Intelligent Fiber Bending Structure Design Method Based on Adaptive Deformation Control

Abstract

The present invention proposes an intelligent fiber bending structure design method based on adaptive deformation control. The method includes: multi-dimensionally monitoring the geological environment where the integral bolt is located by arranging a high-precision sensor array and preprocessing the detected data; obtaining the overall target deformation amount corresponding to the environmental data at the current moment through weighted averaging of the data of each sensor, and adjusting the overall actual deformation amount based on the difference between the overall actual deformation amount and the overall target deformation amount; designing a resistance-increasing unit arranged in parallel, independently adjusting the resistance-increasing force according to local environmental changes, and when the bearing capacity of a certain fiber subunit approaches the limit, gradually transferring the load of this subunit to other fiber subunits that are not close to failure to prevent sudden global failure; integrating the optimized structure and adjusting the structure design optimization method based on real-time feedback. The present invention overcomes the deficiencies in the existing bolt design and improves the adaptability and reliability of the bolt.

Description

Technical Field

[0001] The present invention belongs to the technical field of adaptive deformation control, and particularly relates to a design method for an intelligent fiber bending structure based on adaptive deformation control. Background Art

[0002] With the development of the geotechnical engineering field, large deformation anchor bolts, as an important part of the support system, have been widely used in multiple fields such as mine exploitation, tunnel construction, and geological disaster prevention. The core function of large deformation anchor bolts is to provide support, bear the stress and displacement generated during the deformation of the rock mass, and ensure the stability of the rock mass. Existing large deformation anchor bolt technologies generally use steel. Although these technologies have played a certain role in some applications, there are still obvious defects under complex geological conditions.

[0003] First of all, the insufficient effect of resistance increase in force-bearing sub-units is a prominent problem. Current designs often adopt uniform structural parameters and material properties along the entire length of the anchor bolt, that is, the geometric shapes and mechanical properties (such as bending amplitude, period, and tensile properties, etc.) of each part of the anchor bolt are completely the same. This means that the anchor bolt cannot bear the load by sub-units according to the force requirements of different order units. The result of this uniform design is that when the rock mass undergoes large-scale deformation, the anchor bolt cannot effectively adapt to this deformation, resulting in low load-bearing efficiency. Especially under high stress conditions, the resistance increase effect of the anchor bolt is significantly insufficient, and the overall performance of the anchor bolt cannot be fully improved.

[0004] Secondly, the existing designs generally lack the characteristics of progressive failure. In traditional large deformation anchor bolts, once the fiber matrix material breaks, the fibers often suddenly all fail, showing sudden failure. This failure mode poses a potential risk to the stability and reliability of the project because when the anchor bolt fails, the supporting force will suddenly drop and cannot gradually release the load. This sudden failure behavior limits the adaptability and service life of the anchor bolt in complex environments. Especially in unstable geological environments such as soft rock and loose soil, the failure mode is particularly crucial.

[0005] Finally, the existing designs generally have limited adaptability in complex environments. With the in-depth development of geotechnical engineering projects, more and more projects face complex geological conditions, including rock strata with uneven hardness, high displacement environments, and dynamic load conditions. Existing anchor bolt design schemes are often limited to a single fiber bending method and lack strain adjustment and deformation control for these complex environments. The anchor bolt fails to well adapt to the requirements of local deformation when stressed, resulting in the inability of the anchor bolt to exert its due ability during the deformation process. Especially under high displacement conditions, the anchor bolt is prone to instability or premature failure.

[0006] In summary, although the existing fiber-reinforced bolt technology has provided a certain supporting capacity for geotechnical engineering, it still cannot provide sufficient adaptability and reliability under high stress, multi-stage unit deformation, and complex geological conditions. There is an urgent need for an innovative design solution to solve the above problems. Summary of the Invention

[0007] The object of the present invention is an intelligent fiber bending structure design method based on adaptive deformation control. By means of an intelligent control mechanism, progressive failure characteristics design, and adjustment of the load-bearing capacity of each unit, the deficiencies in the existing bolt design are effectively overcome, and the adaptability and reliability of the bolt are improved.

[0008] To achieve the above object, the present invention provides an intelligent fiber bending structure design method based on adaptive deformation control, and the method includes:

[0009] S1. Monitor the geological environment where the overall bolt is located in multiple dimensions through the arrangement of a high-precision sensor array and preprocess the detection data to obtain the environmental data at the current moment. ;

[0010] S2. Construct an adaptive weighted response prediction model based on the sensor array. By weighted averaging the data of each sensor, respond to different geological environment conditions to obtain the overall target deformation amount corresponding to the environmental data at the current moment. Based on the difference between the overall actual deformation amount and the overall target deformation amount, adjust the overall actual deformation amount to make the overall actual deformation amount closest to the overall target deformation amount. Corresponding overall target deformation amount, and adjust the overall actual deformation amount based on the difference between the overall actual deformation amount and the overall target deformation amount to make the overall actual deformation amount closest to the overall target deformation amount;

[0011] S3. Design multiple resistance increasing units arranged in parallel so that different regions can independently adjust the resistance increasing factor according to local environmental changes, so that the deformation amount of each resistance increasing factor gradually approaches the target deformation amount.

[0012] S4. When the load-bearing capacity of a certain fiber subunit approaches the limit, gradually transfer the load of this subunit to other fiber subunits that are not close to failure, and through progressive load distribution adjustment, realize the transition process of unit-by-unit failure to prevent the sudden occurrence of global failure.

[0013] S5. Integrate the optimized structure, collect real-time feedback, and adjust the design optimization method based on the real-time feedback.

[0014] Furthermore, the high-precision sensor array includes a stress sensor, a displacement sensor, a temperature and humidity sensor, and a crack monitoring sensor;

[0015] The preprocessing includes:

[0016] Collect the data of each sensor as the original data, and introduce a denoising method based on weighted mean to denoise the original data. Filter the original data using the weighted mean method with different weight coefficients according to the measurement accuracy and position of the sensors to remove unnecessary noise;

[0017] Use the normalization method to convert the denoised data of different sensors with different dimensions into standardized data;

[0018] Perform fusion processing on the standardized data to obtain the fused data as the environmental data at the current moment 。

[0019] Furthermore, in the fusion processing, an adaptive data fusion method based on weighted Kalman filtering is adopted. By establishing a Kalman filter model, the true values of the data of different sensors are estimated and fused.

[0020] Furthermore, the adaptive weighted response prediction model obtains the target deformation of the structure by the response prediction formula and by combining the weighted characteristics of the environmental data :

[0021] ;

[0022] where, is the target deformation, represents the deformation response prediction function of the th sensor, is the weight coefficient of the sensor , is the sensor at time acquired environmental data, is a response model trained based on historical data, and this function can predict the deformation response of the structure by combining various factors such as temperature and stress;

[0023] where, the weight coefficient is determined according to the accuracy of the sensor and the contribution degree of its data to the deformation prediction;

[0024] Adjust the overall actual deformation according to the difference between the overall actual deformation and the overall target deformation to make the overall actual deformation close to the overall target deformation. Specifically:

[0025] ;

[0026] where, is the correction value of the actual deformation, is the adjustment factor used to adjust the sensitivity of the model, is the The error of a sensor, is the actual measured value of the th sensor; is the variance of the measured value of the i-th sensor, which is used to measure the volatility or uncertainty of the measurement data of this sensor

[0027] Furthermore, the S3 specifically includes:

[0028] Parallelly arrange resistance increasing units within the structure, where each unit has an independent resistance increasing factor , and adjust the first resistance increasing force based on their respective local deformation amounts ; among them, the local deformation amount adjusts the resistance increasing factor according to the local static error of each resistance increasing unit.

[0029] Furthermore, introduce an environmental adaptability factor to adjust the resistance increasing factor of each resistance increasing unit according to real-time environmental data , to adapt to the local deformation differences in complex environments; among them, the environmental adaptability factor is calculated based on the environmental data at the previous moment and affects the resistance increasing factor in a weighted manner;

[0030] Then the change of the local deformation amount is calculated as follows:

[0031] For the th resistance increasing unit, its deformation adjustment formula is as follows:

[0032] ;

[0033] Among them, is the target deformation amount of the m-th segment, is the weight coefficient of the unit deformation, which controls the sensitivity of this resistance increasing unit, is its static error amount, that is:

[0034] ;

[0035] At the same time, design a regularization term to constrain the change range of the resistance increasing factor, so that it remains smooth during the adjustment process and avoid unreasonable mutations.

[0036] Furthermore, the S4 is specifically:

[0037] Construct a per-unit load control model to dynamically adjust the load of each subunit according to its current load-bearing capacity, and obtain the second load condition , representing the load condition of the subunit at the current moment;

[0038] Introduce environmental factors Adjust the changes in material strength and cross-sectional area, and correct the second load condition to obtain the third load condition ;

[0039] Based on the third load condition Construct a progressive failure control model to avoid failure by gradually transferring the load. When the third load condition approaches the failure load , avoid failure by gradually transferring the load; wherein, the progressive failure control model is through the load transfer factor , representing the load transfer ratio of the subunit corresponding to the third load condition, expressed as:

[0040] ;

[0041] Wherein, is the failure load of the subunit , when approaches , tends to 1, indicating that the load is completely transferred to other subunits. On the contrary, when is much lower than , tends to 0; wherein, is the current load condition;

[0042] According to the load transfer factor decide the load ratio borne by each subunit to obtain the fourth load condition , which is the final load of the unit after load adjustment. The load transfer factor controls the transition of the subunit load from the original load to the total load, ensuring that when a subunit approaches failure, the load can be smoothly transferred to other subunits.

[0043] Furthermore, the S4 further includes:

[0044] Real-time monitor the deformation amount , stress and strain of each subunit, and judge whether load reallocation is required. Specifically:

[0045] Introduce a failure risk factor to represent the failure probability of each subunit. Assume that the strain of a certain subunit changes at a rate of , the failure risk factor is calculated as follows:

[0046] ;

[0047] When reaches a certain threshold, it is considered that the subunit has a relatively high failure risk, and the load transfer mechanism is activated to reduce the load of the subunit;

[0048] Design an objective optimization strategy based on the goal of minimizing the total system deformation to optimize the effect of load transfer. The objective optimization strategy aims to minimize the deformation of each subunit:

[0049] ;

[0050] Among them, is the weight coefficient of the unit deformation, is the target deformation of the m-th segment, is the current local deformation, is the total number of units.

[0051] Furthermore, the S5 specifically includes:

[0052] Based on the fourth load condition , compare the target load under the ideal state to obtain the deviation between the current load and the target load, and design an intelligent adjustment factor , which determines the load adjustment speed and amplitude of each subunit, expressed as:

[0053] ;

[0054] Among them, is the adjustment coefficient, is the failure risk factor of the subunit;

[0055] Based on the deviation and the intelligent adjustment factor real-time adjust the structure design, including the dynamic optimization of material properties and geometric shapes;

[0056] Determine the deformation amount according to the initial bending length , laying period and bending amplitude of the three-unit fiber, determine the bearing capacity of the matrix based on the matrix and fiber, and determine the total elastic modulus of the anchor bolt based on the bearing capacity of the sub-unit matrix ;

[0057] Design the overall optimization objective function , comprehensively considering the load difference, failure risk, and performance after adjusting the structural design, is expressed as:

[0058] ;

[0059] Among them, is the weight of the load difference, is the weight of the failure risk, and the overall optimization objective function 's optimization objective is to balance the load and risk by minimizing to achieve uniform load distribution and efficient operation of the structure.

[0060] Furthermore, the dynamic optimization of the material properties and geometry specifically includes:

[0061] For the subunit with excessive load, increase the cross-sectional area of the unit or change the shape to improve its load-bearing capacity;

[0062] For the subunit with large load difference, adjust the strength of the material , increase its tensile strength and stiffness, and improve its load-bearing capacity.

[0063] The beneficial technical effects of the present invention are at least as follows:

[0064] (1) By introducing an intelligent control system, the present invention enables the bolt to adjust its deformation mode in real time. The real-time data in the geological environment (such as displacement, stress distribution, etc.) is collected through the environmental perception layer, and combined with the reinforcement learning algorithm to intelligently predict and adjust the bending amplitude, period of the fiber, and the laying method of the fiber, so as to achieve the adaptive adjustment of the bolt. This mechanism can not only ensure the stability and load-bearing capacity of the bolt in a complex environment, but also adjust the degree of fiber participation in bearing according to different deformation order units, avoiding the performance defects caused by a single design.

[0065] (2) To solve the problem of insufficient resistance increasing effect of sub-units in the prior art, the present invention rationally designs the bending amplitude and laying period of the fiber, so that during the stretching process of the bolt, the fiber participates in bearing unit by unit, gradually increasing the overall elastic modulus and enhancing the load-bearing capacity of the bolt. Especially during the process of the fiber being straightened unit by unit, the bolt shows the characteristic of gradually increasing resistance. In addition, based on the design of progressive failure characteristics, the present invention ensures that after the matrix of the bolt breaks, the fiber will not fail instantly, but release the load unit by unit, delaying the failure process and enhancing the anti-tensile ability and durability of the bolt.

[0066] (3) The design of the present invention takes into account the adaptability of anchor bolts under different geological conditions. By adjusting the laying period and bending amplitude of the fibers and combining with an intelligent control system to perceive and adapt to the geological environment in real time, the anchor bolt can effectively cope with challenges such as large deformation, high stress, and high displacement in the changing geotechnical environment. Especially in environments with high displacement and high stress such as soft rock and loose soil, the anchor bolt can maintain a high bearing capacity and stability, significantly improving the applicability and reliability of the anchor bolt in complex environments.

[0067] (4) The present invention adopts an intelligent feedback system. Through sensors, it real-time monitors the deformation and stress distribution of the anchor bolt, and combines with a reinforcement learning algorithm for closed-loop optimization to ensure that the fiber layout and deformation characteristics of the anchor bolt can be dynamically adjusted according to the actual situation during operation. This closed-loop system not only improves the adaptive ability of the anchor bolt, but also enhances the reliability and service life of the anchor bolt. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] The present invention will be further described with reference to the accompanying drawings. However, the embodiments in the drawings do not constitute any limitation to the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to the following drawings.

[0069] Figure 1 It is a flowchart of the intelligent fiber bending structure design method based on adaptive deformation control of the present invention.

[0070] Figure 2 It is a design schematic diagram of the overall structure of the anchor bolt and the distribution of bending fibers in the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0071] The embodiments of the present invention will be described in detail below. The examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals represent the same or similar elements or elements with the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention and should not be construed as a limitation to the present invention.

[0072] In one or more embodiments, as Figure 1 shown, a design method of an intelligent fiber bending structure based on adaptive deformation control is disclosed. The method includes the following steps S1 - S5:

[0073] S1. Monitor the geological environment of the overall anchor bolt in multiple dimensions by arranging a high-precision sensor array and preprocess the detection data to obtain the environmental data at the current moment .

[0074] Specifically, in this step, the present invention will conduct multi-dimensional monitoring of the geological environment where the anchor rod is located by arranging a high-precision sensor array to ensure comprehensive acquisition of key information such as rock mass stress, displacement, temperature and humidity, and possible crack propagation. Different sensors have different sensing capabilities. Therefore, they need to be arranged according to specific requirements and transmit data to the central data processing unit in real time through a wireless communication network.

[0075] Stress sensor: The present invention selects a high-precision fiber optic sensor to monitor the stress change of the surrounding rock mass in real time, especially for continuously monitoring the stress condition of the anchor rod.

[0076] Displacement sensor: A high-resolution laser displacement sensor is adopted to be able to obtain the deformation condition of the anchor rod in real time.

[0077] Temperature and humidity sensor: An environmental monitoring sensor is adopted to monitor the temperature and humidity of the environment in real time, which has an important impact on the durability and performance of fiber materials.

[0078] Crack monitoring sensor: Crack sensors are deployed at positions where crack propagation may occur to monitor the development trend of rock mass cracks in real time.

[0079] Furthermore, since the data collected by the sensors may be affected by environmental noise and errors, the present invention first needs to process the collected raw data to ensure its accuracy. In order to better adapt to complex environments, the present invention proposes an improved denoising and normalization method, which can effectively filter noise and balance the dimensional differences between various types of data.

[0080] Denoising processing:

[0081] To ensure the reliability of the data, the present invention introduces a denoising processing method based on weighted mean. Let the raw data collected by each sensor be , and each sensor has a different weight coefficient according to its measurement accuracy and position. The weighted mean method is used to filter the raw data to remove unnecessary noise. Assume that the present invention has sensor data, and the denoised values of these data can be expressed by the following formula:

[0082] ;

[0083] where is the denoised data of the th sensor at time , is the standard deviation of the th sensor, is the weight coefficient based on the standard deviation. Through this weighted mean method, the present invention can more accurately eliminate environmental interference and improve the reliability of data.

[0084] Data normalization:

[0085] To ensure the comparability of data from different types of sensors in subsequent calculations, the present invention uses a normalization method to convert data with different dimensions into standardized data. Through the maximum-minimum normalization method, each data point is within the range of [0,1], and the formula is as follows:

[0086] ;

[0087] where, is the normalized data, and are respectively the minimum and maximum values of the data of the -th sensor.

[0088] Furthermore, after aggregating the data from multiple sensors, the present invention needs to perform data fusion processing to ensure that the contribution of each data source can be reasonably reflected in the subsequent model. Considering the possible deviations and redundancies between sensors, the present invention adopts an innovative weighted Kalman filtering method for data fusion.

[0089] Weighted Kalman filtering:

[0090] To achieve optimal fusion in multi-source data, the present invention designs an adaptive data fusion method based on weighted Kalman filtering. By establishing a Kalman filtering model, the present invention can effectively estimate the true values of data from different sensors and fuse them. The specific Kalman filtering update process can be expressed as:

[0091] ;

[0092] where, represents the estimated state at time , is the measured value, is the measurement matrix, is the Kalman gain, is calculated as follows:

[0093] ;

[0094] where, is the estimated covariance at the previous moment, is the measurement noise covariance.

[0095] Through this method, the present invention can perform weighted fusion among various sensors, enhance the credibility of the entire data set, and reduce the impact caused by measurement errors of individual sensors.

[0096] Furthermore, finally, after denoising, normalization, and data fusion, the present invention transmits the processed data in real time to the control system and the design optimization module. In the entire system, the real-time nature of the data is crucial. Therefore, the present invention uses wireless data transmission technologies (such as 5G networks or Wi-Fi) to ensure the stability and low latency of data transmission.

[0097] The data at each moment will be used as the input for subsequent steps, supporting the adjustment and optimization of the adaptive deformation control model, and helping the anchor bolt to adjust its own shape and bearing capacity according to environmental changes.

[0098] The core task of Step 1 is to collect and process environmental data in real time through a high-precision sensor system, providing basic support for subsequent intelligent control and design optimization. Through innovative denoising processing, normalization, and weighted Kalman filtering methods, the present invention ensures the high precision and reliability of the data. Step 1 not only provides accurate data input for subsequent steps but also ensures that the system can dynamically respond to changes in complex environments through a real-time feedback mechanism, thus laying a solid foundation for the successful implementation of this patent solution.

[0099] S2. Construct an adaptive weighted response prediction model based on the sensor array, make responses to different geological environment conditions through weighted averaging of the data of each sensor, and obtain the environmental data at the current moment the corresponding overall target deformation amount, and adjust the overall actual deformation amount based on the difference between the overall actual deformation amount and the overall target deformation amount, so that the overall actual deformation amount is closest to the overall target deformation amount.

[0100] Specifically, the core task of this step is to design an adaptive control model based on environmental data (such as stress, displacement, temperature, etc.), enabling the intelligent fiber structure to effectively adjust its curvature under different environmental conditions, ensuring the bearing capacity and stability of the anchor bolt.

[0101] The present invention designs an innovative control model that takes into account the real-time changes in environmental factors and combines the preprocessing and fusion results of the previous data to form an adaptive feedback mechanism that can automatically adjust the deformation of the fiber structure according to changes in the external environment. Specifically, the model is based on the input environmental data , and adjusts the target deformation amount through a weighted adjustment mechanism .

[0102] Furthermore, according to the output data in the previous Step 1 , the present invention first needs to predict the deformation response of the current structure based on real-time data. Here, the present invention adopts an adaptive weighted response prediction model. The core of this model is to make a response to different geological environment conditions through the weighted average of the data of each sensor, and then determine the deformation amount of the target.

[0103] The present invention designs a response prediction formula to obtain the target deformation amount of the structure by combining the weighted characteristics of the environmental data :

[0104] ;

[0105] Wherein, is the target deformation amount of the m-th segment, represents the deformation response prediction function of the -th sensor, is the weight coefficient of the sensor , is the sensor at time acquired environmental data. is a response model trained based on historical data, and this function can predict the deformation response of the structure by combining various factors such as temperature and stress.

[0106] The weight coefficient is determined based on the accuracy of the sensor and the contribution degree of its data to the deformation prediction. The weight coefficient is updated by the following method:

[0107] ;

[0108] Wherein, is the variance of the -th sensor response function , used to quantify its contribution to the target deformation prediction.

[0109] Furthermore, after the prediction of the target deformation amount is completed, the present invention adjusts the actual bending amount of the fiber structure through a feedback mechanism to ensure that it can be adaptively adjusted to a predetermined target shape. To achieve this goal, the present invention proposes a feedback mechanism based on weighted adjustment, and this mechanism combines the difference between the target deformation amount and the actual deformation amount to adjust the control parameters of the model.

[0110] Specifically, the present invention adopts a weighted error correction model to make the actual bending amount closer to the target deformation amount , and its control formula is as follows:

[0111] ;

[0112] Among them, is the correction value of the actual deformation amount by the control system, is the adjustment factor used to adjust the sensitivity of the model, is the error of the th sensor, is the th actual measured value of the sensor. By weighted averaging the errors, the present invention can adaptively adjust the degree of deformation of the fiber to ensure the stable operation of the bolt under changing environmental conditions.

[0113] S3. Design multiple resistance increasing units arranged in parallel so that different regions can independently adjust the resistance increasing factor according to local environmental changes, thereby making the deformation amount of each resistance increasing factor gradually approach the target deformation amount.

[0114] In step S2, we established the feedback relationship between the target deformation amount and the actual deformation amount through the adaptive control model, ensuring precise adjustment based on environmental data. However, simply adjusting through the global resistance increasing factor is not sufficient to cope with the differences in local deformations in a complex environment. In the fiber structure, the deformation responses of different regions may be affected by local stress, temperature and other factors. Therefore, it is insufficient to control the overall deformation effect with only one unified resistance increasing factor. The goal of step 3 is to introduce a method of arranging multiple resistance increasing units in parallel so that different regions can independently adjust the resistance increasing factor to ensure that each unit can accurately respond to external environmental changes and achieve precise control of deformation.

[0115] As Figure 2 shown, in this first-order unit, we arranged resistance increasing units in parallel within the structure. Each unit has an independent resistance increasing factor ( ), and adjusts the resistance increasing force based on its respective local deformation amount . Finally, all the resistance increasing units act together, and their combined resistance increasing force is calculated by weighted averaging:

[0116] ;

[0117] Among them, is the weight factor, satisfying , and is used to balance the contributions of different resistance increasing units. For the th resistance increasing unit, its resistance increasing force is calculated as follows:

[0118] ;

[0119] Among them, is the resistance increasing factor of the -th resistance increasing unit, is the actual deformation amount of this unit, representing its deviation from the target deformation amount .

[0120] Furthermore, in order to adapt to local deformation differences in complex environments, we introduce an environmental adaptability factor , which can adjust the resistance increasing factor of each resistance increasing unit according to real-time environmental data (such as temperature, humidity, stress) . In this way, under the condition of changing environmental conditions, each resistance increasing unit can dynamically adjust its resistance increasing performance to ensure the deformation accuracy of the overall structure.

[0121] The calculation of the environmental adaptability factor needs to combine environmental data , such as temperature , humidity , etc., which affect the resistance increasing factor in a weighted manner. The specific adjustment formula is:

[0122] ;

[0123] Among them, is the basic resistance increasing factor, is the environmental adaptability factor of the -th resistance increasing unit, which considers the correction effect of current environmental data on the resistance increasing factor.

[0124] The calculation of the environmental adaptability factor is based on the weighted average of environmental factors:

[0125] ;

[0126] Among them, is the weight coefficient of the environmental factor , is the measured value of the -th environmental factor, is the maximum value of this environmental factor, ensuring the normalization of all environmental data.

[0127] Furthermore, based on the above resistance increasing design and environmental adaptability factor, we designed an adaptive control algorithm to make the deformation amount of each resistance increasing unit gradually approach the target deformation amount . The core idea of this algorithm is to adjust the resistance increasing factor according to the local static error of each resistance increasing unit, so as to achieve precise control.

[0128] For the A resistance increasing unit, whose deformation adjustment formula is as follows:

[0129] ;

[0130] wherein, is the adjustment factor, which controls the sensitivity of this resistance increasing unit, is its static error amount, that is:

[0131] ;

[0132] Through the static error correction mechanism, the system can adjust the resistance increasing force of each resistance increasing unit according to the local environmental influence, so as to ensure the overall deformation accuracy.

[0133] Furthermore, in order to avoid overcorrection and prevent the system from becoming unstable due to too large or too small resistance increasing factors, we designed a regularization term to constrain the change range of the resistance increasing factors, making it smooth during the adjustment process and avoiding unreasonable mutations.

[0134] The regularization term is designed as follows:

[0135] ;

[0136] wherein, is the regularization coefficient of the th resistance increasing unit, is the basic resistance increasing factor, is the resistance increasing factor after the current adjustment.

[0137] The regularization term can be used as a smoothing constraint term in the resistance increasing calculation formula, making the system more stable during the optimization process and avoiding overregulation.

[0138] S4. Design a unit-by-unit load bearing and progressive failure control strategy. When the load bearing capacity of a certain fiber subunit approaches the limit, gradually transfer the load of this subunit to other fiber subunits that are not close to failure, and through progressive load distribution adjustment, achieve the transition process of unit-by-unit failure and prevent the sudden occurrence of global failure.

[0139] Specifically, the core goal of this unit is: when the load bearing capacity of a certain fiber subunit approaches the limit, the system can gradually transfer the load of this subunit to other fiber subunits that are not close to failure, and through progressive load distribution adjustment, achieve the transition process of unit-by-unit failure and prevent the sudden occurrence of global failure.

[0140] Unit-by-unit load bearing control model:

[0141] The key to per-unit load control lies in dynamically adjusting the load according to the current load-bearing capacity of each subunit. Assume that the maximum load-bearing capacity of each subunit is , while its current actual load-bearing capacity is . This load-bearing capacity is associated with the stress and strain of the subunit. Using the relationship of elasticity theory, the actual load-bearing capacity of the subunit can be expressed as:

[0142] ;

[0143] where is the cross-sectional area of the th unit, is the yield limit of the material, and is the strain of the th unit. At this time, the system calculates the actual load-bearing capacity of each subunit by real-time monitoring the strain value of the subunit, thereby obtaining the second load condition , indicating the load condition of the subunit at the current moment.

[0144] Further, assume that each subunit is composed of different materials, and the mechanical properties of these materials may change during the load-bearing process. Therefore, and may also be adjusted according to the environment. To more accurately reflect this change, the present invention introduces an environmental factor to adjust the changes in material strength and cross-sectional area:

[0145] ;

[0146] where represents the dynamic adjustment coefficient of material properties based on environmental factors (such as temperature, humidity, etc.). By introducing this factor, the model can more accurately reflect the change in the load-bearing capacity of the subunit under actual operating conditions.

[0147] Progressive failure control model:

[0148] To avoid the impact of local failure on the entire structure, the present invention designs a progressive failure control mechanism. In a traditional structure, when a certain part reaches its maximum load-bearing capacity, sudden failure will occur. In the design of the present invention, the present invention avoids the sudden collapse of the system by gradually transferring the load. Set the failure load of each subunit. When approaches , the subunit will enter the progressive failure stage, and at this time, the load of the subunit will be gradually transferred.

[0149] To quantify this transfer process, the present invention introduces a load transfer factor , representing the load transfer ratio of subunit . The calculation method is as follows:

[0150] ;

[0151] Wherein, is the failure load of subunit . When is close to , the value of tends to 1, indicating that the load is completely transferred to other subunits. On the contrary, when is much lower than , tends to 0.

[0152] Furthermore, at each moment, the system determines the load ratio borne by each subunit according to the current load transfer factor . To achieve more precise load distribution, the present invention designs the following load adjustment formula:

[0153] ;

[0154] Wherein, is the total system load, while is the load of the th unit after load adjustment. The load transfer factor controls the transition of the subunit load from the original load ( ) to the total load ( ), ensuring that when a certain subunit is approaching failure, the load can be smoothly transferred to other subunits.

[0155] Furthermore, to improve the reliability of the system, the present invention also designs a per-unit failure prediction and dynamic adjustment mechanism. This mechanism determines whether load redistribution is needed by real-time monitoring of the deformation , stress and strain of each subunit. By predicting the strain change trend of each subunit, the system can identify possible failure risks in advance.

[0156] The present invention introduces a "failure risk factor" to represent the failure probability of each subunit. Assuming that the strain of a certain subunit changes at a rate of in one unit of time, the calculation formula of the failure risk factor is:

[0157] ;

[0158] When reaches a certain threshold, the system will consider that there is a high failure risk in this subunit and activate the load transfer mechanism to reduce the load on this subunit.

[0159] Furthermore, in order to optimize the effect of load transfer, the present invention introduces an objective optimization strategy based on minimizing the total deformation of the system. Set the objective function as the weighted sum of squares of the overall deformation of the system, and the goal is to minimize the deformation of each subunit:

[0160] );

[0161] wherein, is the weight coefficient of the unit deformation, is the target deformation amount expected for each subunit during load transfer, is the current actual deformation amount. By minimizing this objective function, the system can dynamically adjust the load distribution so that the deformation of each subunit is evenly distributed, thereby avoiding local overload and improving the stability of the structure.

[0162] S5. Integrate the optimized structure, collect real-time feedback, and adjust the design optimization method based on the real-time feedback.

[0163] Specifically, in step 4, the design of unit-by-unit load bearing and progressive failure control is completed. In order to further improve the performance and adaptability of the system, the present invention conducts design optimization and adjustment by introducing an intelligent feedback mechanism. The core purpose of this step is to dynamically optimize the load distribution, material distribution, geometric shape, etc. of the system based on real-time feedback, so that the structure can maintain the best stability and performance in a changing environment.

[0164] The input data comes from the design output obtained in step 4, mainly including:

[0165] The current load of each subunit ;

[0166] Failure risk factor ;

[0167] Load transfer factor .

[0168] In addition, considering the material and geometric properties of the three-unit progressive bending design, the present invention also needs to input the following parameters:

[0169] Total length of anchor bolts , matrix elastic modulus , fracture elongation ;

[0170] Elastic modulus of the fiber , initial bending amplitude , fiber laying period (where ), where the fiber laying period of Unit I is the shortest, that of Unit II is the second, and that of Unit III is the longest;

[0171] Tensile rate of the fibers in each unit , , , and satisfy ;

[0172] Among them, the present invention assumes that the behavior of the fiber after being straightened presents linear elasticity, and the fibers of the three units are in parallel with the same displacement.

[0173] Furthermore, the present invention establishes a feedback mechanism based on load difference and failure prediction. First, the present invention calculates the load difference of each subunit , that is, the deviation between the current load and the target load:

[0174] ;

[0175] Among them, is the target load, is the current load. Based on the load difference, the present invention designs an intelligent adjustment factor , which determines the load adjustment speed and amplitude of each subunit. The calculation formula of the adjustment factor is as follows:

[0176] ;

[0177] Among them, is the adjustment coefficient, is the failure risk factor. The subunit with a higher failure risk will increase the load adjustment amplitude by increasing the adjustment factor so as to quickly return to the target load state.

[0178] Furthermore, based on the calculated load difference and the adjustment factor , the feedback control mechanism will adjust the structural design according to the following two methods:

[0179] Geometric adjustment:

[0180] For the subunit with an excessive load, the system will increase the cross-sectional area of this unit or change its shape to improve the bearing capacity. The calculation formula of geometric adjustment is:

[0181] ;

[0182] Among them, is the geometric adjustment factor, which is used to control the adjustment ratio of the cross-sectional area. In this way, the present invention can effectively distribute the load and relieve the pressure on the overloaded part.

[0183] Material property adjustment:

[0184] For sub-units with large load differences, the system will dynamically adjust the strength of the material , and improve the tensile strength and stiffness. For example, assume that the material strength is adjusted according to the increase in load difference and failure risk, and the formula is as follows:

[0185] ;

[0186] Among them, is the material strength adjustment factor. Through the dynamic adjustment of this factor, it is ensured that the material in the area with a higher load can effectively provide additional bearing capacity.

[0187] Furthermore, the present invention combines the calculation of the deformation amount and the load, and further conducts design optimization.

[0188] Calculation of deformation amount: According to the initial bending length , laying period and bending amplitude of the three-unit fiber, the initial bending length of each unit can be approximated as:

[0189] ;

[0190] The stretching rate after straightening is calculated by the formula:

[0191] ;

[0192] Load calculation: The total bearing capacity of the anchor bolt is jointly borne by the matrix and the fiber. The bearing capacity of the matrix is:

[0193] ;

[0194] Among them, is the cross-sectional area of the matrix, is the ultimate stress of the matrix.

[0195] For the fiber unit , its bearing capacity is:

[0196] ;

[0197] Among them, is the cross-sectional area of the unit fiber, is the elastic modulus of the fiber, is the fiber strain.

[0198] Furthermore, the total elastic modulus of the anchor bolt gradually increases during the deformation process and is calculated based on the following formula:

[0199] ;

[0200] This calculation is based on the situation where the matrix and the fiber jointly bear the load. And during the process of the fiber being gradually straightened, the elastic modulus will gradually increase, thereby improving the load-bearing capacity and stiffness of the anchor bolt.

[0201] Furthermore, in order to ensure that the load distribution and design adjustment of the entire system are optimal, the present invention defines an optimization objective function , which comprehensively considers the load difference, failure risk, material properties (such as elastic modulus) after design adjustment, and the synergistic effect between the matrix and the fiber. By minimizing , the present invention aims to balance the load distribution, improve the structural stiffness, and ensure the minimization of the failure risk.

[0202] Specifically, the optimization objective function includes the following parts:

[0203] Minimization of the load difference

[0204] The load difference (i.e., the deviation between the current load and the target load) is the key to system optimization. The present invention hopes to make the load distribution tend to be uniform by adjusting the design parameters. The contribution of the load difference part is represented by the following term:

[0205] ;

[0206] Among them, is the weight of the load difference, is the subunit 's load difference.

[0207] Furthermore, the failure risk is calculated by real-time monitoring of the state of the subunit. To prevent the system from failing prematurely in some areas, the present invention takes the failure risk as the optimization object of the objective function to minimize the failure risk. The contribution of the failure risk part is represented by the following term:

[0208] ;

[0209] Among them, is the weight of the failure risk, is the subunit 's failure risk factor.

[0210] Furthermore, the above-mentioned synergistic effect between the matrix and the fiber on the elastic modulus of the anchor bolt has been calculated , especially the gradual increase in the elastic modulus when the fiber is gradually straightened. The optimization process should consider the change in the stiffness of the structure, which affects the overall load-bearing capacity. The adjustment of the elastic modulus will directly affect the load distribution and the overall stability of the structure. Therefore, the change in the stiffness of the structure needs to be comprehensively considered in the optimization objective function.

[0211] Combining these factors, the optimization objective function has the following final form:

[0212] ;

[0213] where is the weight of the elastic modulus adjustment, is the total calculated elastic modulus of the anchor bolt, and is the target elastic modulus. In this objective function, the load difference, failure risk, and optimization of the elastic modulus work together to ensure that the system is not only balanced in load distribution but also reaches the optimal state in terms of material properties and structural stiffness.

[0214] By minimizing , the present invention can achieve uniform load distribution, minimize the failure risk, and improve the stiffness of the overall system, thereby ensuring the optimal performance of the structure under various complex loads.

[0215] Furthermore, the feedback loop is the core of this optimization process. By continuously collecting feedback data and adjusting the structural design, the continuous improvement of the system performance is ensured. The specific process is as follows:

[0216] Calculate the load difference of each subunit .

[0217] Based on the load difference and failure risk, adjust the intelligent adjustment factors and geometric / material parameters.

[0218] Update the design and load distribution of each subunit.

[0219] Re-evaluate the load distribution and failure risk of the system, and transfer the feedback information back to the optimization algorithm as new input.

[0220] At each time step, this process will perform an optimization adjustment in real time. Through continuous feedback loops, each subunit of the structure can be dynamically optimized according to its load and failure risk, ultimately achieving the best stability and performance of the system.

[0221] In summary, through innovative technologies such as adaptive deformation control, unitized resistance increase, progressive failure design, and intelligent control mechanisms, the present invention has successfully overcome the deficiencies of existing large-deformation anchor technologies in terms of unitized resistance increase, failure modes, and adaptability. It not only improves the bearing capacity and reliability of the anchor, but also enables it to exhibit stronger adaptability and durability in complex geotechnical environments.

[0222] The above is the preferred embodiment of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements are also regarded as the protection scope of the present invention.

Claims

1. An intelligent fiber bending structure design method based on adaptive deformation control, characterized in that, The method includes: S1. Deploy a high-precision sensor array to conduct multi-dimensional monitoring on the geological environment where the integral bolt is located and preprocess the detection data to obtain the environmental data at the current moment ; S2. Construct an adaptive weighted response prediction model based on the sensor array, make responses to different geological environment conditions through weighted averaging of the data of each sensor, and obtain the environmental data at the current moment. The corresponding overall target deformation amount, adjust the overall actual deformation amount based on the difference between the overall actual deformation amount and the overall target deformation amount, so that the overall actual deformation amount is closest to the overall target deformation amount. S3. Design multiple resistance increasing units arranged in parallel, enabling different regions to independently adjust the resistance increasing factor according to local environmental changes, so that the deformation amount of each resistance increasing factor gradually approaches the target deformation amount; S4. Design a strategy for unit-by-unit load bearing and progressive failure control. When the load bearing capacity of a certain fiber subunit approaches the limit, gradually transfer the load of this subunit to other fiber subunits that are not close to failure, and through progressive load distribution adjustment, achieve the transition process of unit-by-unit failure to prevent the sudden occurrence of global failure; S5. Integrate the optimized structure, collect real-time feedback, and adjust the design optimization method based on the real-time feedback; Among them, the adaptive weighted response prediction model obtains the target deformation amount of the structure by combining the weighted features of environmental data through a response prediction formula : ； Among them, is the target deformation amount, represents the deformation response prediction function of the -th sensor, is the weight coefficient of the sensor , is the environmental data collected by the sensor at time , is a response model trained based on historical data, and this function can combine temperature and stress to predict the deformation response of the structure; Among them, the weight coefficient is determined based on the accuracy of the sensor and the contribution degree of its data to the deformation prediction; Adjust the overall actual deformation amount based on the difference between the overall actual deformation amount and the overall target deformation amount to make the overall actual deformation amount close to the overall target deformation amount. Specifically: ； Among them, is the correction value of the actual deformation amount, is the adjustment factor used to adjust the sensitivity of the model, is the error of the th sensor, is the actual measured value of the th sensor.

2. The intelligent fiber bending structure design method based on adaptive deformation control according to claim 1, wherein, The high-precision sensor array includes stress sensors, displacement sensors, temperature and humidity sensors, and crack monitoring sensors; The preprocessing includes: Collect the data of each sensor as the original data, introduce a denoising processing method based on weighted mean to denoise the original data, and use the weighted mean method to filter the original data according to different weight coefficients for different sensor measurement accuracies and positions to remove noise; Use the normalization method to convert the denoised data of different sensors with different dimensions into standardized data; Fuse the standardized data to obtain the fused data, which is used as the environmental data at the current moment 。 3. The method for designing an intelligent fiber bending structure based on adaptive deformation control according to claim 2, wherein In the fusion processing, adopt an adaptive data fusion method based on weighted Kalman filtering. By establishing a Kalman filter model, estimate the true values of different sensor data and fuse them.

4. The intelligent fiber bending structure design method based on adaptive deformation control according to claim 1, wherein The specific content of S3 includes: Arrange in parallel within the structure Resistance increasing units, each unit has an independent resistance increasing factor , and based on their respective local deformation Adjust the first resistance ; Wherein, the local deformation The resistance increasing factor is adjusted according to the local static error of each resistance increasing unit.

5. The method for designing an intelligent fiber bending structure based on adaptive deformation control according to claim 4, wherein Introduce environmental adaptability factors Adjust the resistance increasing factor of each resistance increasing unit according to real-time environmental data , to adapt to local deformation differences in complex environments; wherein, the environmental adaptability factor is calculated based on environmental data at the previous moment affects the resistance increasing factor in a weighted manner; Then the local deformation amount changes as calculated below: For the -th resistance increasing unit, its deformation adjustment formula is as follows: ； Among them, is a regulating factor that controls the sensitivity of the resistance increasing unit, is its static error amount, that is: ； Design the regularization term simultaneously Constrain the variation range of the resistance increasing factor to keep it smooth during the adjustment process and avoid unreasonable mutations.

6. The intelligent fiber bending structure design method based on adaptive deformation control according to claim 4, characterized in that The specific content of S4 is: Build a per-unit load control model to dynamically adjust the load of each subunit according to its current load-bearing capacity, and obtain the second load condition , representing the load condition of the subunit at the current moment; Introduce environmental factors Adjust the changes in material strength and cross-sectional area, correct the second load condition, and obtain the third load condition ; Based on the third load condition Construct a progressive failure control model to avoid failure by gradually transferring the load. When the third load condition approaches the failure threshold avoid failure by gradually transferring the load; wherein, the progressive failure control model is based on a load transfer factor , representing the load transfer ratio of the corresponding subunit for the third load condition and is expressed as: ； wherein, is the failure load of the sub-unit . When is close to , tends to 1, indicating that the load is completely transferred to other sub-units. On the contrary, when is much lower than , tends to 0; According to the load transfer factor determine the load ratio borne by each subunit to obtain the fourth load condition , which is the load of the unit after final load adjustment. The load transfer factor controls the transition of the subunit load from the original load to the total load, ensuring that when a certain subunit is approaching failure, the load can be smoothly transferred to other subunits.

7. The method for designing an intelligent fiber bending structure based on adaptive deformation control according to claim 6, wherein S4 also includes: Monitor the deformation of each subunit in real time , stress and strain , and determine whether load redistribution is required. Specifically, there are: Introduce a failure risk factor to represent the failure probability of each subunit. Assume that the strain of a certain subunit changes at a rate of within a unit time. Then the calculation formula for the failure risk factor is as follows: ； When When a certain threshold is reached, it is considered that there is a high risk of failure in this subunit, and the load transfer mechanism is activated to reduce the load on this subunit; Design an objective optimization strategy based on the goal of minimizing the total deformation of the system to optimize the effect of load transfer. The objective optimization strategy is The weighted sum of squares of the overall system deformation, with the goal of minimizing the deformation of each subunit: ； Among them, is the weight coefficient of the unit deformation, is the expected target deformation of each sub-unit in the later stage of load transfer, is the current actual deformation.

8. The design method of the intelligent fiber bending structure based on adaptive deformation control according to claim 6, characterized in that The specific content of S5 includes: Based on the fourth load condition , the deviation between the current load and the target load is obtained by comparing with the target load in the ideal state , and an intelligent adjustment factor is designed , which determines the load adjustment speed and amplitude of each subunit, and is expressed as: ； Among them, is the adjustment coefficient, and is the failure risk factor of the subunit; Based on deviation and intelligent adjustment factor Real-time adjustment of the structural design, including dynamic optimization of material properties and geometric shapes; According to the initial bending length of the three-unit fiber , laying period and bending amplitude Determine the deformation amount, determine the bearing capacity of the matrix based on the matrix and the fiber, and determine the total elastic modulus of the anchor bolt based on the bearing capacity of the sub-unit matrix ; Overall optimization objective function of the design , comprehensively considering the load difference, failure risk, and performance after adjusting the structural design, is expressed as: ； Among them, is the weight of the load difference, is the weight of the failure risk, and the overall optimization objective function has an optimization objective of balancing the load and risk by minimizing to achieve uniform distribution of the load and efficient operation of the structure.

9. The method for designing an intelligent fiber bending structure based on adaptive deformation control according to claim 8, characterized in that, The dynamic optimization of the material properties and geometric shapes specifically includes: For a subunit with excessive load, increase the cross-sectional area of the unit or change the shape to improve its load-bearing capacity; Adjust the strength of the material for the subunits with large load differences to increase their tensile strength and stiffness and improve their load-bearing capacity.

Images