavity is arranged at the top end of the external protective shell, a core processor is arranged outside the closed cavity, and a wireless communicator is arranged inside the closed cavity.

[0052] The beneficial effects of the present invention are:

[0053] 1. The anchor rod designed by the present invention can realize real-time monitoring, dynamic adjustment and remote management of the anchor rod status. At the same time, a data-driven method is adopted to optimize the anchor rod design to ensure its safety and reliability under complex geological conditions. By introducing intelligent and adaptive technologies, the adaptability and real-time monitoring capabilities of the anchor rod are significantly improved, and the control accuracy and response speed of the anchor rod are enhanced. Through data-driven design optimization and intelligent decision-making, not only the design efficiency and reliability are improved, but also a closed-loop optimization process is formed to continuously iterate and improve the design level.

[0054] 2. The present invention adopts a modular design concept, which enables the anchor rod to flexibly adapt to different geological conditions. It not only allows the function of the anchor rod to be expanded by replacing or adding modules, but also provides the advantage of easy maintenance and upgrading, which helps to reduce long-term operating costs. Through the built-in telescopic section and microprocessor, the anchor rod can automatically adjust its length and prestress according to the changes in geological conditions monitored in real time, so as to maintain optimal performance in a dynamic environment, thereby greatly improving the applicability and reliability of the anchor rod under complex geological conditions.

[0055] 3. The present invention uses deep learning technology to establish an optimization configuration model for anchor rod parameters. Different from the traditional manual adjustment method based on experience or limited data sets, the optimization configuration model is trained with a large amount of geological parameter data, and can automatically identify the complex relationship between geological conditions and anchor rod configuration, so that it can more accurately capture the important details of geological characteristics and generate the best anchor rod configuration parameters accordingly. At the same time, the PID controller parameters are optimized through the optimization algorithm, which further improves the control accuracy and response speed of the anchor rod, so that the anchor rod can maintain the best working state in a constantly changing environment, providing a more accurate and efficient solution for slope reinforcement projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0057] Figure 1 is a flow chart of a design method for an intelligent adaptive slope reinforcement anchor according to an embodiment of the present invention;

[0058] Figure 2 is a structural schematic diagram of an intelligent adaptive slope reinforcement anchor according to an embodiment of the present invention;

[0059] Figure 3 yes Figure 2 A partial enlarged view of point A in the middle.

[0060] In the figure:

[0061] 1. External protective shell; 2. Telescopic section; 3. Support rod; 4. Sensor; 5. Enclosed cavity; 6. Core processor; 7. Wireless communicator; 8. Mechanical connector; 9. Centering bracket; 10. Sheath pipe; 11. Anchor; 12. Power supplier; 13. Reinforced concrete slab; 14. Slope. DETAILED DESCRIPTION

[0062] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention and are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, ordinary technicians in the field should be able to understand other possible implementations and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0063] According to an embodiment of the present invention, a design method for an intelligent adaptive slope reinforcement anchor rod and an anchor rod are provided.

[0064] The present invention will now be further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to the design method of the intelligent adaptive slope reinforcement anchor rod according to the embodiment of the present invention, the design method includes:

[0065] Step S1, defining the structure of the reinforcement anchor rod based on the slope reinforcement requirements, and determining the optimal control parameters of the reinforcement anchor rod structure using adaptive adjustment technology;

[0066] Step S2, establishing a label system based on the slope reinforcement environmental information data, and establishing a geological parameter database according to the label system and one-hot encoding;

[0067] Step S3, evaluating the operating status of the reinforcement anchor rod based on the geological parameter database and the neural network technology, and generating the reinforcement anchor rod stress controller based on the operating status and the PID control technology;

[0068] Step S4, determining the final design result of the reinforcement anchor based on the reinforcement anchor stress controller and the optimal control parameters.

[0069] In this embodiment, when the structure of the reinforcement anchor is defined based on the slope reinforcement requirement and the optimal control parameters of the reinforcement anchor structure are determined using the adaptive adjustment technology, the outer protective shell of the reinforcement anchor can be defined based on the slope reinforcement requirement and the modular processing technology, and the positions of the core processor and the telescopic section can be determined inside the outer protective shell; sensors are placed inside the outer protective shell to determine the structure of the reinforcement anchor, and the sensors are used to collect the working information of the reinforcement anchor and the environmental information within the preset working range to obtain sensor data; the state of the reinforcement anchor is judged based on the sensor data and the fusion technology, and the optimal control parameters for determining the structure of the reinforcement anchor are generated based on the state of the reinforcement anchor and the adaptive adjustment.

[0070] It should be explained that the reinforced anchor specifically includes: core processor 6, telescopic section 2, sensor 4, wireless communicator 7, power supplier 12, mechanical connector 8, external protective layer 1, etc., to achieve modular structure. The modular design allows the reinforced anchor to adapt to different geological conditions and working environments. By replacing or adding different modules, the function of the anchor can be expanded and its performance can be optimized. The modular design also includes the characteristics of easy maintenance and upgrading, which helps to reduce long-term operating costs.

[0071] The core processor 6 is the heart of the entire anchor rod. It includes key components such as the central processing unit (CPU), power management unit and main control circuit board. The core processor 6 is designed to be compact so that it can be installed at any position of the anchor rod. It is also waterproof and dustproof to ensure that it can work normally in harsh environments.

[0072] The telescopic section 2 allows the anchor rod to be adjusted in length according to actual needs after installation, so as to better adapt to different geological structures and load requirements. The telescopic section 2 contains a driving mechanism, which is controlled by the core processor 6 and can accurately adjust the length of the anchor rod after receiving instructions.

[0073] Sensors 4 are distributed at different positions of the anchor rod and are used to monitor the working status and surrounding environmental conditions of the anchor rod in real time, including but not limited to: pressure sensors to monitor the pressure changes on the anchor rod; temperature sensors to monitor the temperature changes around the anchor rod; humidity sensors to monitor the humidity changes around the anchor rod; acceleration sensors to monitor the vibration of the anchor rod and determine its stability; displacement sensors to monitor the displacement changes of the anchor rod and evaluate its fixing effect.

[0074] Wireless communicator 7 provides multiple wireless communication modes to ensure reliable communication means in different scenarios. Wireless communicator 7 is designed as an independent pluggable module, which can flexibly select the appropriate communication mode according to actual application requirements and is convenient for future upgrades.

[0075] The power supply 12 supports a built-in battery and an external power supply interface. The mechanical connector 8 ensures a safe and reliable connection between the various structures. The mechanical connector 8 is designed to be detachable for easy maintenance and replacement.

[0076] The outer protective layer 1 is made of high-strength material and has good anti-corrosion performance. For the part that needs to be buried in the soil layer, consider using materials with higher compressive strength.

[0077] In this embodiment, when judging the state of the reinforcement anchor based on sensor data and fusion technology, and generating the optimal control parameters for the reinforcement anchor structure based on the state of the reinforcement anchor and adaptive adjustment, the initial weight of the sensor can be set according to the accuracy and reliability information of the sensor, and the variables for measuring data quality can be defined based on preset rules; the sensor weight is updated based on the variable definition result, the sensor data fusion operation is performed according to the sensor weight and data fusion formula, and the fused sensor data is transmitted to the microprocessor; the microprocessor receives the sensor data, and adjusts the length and thread density of the telescopic section according to the sensor data and the control parameters of the reinforcement anchor; the microprocessor and the optimization algorithm are used to generate a control parameter population, and the tournament selection technology is combined to select the optimal solution to adjust the control parameters of the telescopic section, so as to optimize the functional state of the reinforcement anchor structure.

[0078] It should be explained that the sensor data fusion technology is used to comprehensively analyze the data of each sensor to more accurately determine the state of the anchor. The data fusion formula is S fused =w1.S1+w2.S2+…+w n S n Where S fused is the fused data, S1, S2, …, S n They are the raw data of each sensor, w1, w2, ..., w n is the weight coefficient, which represents the weight associated with the i-th sensor, and reflects the reliability and accuracy of the sensor.

[0079] In sensor data fusion, the weight coefficients w1, w2, ..., w n The determination of is a key step, which affects the accuracy and reliability of the final fusion result. n Initial weights are set based on the accuracy and reliability of the sensors, and then further optimized through adaptive adjustment and machine learning. The specific steps are as follows:

[0080] Step 1: Make a preliminary determination based on accuracy and reliability;

[0081] According to the accuracy and reliability information of each sensor, the sensor with higher accuracy and reliability is given a greater weight. The specific formula is as follows:

[0082]

[0083] In the formula, ω i Represents the weight of the i-th sensor, Reliability irepresents the reliability score of the ith sensor. Here, if reliability is measured by a certain error range or confidence level, a smaller error range or a higher confidence level corresponds to a higher reliability score. In practical applications, the reliability score is obtained by expert experience, historical data analysis, or through the sensor self-diagnosis mechanism. j represents the index variable, and n represents the total number of sensors.

[0084] Step 2: Further optimization based on adaptive weight adjustment;

[0085] Adaptive weight adjustment is a dynamic method that adjusts weights in real time according to the quality of sensor data. During the fusion process, the data quality of each sensor is continuously monitored and the corresponding weights are adjusted according to the quality. This method can make the system more flexible and better able to cope with changes in sensor performance.

[0086] Specifically, we first define a variable Quality to measure data quality. i (t), Quality i (t) represents the data quality score of the i-th sensor at time t, Quality i (t) can be measured based on a variety of factors, such as measurement error, data update frequency, consistency check results, etc.

[0087] Define a weight update rule that assigns weights based on the data quality score of each sensor. Specifically, the normalized quality score is used as the basis for the weight. The formula is as follows:

[0088]

[0089] In the formula, ω i (t) represents the weight of the i-th sensor at time t; Quality i (t) represents the data quality score of the i-th sensor at time t; j represents the index variable, and n represents the total number of sensors.

[0090] In this embodiment, when a control parameter population is generated by using a microprocessor and an optimization algorithm, and the control parameters of the telescopic section are selected to optimize the functional state of the reinforced anchor structure by combining the tournament selection technology, an initial population can be generated based on a random number generator, and the control parameters of the reinforced anchor are integrated into the initial population as a parameter individual structure to generate a control parameter population; a fitness function and tournament parameters are defined, and a number of groups of parameter individuals are randomly selected from the control parameter population to form a parameter individual set; the fitness value of each parameter individual in the parameter individual set is calculated according to the fitness function, and the parameter individual with the highest fitness value result is selected as The winner of the tournament; repeatedly generate the control parameter population to select the winner of the tournament, and randomly select two groups of individuals from the winner individuals to perform crossover operations according to the crossover probability to generate new parameter offspring individuals; perform mutation operations on the parameter offspring individuals, and introduce adaptive mutation probability to dynamically adjust the mutation probability, and mix the parameter offspring individuals with the parameter individuals in the control parameter population according to the mutation results; combine the mixing results with the adaptive termination conditions to generate the final control parameter population, and output the parameter individual with the highest fitness in the final control parameter population as the optimal control parameter of the telescopic section, and use the optimal control parameters to optimize the functional state of the reinforced anchor structure.

[0091] It should be explained that the microprocessor is responsible for receiving sensor data and dynamically adjusting the length and thread density of the anchor rod according to the data collected by the sensor. Specifically:

[0092] The length formula is: L adjusted =L initial +K p .e p +K i .∫e p dt+K d .de p / d t ;

[0093] The thread density formula is: D adjusted =D initial +K p .e d +K i .∫e d dt+K d .de d / d t ;

[0094] Where, L adjusted Indicates the adjusted anchor length; L initial Indicates the initial length; e p Indicates the error (i.e. the difference between the target length and the actual length); K p , K i and K dRespectively represent the proportional, integral and differential gains of the PID (proportional-integral-differential) controller; D adjusted is the adjusted thread density; D initial is the initial thread density; e d is the thread density error (i.e. the difference between the target thread density and the actual thread density).

[0095] In order to improve the control accuracy and response speed, the microprocessor uses an improved genetic algorithm to optimize the parameters of the PID controller. The specific steps are as follows:

[0096] Step 1: Generate an initial population P by using a random number generator. Each individual (chromosome) represents a set of PID controller parameters (K p , K i and K d ), the population size N can be determined according to the actual situation. Generally, a larger population size can achieve better search results.

[0097] Step 2: Define the fitness function to measure the performance of each individual. The fitness function can be the sum of the squares of the control errors, that is:

[0098]

[0099] In the formula, K p , K i and K d Respectively represent the proportional, integral and differential gains of the PID (proportional-integral-differential) controller, y t represents the target length, represents the actual length and T represents the time step.

[0100] Step 3: Implement tournament selection. Randomly select a number of individuals from the population each time, and select the individuals with the highest fitness to enter the next generation. The specific tournament selection method aims to efficiently select individuals with higher fitness from the current population as "parents" to breed the next generation of individuals, thereby improving the overall performance of the genetic algorithm. It is necessary to define a parameter k (tournament size), which is usually a small positive integer. From the current genetic algorithm population P, k individuals are randomly selected according to a uniform distribution to form the set C of participating individuals for the tournament.

[0101] According to the preset fitness function, calculate the fitness value of each individual in set C, and find the individual p with the highest fitness value from set C. * , and regard it as the winner of the tournament, this individual p * Will be selected as part of the next generation population or used in subsequent crossover and mutation operations.

[0102] Sampling and fitness evaluation are repeated until a sufficient number of individuals are selected to form the next generation of the population. This process ensures that the individuals selected in each generation are high-fitness individuals that have emerged from competition.

[0103] Step 4: Randomly select two individuals from the selected individuals for crossover operation to generate new offspring individuals, and dynamically adjust the crossover probability P according to the individual fitness c If the fitness gap in the population is large, reduce P c ; Otherwise it will increase.

[0104] Step 5: Perform mutation operations on the newly generated offspring individuals to increase the diversity of the population. Adaptive mutation probability is used to dynamically adjust the mutation probability according to the population diversity to avoid premature maturation.

[0105] Step 6: Mix some individuals in the old population with the new individuals to maintain population diversity, and introduce adaptive termination conditions. Dynamically adjust the termination conditions based on the fitness change rate ΔF. If ΔF is small, terminate early. The specific calculation formula for ΔF is as follows: ΔF = |F current -F previous ∣ / F previous , where F current and F previous Represent the average fitness values ​​of the current and previous iterations respectively.

[0106] Step 7: Output the individual with the highest fitness value in the current population as the optimal solution, that is, the optimal PID parameters.

[0107] At the same time, the microprocessor has a built-in storage module that can record the history of sensor data and control instructions, which is convenient for later data analysis and troubleshooting.

[0108] Specifically, in order to enhance the flexibility and reliability of communication, the anchor rod adopts a multi-mode communication protocol, supporting multiple wireless communication technologies such as Wi-Fi, Bluetooth, LoRa, Zigbee, NFC, and dedicated low-power wide area network (LPWAN), ensuring that the most suitable communication method can be found in different application scenarios.

[0109] In addition, the anchor rod has the function of adaptive data transmission frequency. The wireless communicator can dynamically adjust the transmission frequency according to the network status and the importance of the data. When changes in geological conditions are monitored, the system will automatically increase the data transmission frequency to ensure that the latest situation is reflected in a timely manner.

[0110] The adjustment formula is as follows: transmit_new =f transmit_base ×k adjust ; Among them, f transmit_new represents the adjusted transmission frequency, f transmit_baserepresents the basic transmission frequency, k adjust Indicates the adjustment coefficient, which can be set according to actual conditions (for example, when geological conditions change, k adjust >1).

[0111] The wireless communicator can send sensor data to the central server at a predetermined frequency, as shown in the following formula: transmit =1 / t interval ; Among them, f transmit represents the transmission frequency, t interval Indicates the interval time between two data transmissions.

[0112] In addition, in order to achieve energy consumption optimization, the anchor adopts intelligent energy-saving technology. When it is detected that the environment has not changed much within a period of time, the data transmission frequency is reduced or the sleep mode is entered to save energy. The specific formula is as follows: P active =P max ×(1-Δt / T threshold );where P active Indicates the power consumption in active mode, P max represents the maximum power, Δt represents the time interval for detecting environmental changes, T threshold Indicates the threshold time.

[0113] In this embodiment, when a label system is established based on slope reinforcement environmental information data, and a geological parameter database is established based on the label system and one-hot coding, address radar, drilling sampling and physical detection methods can be used to collect underground structure environmental data at the slope reinforcement location to obtain geological parameters, and the geological parameters are cleaned and standardized; based on the preset geological parameter classification rules, the geological parameters are divided into different geological types, the geological type is used as the initial label, and the initial label is converted into a numerical vector form using one-hot coding to generate the corresponding one-hot coding; after normalizing the numerical vector, the one-hot coding and the numerical value are combined into a completed vector parameter to obtain a composite label, and the consistency of the composite label is verified; based on the verification result, it is determined whether there is a conflict in the geological parameters in the composite label, and the composite label is adjusted according to the judgment result, and the composite label that meets the requirements is stored in the database to generate a geological parameter database.

[0114] It should be explained that data collection is a crucial task at all stages of slope reinforcement projects. Through on-site measurements and laboratory analysis of reinforcement anchors, researchers are able to obtain a series of key geological parameters, including hardness (H), moisture (W) and permeability (P). These parameters not only provide the basic physical properties of rocks, but also provide a scientific basis for predicting the behavior of rocks and their response to engineering measures. At the same time, underground structural information is collected using geological radar, borehole sampling and other geophysical detection methods, including: geological layer thickness, crack density and groundwater level.

[0115] At the same time, the collected data should be cleaned to remove outliers and missing values ​​to ensure data quality, and the geological parameters should be standardized or normalized. The formula is as follows:

[0116]

[0117] In the formula, x′ represents the standardized data; x represents the original data, min(x) and max(x) represent the minimum and maximum values ​​in the data set, respectively.

[0118] When establishing a data labeling system, it is necessary to assign corresponding geological condition classification labels to each set of geological data to facilitate subsequent data analysis and model training. This includes not only simple classification, but also feature extraction of more useful label information. The specific detailed steps are as follows:

[0119] Step 1: Simple classification of geological conditions;

[0120] Based on engineering experience and practical knowledge, geological data are divided into different categories according to the main geological parameters of rock type, hardness, humidity, and permeability. For specific reference, please refer to the Technical Specifications for Building Slope Engineering, such as: Hardness (H) is divided into:

[0121] Very Hard Rock (VHR): Saturated uniaxial compressive strength of rock R c >60MPa;

[0122] Hard rock (HR): 30MPa <R c ≤60MPa;

[0123] Soft rock (SSR): 30MPa <R c ≤60MPa;

[0124] Soft rock (SR): 5MPa <R c ≤15MPa;

[0125] Very soft rock (VSR): R c ≤5MPa.

[0126] Step 2: Feature extraction and label expansion;

[0127] In addition to basic geological parameters, more features such as geological layer thickness, fracture density and groundwater level are extracted and used as part of the label.

[0128] (1) The thickness of a geological layer refers to the vertical thickness of a specific geological layer, which is usually determined by drilling sampling or geophysical exploration data. Specifically, it is: T = z top -z bottom ; where T represents the thickness of the geological layer, z topIndicates the depth of the top of the geological layer, z bottom Indicates the depth of the bottom of a geological layer.

[0129] (2) Crack density can be defined as the number of cracks per unit volume, specifically: D c =N c / V; where D c Indicates crack density; N c represents the number of cracks, and V represents the volume under investigation.

[0130] (3) Groundwater level refers to the height of the groundwater surface relative to the ground. This value can be obtained through water well measurement or other detection technology. Specifically: W = h ground -h water ; Where W represents the groundwater level; h ground Indicates the ground height, h water Indicates the height of the groundwater level.

[0131] At the same time, for the encoding of composite labels, since each label contains multiple attributes, these attributes need to be encoded separately and then combined into a complete vector, and then one-hot encoding needs to be used to convert the geological type into a numerical vector.

[0132] One-hot encoding is a commonly used encoding technique in machine learning and data processing. It is used to convert categorical variables (non-numeric attributes such as geological type in this example) into numerical vector form. This method can make it easier for computers to process this information and can be directly used in many machine learning algorithms.

[0133] In one-hot encoding, each category value is converted into a binary vector. The number of elements in the vector is equal to the total number of categories of the categorical variable. Each position in the vector corresponds to a possible category. For a specific data point, the value at the corresponding category position is 1, and the values ​​at all other positions are 0. This achieves the "one-hot" state, that is, only one position is "hot" (value 1), and the rest are "cold" (value 0).

[0134] The codes corresponding to the five categories of geological types are:

[0135] Very hard rock (VHR): [1, 0, 0, 0, 0];

[0136] Hard rock (HR): [0, 1, 0, 0, 0];

[0137] Softer rock (SSR): [0, 0, 1, 0, 0];

[0138] Soft rock (SR): [0, 0, 0, 1, 0];

[0139] Very soft rock (VSR): [0, 0, 0, 0, 1];

[0140] For numerical attributes, normalization is performed to unify the scale, and then the one-hot encoded and normalized numerical attributes are combined into a complete vector. Assume that the composite label is (HR, 120MPa, 0.8, 0.005, 10m, 0.8 / m^3, 8m), which correspond to hard rock, rock hardness, humidity, permeability coefficient, geological layer thickness, fracture density and groundwater level, respectively.

[0141] The consistency of the labels is checked to ensure that there are no obvious contradictions between different parameters under the same geological conditions. For example, if the rock hardness is high, the permeability is usually low, and vice versa. The label information is stored in a structured database (such as MySQL) for easy query and management.

[0142] In this embodiment, when evaluating the operating status of the reinforcement anchor rod based on the geological parameter database combined with the neural network technology, and generating the reinforcement anchor rod stress controller based on the operating status and PID control technology, a configuration parameter acquisition model can be constructed based on the geological parameter database and the convolutional neural network technology, and the optimal configuration parameters of the telescopic section in the reinforcement anchor rod can be output according to the configuration parameter acquisition model; sensors are used to collect the stress, displacement and vibration data of the reinforcement anchor rod during the slope reinforcement process, and an evaluation model is established based on the collected data to evaluate the operating status of the reinforcement anchor rod; an intelligent control strategy is designed based on the geological parameters and the evaluation results of the operating status of the reinforcement anchor rod, and the intelligent control strategy is combined with the PID control technology to generate a reinforcement anchor rod stress controller.

[0143] It should be explained that when constructing the configuration parameter acquisition model, a convolutional neural network (CNN) is used as the basic model, and on this basis, an attention mechanism integrating multi-scale and knowledge guidance is introduced. The specific complete model architecture is input and output;

[0144] The inputs include geological parameters, including rock type, rock hardness, moisture, permeability coefficient, thickness of geological layers, fracture density and groundwater level.

[0145] The output includes the optimal configuration parameters of the anchor rod, such as length and thread density.

[0146] Multi-scale attention layer: For each scale k i , calculate Q i , k i 、V i , and use w prior Conduct knowledge-guided attention computation.

[0147] Concatenation layer: Concatenate the results Z of each scale i Stitched together.

[0148] Fully connected layer: The output Z of the attention layer is fully connected to obtain the final anchor configuration parameters.

[0149] Specifically, we assume that the input feature matrix is ​​X∈R n ×d, where n is the number of samples and d is the number of features. Several receptive fields of different sizes are set, such as k1, k2, ..., k m , corresponding to different window sizes, each receptive field will calculate its own attention distribution and merge these distributions.

[0150] For each scale k i , first apply the linear transformation The details are as follows:

[0151]

[0152] Attention Distribution A i The calculation is as follows:

[0153]

[0154] Specifically, this embodiment uses the softmax function to convert the original attention score into a probability distribution to obtain the attention-weighted feature matrix Z i =A i .V i , A i represents the attention distribution; V i Represents the corresponding value matrix.

[0155] At the same time, assume that there is a prior importance weight vector w prior ∈R d , and multiply it with the attention distribution to enhance the weights of important features:

[0156]

[0157] In the formula, . represents element-by-element multiplication, using the improved attention distribution A i ′ to calculate the final feature matrix Z i ′, the specific formula is: Z i ′=A i ′V i .

[0158] The results of all scales are merged and knowledge-guided attention is applied to each scale, and the result of Z (Z means stitching the results of each scale together) is expressed as:

[0159] Z=[Z′1;Z′2;...;Z′ m ];

[0160] The attention mechanism that combines multi-scale and knowledge-guided can be expressed as:

[0161]

[0162] In the formula, A i ′ represents the attention distribution guided by knowledge; V i Represents the corresponding value matrix.

[0163] The gradient descent method is used to adjust the model parameters to minimize the prediction error. The loss function is defined as the mean square error (MSE):

[0164]

[0165] In the formula, l e represents the actual output of the e-th sample, represents the output predicted by the model, N is the number of samples, and K-Fold Cross Validation is used to evaluate the generalization ability and robustness of the model. The data set is divided into K subsets, and K-1 subsets are used for training each time, and the remaining 1 subset is used for validation to prevent overfitting.

[0166] Through the grid search method, we find the optimal combination of model hyperparameters, such as learning rate, batch size, number of hidden layer units, etc., and apply the Bayesian optimization algorithm to improve model performance. The objective function of Bayesian optimization is expressed as: * =argmin a f(a); where a * represents the hyperparameter vector, and f(a) represents the evaluation function.

[0167] In this embodiment, when using sensors to collect stress, displacement and vibration data of reinforcement anchor rods during the slope reinforcement process, and establishing an evaluation model based on the collected data to evaluate the operating status of the reinforcement anchor rods, sensors can be used to measure the stress values ​​and displacement change data of the reinforcement anchor rods at different positions during the slope reinforcement process, and the vibration data of the reinforcement anchor rods can be recorded through vibration measuring equipment; the stress values, displacement change data and vibration data are used as input, and the stress distribution, deformation status and stability score of the reinforcement anchor rods are used as output to construct an evaluation model; the generalization ability and robustness of the evaluation model are analyzed based on the gradient descent method and cross-validation technology, and the evaluation model parameters are optimized based on the analysis results, and the evaluation function is defined based on the optimized model parameters; the optimal hyperparameter combination is found according to the evaluation function and the grid search method to determine the final evaluation model, and the evaluation model is used to evaluate the operating status of the reinforcement anchor rods during the slope reinforcement process.

[0168] It should be explained that in the process of evaluating the stress distribution, deformation and stability of the anchor rod, it is necessary to collect the stress σ of the anchor rod during the slope reinforcement process. i , displacement ΔL i And vibration data:

[0169] Stress σ i Use sensors to measure stress values ​​at different locations of the anchor;

[0170] Displacement ΔL i Install displacement meters or other appropriate sensors to monitor displacement changes of the anchor bolts;

[0171] Vibration data can be recorded by accelerometers or other vibration measuring devices to record the vibration frequency, amplitude and other information of the anchor.

[0172] After preprocessing the data, including denoising, filtering and normalization, the collected data is used to establish an assessment model for anchor stress distribution, deformation and stability;

[0173] The model input is the stress, displacement and vibration data of the anchor.

[0174] The model output is the stress distribution, deformation and stability score of the anchor.

[0175] Specifically, the expression of the evaluation model is:

[0176]

[0177] In the formula, S s Represents the stress distribution score, S d Indicates the deformation score, S t represents the stability score, σ i represents the stress at each point of the anchor, σ max Indicates the maximum stress, △L i represents the displacement of each point of the anchor rod, L max represents the maximum displacement, S v represents the vibration influence, α represents the weight coefficient, which needs to be determined based on experimental data or expert experience, S v The larger the value, the more serious the vibration impact. t will decrease accordingly.

[0178] In this embodiment, when designing an intelligent control strategy based on geological parameters and the evaluation results of the operating status of the reinforcement anchor rod, and combining the intelligent control strategy with the PID control technology to generate a reinforcement anchor rod stress controller, the target prestress of the reinforcement anchor rod can be obtained based on the geological parameters, and the actual prestress of the reinforcement anchor rod can be analyzed according to the operating status of the reinforcement anchor rod, and the deviation between the target prestress and the actual prestress can be judged; the deviation result is used as the state variable of the reinforcement anchor rod prestress, and the PID parameters are randomly generated in combination with the PID control theory, and the PID parameters, the optimal configuration parameters of the expansion section of the reinforcement anchor rod and the geological parameters are used as individuals to generate an initial population; based on the tournament selection technology combined with the calculation of the average fitness, maximum fitness and minimum fitness of the initial population, the convergence state of the initial population is analyzed; the adaptive crossover rate and the adaptive mutation rate are adjusted according to the convergence state, and the crossover operation is performed based on the adaptive crossover rate and the adaptive mutation rate to generate a new population and output the optimal individual; based on the optimal individual, the prestress control parameters of the reinforcement anchor rod during the slope reinforcement are determined to generate a reinforcement anchor rod stress controller.

[0179] It should be explained that the adaptive function is designed according to the environmental changes of the slope, and the anchor prestress is adaptively adjusted to cope with different slope changes and environmental influences. An intelligent control system is designed to automatically adjust the anchor prestress according to the real-time monitored geological parameters to ensure that the anchor maintains the best working state in a changing environment. The control system is based on a PID controller. The specific steps are as follows:

[0180] Step 1: Real-time PID control;

[0181] State variable: Define the state variable e(t) of anchor prestress, that is, the deviation between the actual anchor prestress and the target prestress.

[0182] Control input: Define the control input u(t), which is the adjustment amount for the anchor prestress.

[0183] PID formula: According to PID control theory, the control input is calculated as follows:

[0184]

[0185] Where Kp, Ki and Kd represent proportional, integral and differential gains respectively. represents the derivative of the error with respect to time, which represents the rate of change of the error with time. It represents the cumulative error from 0 to t, representing the sum of past errors.

[0186] Step 2: Parameter optimization;

[0187] Combined with the adaptive crossover rate (C r ) and adaptive mutation rate (M r) is used as follows:

[0188] Randomly generate a set of PID parameters (K p , K i , K d ) as the initial population, the population size is set to W, and each individual consists of three parameters. In addition, each individual also needs to contain the geometric parameters of the anchor (such as length L, thread density D, etc.) and the geological parameters of the slope (such as hardness H, moisture W, permeability P, etc.).

[0189] Initialize the adaptive crossover rate C r and adaptive mutation rate M r Initial value of:

[0190]

[0191]

[0192] In the formula, C rmin represents the minimum value of the crossover rate, C rmax Indicates the maximum value of the crossover rate. rmin Represents the minimum value of the mutation rate, M rmax Indicates the maximum value of the mutation rate.

[0193] Define the fitness function F, such as the sum of squares of the control error: F = α1×Σ(e(t)^2) / T+α2×S s +α3S d +α4S t .

[0194] Among them, S s Represents stress distribution score; S d Indicates the deformation score; S t represents the stability score; α1, α2, α3, α4 represent weight coefficients; e(t) represents the deviation between the actual value and the target value of the anchor prestress, and T represents the time step.

[0195] Use tournament selection and other methods to select parent individuals to ensure that excellent individuals have a higher probability of being selected and then calculate the average fitness value f of the current population avg , the minimum fitness value f min And the maximum fitness value f max .

[0196] Dynamically adjust the adaptive crossover rate C according to the convergence degree of the population r , the formula is as follows: C r (g1) = C rmin +(C rmax -C rmin )×(f max -favg ) / (f max -f min );

[0197] Among them, f max is the maximum fitness of the population, f avg is the average fitness of the population, f min is the minimum fitness of the population, C rmin is the minimum value of the crossover rate, C rmax is the maximum value of the crossover rate, and g1 is the current number of iterations.

[0198] Dynamically adjust the adaptive mutation rate M according to the convergence degree of the population r , the formula is as follows: M r (g2) = M rmin +(M rmax -M rmin )×(1-(f max -f avg ) / ((f max -f min )));

[0199] Among them, M rmin is the minimum mutation rate, M rmax is the maximum value of the mutation rate, and g2 is the current number of iterations.

[0200] Adaptation crossover rate C r (g1) Perform a crossover operation to generate a new generation of population. The geometric parameters of the anchor and the geological parameters of the slope are considered during the crossover operation. Specifically, the parent individuals are selected from the current population according to the adaptive crossover rate C. r (g1) Select two parent individuals.

[0201] One or more intersection points are randomly selected between the selected parent individuals. For each parameter (including PID parameters, anchor geometry parameters and slope geological parameters), the intersection points can be different, and the parts of the parent individuals between the intersection points are exchanged to generate two new individuals.

[0202] For PID parameters (K p , K i , K d ), you can select an intersection point and exchange the part after this point. For the geometric parameters of the anchor (such as length L, thread density D, etc.), select the intersection point separately, because these parameters may have different ranges and importance; for the geological parameters of the slope (such as hardness H, moisture W, permeability P, etc.), select the intersection point separately, because these parameters directly affect the working environment and performance of the anchor.

[0203] Use adaptive mutation rate M r(g2) Perform mutation operation to increase the diversity of the population. The geometric parameters of the anchor and the geological parameters of the slope are also considered during the mutation operation. The specific adaptive mutation rate M r (g2) Individuals to be mutated are selected from the current population. For each parameter, the variation range is determined according to its range and importance. For example, PID parameters can be fine-tuned within a certain range, while geometric parameters and geological parameters require a larger variation range. For the selected mutated individuals, their parameter values ​​are randomly changed according to the variation range. For the PID parameters, a small offset can be randomly added within a certain range. For the geometric parameters of the anchor, their values ​​can be appropriately increased or decreased to explore different design options. For the geological parameters of the slope, random adjustments can be considered within a reasonable range to simulate different geological conditions.

[0204] The newly generated population is compared with the parent population, and individuals with higher fitness are selected to form a new generation of population. In the selection process, the stress distribution S of the anchor is considered. s , deformation condition S d And stability score S t , check whether the termination conditions are met, such as reaching the predetermined number of iterations, the fitness value reaching a certain threshold, or the population change being less than a certain threshold. If the termination conditions are met, the optimal solution is output; otherwise, the iteration continues. Therefore, by optimizing the PID controller parameters through the optimization algorithm, the control accuracy and response speed of the anchor rod are further improved, so that the anchor rod can maintain the best working state in a constantly changing environment, providing a more accurate and efficient solution for slope reinforcement projects.

[0205] like Figure 2 and Figure 3 As shown, according to another embodiment of the present invention, the present invention further provides an intelligent adaptive slope reinforcement anchor rod, the slope reinforcement anchor rod comprises an external protective shell 1, a telescopic section 2 is arranged at the bottom end of the external protective shell 1, a support rod 3 is arranged inside the external protective shell 1, a plurality of groups of sensors 4 are arranged outside the support rod 3 and inside the external protective shell 1, a closed cavity 5 is arranged at the top end of the external protective shell 1, the closed cavity 5 is closed by C30 concrete, a core processor 6 is arranged outside the closed cavity 5, and a wireless communicator 7 is arranged inside the closed cavity 5.

[0206] like Figure 2 and Figure 3As shown, the slope reinforcement anchor rod also includes a mechanical connector 8 in specific applications, and the telescopic section 2 is connected to the external protective shell 1 through the mechanical connector 8, a centering bracket 9 is arranged on the outer side of the support rod 3, the support rod 3 and the external protective shell 1 are connected through a sheath tube 10, an anchor 11 is arranged on the top of the sheath tube 10, and the closed cavity 5 is installed on the outer side of the anchor 11, a power supplier 12 is arranged inside the closed cavity 5, and the slope reinforcement anchor rod is installed on the slope surface 14 during slope reinforcement through a reinforced concrete slab 13 when in application.

[0207] In summary, with the help of the above technical solutions of the present invention, the anchor rod designed by the present invention can realize real-time monitoring, dynamic adjustment and remote management of the anchor rod state, and at the same time, a data-driven method is used to optimize the anchor rod design to ensure its safety and reliability under complex geological conditions. By introducing intelligent and adaptive technologies, the adaptability and real-time monitoring capabilities of the anchor rod are significantly improved, and the control accuracy and response speed of the anchor rod are enhanced. Through data-driven design optimization and intelligent decision-making, not only the design efficiency and reliability are improved, but also a closed-loop optimization process is formed, and the design level is continuously iterated to improve. The present invention adopts a modular design concept, so that the anchor rod can flexibly adapt to different geological conditions, thereby not only allowing the function of the anchor rod to be expanded by replacing or adding modules, but also providing the advantages of easy maintenance and upgrading, which helps to reduce long-term operating costs. Through the built-in telescopic section and microprocessor, the anchor rod can automatically adjust its length and prestress according to the changes in geological conditions monitored in real time, so as to maintain optimal performance in a dynamic environment, thereby greatly improving the applicability and reliability of the anchor rod under complex geological conditions.

[0208] The present invention uses deep learning technology to establish an optimization configuration model for anchor rod parameters. Different from the traditional manual adjustment method based on experience or limited data sets, the optimization configuration model is trained with a large amount of geological parameter data, and can automatically identify the complex relationship between geological conditions and anchor rod configuration, so that it can more accurately capture the important details of geological characteristics and generate the best anchor rod configuration parameters accordingly. At the same time, the PID controller parameters are optimized through the optimization algorithm, which further improves the control accuracy and response speed of the anchor rod, so that the anchor rod can maintain the best working state in a constantly changing environment, providing a more accurate and efficient solution for slope reinforcement projects.

[0209] In the present invention, unless otherwise clearly stipulated and limited, the terms such as "installation", "setting", "connection", "fixation" and "screw-on" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral one; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be the internal connection of two elements or the interaction relationship between two elements. Unless otherwise clearly defined, ordinary technicians in this field can understand the specific meanings of the above terms in the present invention according to the specific circumstances.

[0210] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A design method for an intelligent adaptive slope reinforcement anchor, characterized in that: The design approach includes: Define the structure of the reinforcement anchor based on the slope reinforcement requirements, and use adaptive adjustment technology to determine the optimal control parameters of the reinforcement anchor structure; A labeling system is established based on slope reinforcement environmental information data, and a geological parameter database is established based on the labeling system and one-hot coding; Based on the geological parameter database and combined with neural network technology, the operating status of the reinforcement anchor is evaluated, and the reinforcement anchor stress controller is generated based on the operating status and PID control technology; The final design result of the reinforcement anchor is determined based on the reinforcement anchor stress controller and the optimal control parameters.

2. The design method of an intelligent adaptive slope reinforcement anchor according to claim 1 is characterized in that: Defining the structure of the reinforcement anchor rod based on the slope reinforcement requirements and determining the optimal control parameters of the reinforcement anchor rod structure using the adaptive adjustment technology includes: Based on the slope reinforcement requirements and modular processing technology, the outer protective shell of the reinforcement anchor is defined, and the positions of the core processor and the telescopic section are determined inside the outer protective shell; A sensor is placed inside the outer protective shell to determine the structure of the reinforcement anchor rod, and the sensor is used to collect working information of the reinforcement anchor rod and environmental information within a preset working range to obtain sensor data; The state of the reinforcement anchor is judged based on sensor data and fusion technology, and the optimal control parameters of the reinforcement anchor structure are generated based on the state of the reinforcement anchor and adaptive adjustment.

3. The design method of an intelligent adaptive slope reinforcement anchor according to claim 2 is characterized in that: The method of judging the state of the reinforcement anchor rod based on sensor data and fusion technology, and generating and determining the optimal control parameters of the reinforcement anchor rod structure based on the state of the reinforcement anchor rod and adaptive adjustment includes: Set the initial weight of the sensor according to the accuracy and reliability information of the sensor, and define the variables for measuring data quality based on preset rules; Update sensor weights based on variable definition results, perform sensor data fusion operations according to sensor weights and data fusion formulas, and transmit fused sensor data to a microprocessor; The microprocessor receives the sensor data and adjusts the length and thread density of the telescopic section according to the sensor data and the control parameters of the reinforcement anchor; A microprocessor and optimization algorithm are used to generate a control parameter population, and the tournament selection technology is used to select the optimal solution to adjust the control parameters of the telescopic section and optimize the functional state of the reinforced anchor structure.

4. The design method of an intelligent adaptive slope reinforcement anchor according to claim 3 is characterized in that: The method of generating a control parameter population by using a microprocessor and an optimization algorithm, and selecting the optimal solution to adjust the control parameters of the telescopic section by combining the tournament selection technology, and optimizing the functional state of the reinforced anchor structure includes: Generate an initial population based on a random number generator, and integrate the control parameters of the reinforcement anchor into the initial population as a parameter individual structure to generate a control parameter population; Define the fitness function and tournament parameters, and randomly select several groups of parameter individuals from the control parameter population to form a parameter individual set; Calculate the fitness value of each parameter individual in the parameter individual set according to the fitness function, and select the parameter individual with the highest fitness value as the winner of the championship; Repeatedly generate the control parameter population, select the tournament winner, and randomly select two groups of individuals from the winners to perform crossover operations according to the crossover probability to generate new parameter offspring individuals; The parameter offspring individuals are mutated, and the adaptive mutation probability is introduced to dynamically adjust the mutation probability. The parameter offspring individuals are mixed with the parameter individuals in the control parameter population according to the mutation results. The mixed results are combined with the adaptive termination conditions to generate the final control parameter population, and the parameter individual with the highest fitness in the final control parameter population is output as the optimal control parameter of the telescopic section. The optimal control parameter is used to optimize the functional state of the reinforced anchor structure.

5. The design method of an intelligent adaptive slope reinforcement anchor according to claim 1 is characterized in that: The method of establishing a label system based on the slope reinforcement environmental information data and establishing a geological parameter database based on the label system and the unique hot coding includes: Use geo-radar, borehole sampling and physical detection to collect underground structural environmental data at the slope reinforcement location to obtain geological parameters, and clean and standardize the geological parameters; Based on the preset geological parameter classification rules, the geological parameters are divided into different geological types, the geological types are used as initial labels, and the initial labels are converted into numerical vectors using one-hot encoding to generate corresponding one-hot encodings; After normalizing the numerical vector, the one-hot encoding and the numerical value are combined into a completed vector parameter to obtain a composite label, and the consistency of the composite label is verified; Based on the verification results, it is determined whether there is a conflict in the geological parameters in the composite label, and the composite label is adjusted according to the judgment result, and the composite label that meets the requirements is stored in the database to generate a geological parameter database.

6. The design method of an intelligent adaptive slope reinforcement anchor according to claim 5 is characterized in that: The method of evaluating the operating state of the reinforcement anchor rod based on the geological parameter database and the neural network technology, and generating the reinforcement anchor rod stress controller based on the operating state and the PID control technology includes: A configuration parameter acquisition model is constructed based on the geological parameter database and convolutional neural network technology, and the optimal configuration parameters of the expansion section in the reinforcement anchor are output according to the configuration parameter acquisition model; Sensors are used to collect stress, displacement and vibration data of reinforcement anchors during the slope reinforcement process, and an evaluation model is established based on the collected data to evaluate the operating status of the reinforcement anchors; An intelligent control strategy is designed based on geological parameters and the evaluation results of the operating status of the reinforcement anchor, and the intelligent control strategy is combined with PID control technology to generate a reinforcement anchor stress controller.

7. The design method of an intelligent adaptive slope reinforcement anchor according to claim 6, characterized in that: The method of collecting stress, displacement and vibration data of the reinforcement anchor rods during the slope reinforcement process using sensors, and establishing an evaluation model based on the collected data to evaluate the operating status of the reinforcement anchor rods includes: The stress value and displacement change data of the reinforcement anchor rod at different positions during the slope reinforcement process are measured by sensors, and the vibration data of the reinforcement anchor rod is recorded by vibration measurement equipment; The evaluation model is constructed by taking stress values, displacement change data and vibration data as input and stress distribution, deformation status and stability score of the reinforcement anchor as output; The generalization ability and robustness of the evaluation model are analyzed based on the gradient descent method and cross-validation technology, and the evaluation model parameters are optimized according to the analysis results, and the evaluation function is defined based on the optimized model parameters; The optimal hyperparameter combination is found according to the evaluation function and grid search method to determine the final evaluation model, and the evaluation model is used to evaluate the operating status of the reinforcement anchor during the slope reinforcement process.

8. The method for designing an intelligent adaptive slope reinforcement anchor according to claim 7, characterized in that: The expression of the evaluation model is: In the formula, S s Represents the stress distribution score, S d Indicates the deformation score, S t represents the stability score, σ i represents the stress at each point of the anchor, σ max Indicates the maximum stress, △L i represents the displacement of each point of the anchor rod, L max represents the maximum displacement, S v represents the vibration influence, and α represents the weight coefficient.

9. The method for designing an intelligent adaptive slope reinforcement anchor according to claim 8, characterized in that: The design of the intelligent control strategy based on the geological parameters and the reinforcement anchor operation status evaluation results, and the combination of the intelligent control strategy with the PID control technology to generate the reinforcement anchor stress controller include: The target prestress of the reinforcement anchor is obtained based on the geological parameters, and the actual prestress of the reinforcement anchor is analyzed according to the operating status of the reinforcement anchor, and the deviation between the target prestress and the actual prestress is determined; The deviation results are used as the state variables of the prestressing of the reinforced anchor rods, and the PID parameters are randomly generated in combination with the PID control theory. The PID parameters, the optimal configuration parameters of the expansion section of the reinforced anchor rods and the geological parameters are used as individuals to generate the initial population. Based on the tournament selection technique, the convergence state of the initial population is analyzed by calculating the average fitness, maximum fitness and minimum fitness of the initial population; Adjust the adaptive crossover rate and adaptive mutation rate according to the convergence state, perform crossover operation based on the adaptive crossover rate and adaptive mutation rate to generate a new population and output the optimal individual; The prestress control parameters of the reinforcement anchor rods during slope reinforcement are determined based on the optimal individual and a reinforcement anchor rod stress controller is generated.

10. An intelligent adaptive slope reinforcement anchor rod, implemented based on the design method of the intelligent adaptive slope reinforcement anchor rod according to any one of claims 1 to 9, characterized in that: The slope reinforcement anchor includes an external protective shell, a telescopic section is arranged at the bottom end of the external protective shell, a support rod is arranged inside the external protective shell, a sensor is arranged outside the support rod and inside the external protective shell, a closed cavity is arranged at the top end of the external protective shell, a core processor is arranged outside the closed cavity, and a wireless communicator is arranged inside the closed cavity.

Images