Intelligent dynamic anti-loose control method and system for high-speed rail fastener

Abstract

The invention discloses an intelligent dynamic anti-loosening control method and system for a high-speed rail fastener, and the method comprises the steps: obtaining the working parameters of multiple dimensions, judging the loosening type of the high-speed rail fastener according to the working parameters of multiple dimensions, generating a pre-tightening force adjustment scheme, and achieving the correction of a fastening state. According to the intelligent dynamic anti-loose control method and system for the high-speed rail fastener, the accuracy and timeliness of risk identification are remarkably improved; the working parameters of the high-speed rail fastener are continuously received within the preset duration, and optimization and correction are continuously carried out; experiments prove that the pre-tightening force of the high-speed rail fastener can be kept at 95% or above of an initial value, and the loosening rate is reduced by 80%; in a cyclic vibration test, the fatigue life of the high-speed rail fastener is prolonged by 40%; compared with a traditional method, under the conditions of low pre-tightening force and high vibration, the anti-loosening performance is improved by 60% or above. Besides the field of high-speed rails, the control method can also be applied to the fields of bridges, aerospace and the like with high requirements on the anti-loosening performance of fasteners.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of fastener state processing, in particular to a high-speed rail fastener intelligent dynamic anti-loose control method and system. BACKGROUND

[0002] In the prior art, in the high-speed rail system, threaded fasteners as key connecting elements are widely used in important parts such as track fasteners, vehicle suspension systems, bogies and vehicle body connections. During high-speed rail operation, these fasteners are subjected to complex vibration, impact and alternating load for a long time, which can easily cause loosening and lead to connection failure, and even cause serious safety accidents.

[0003] Research shows that the loosening process of threaded connection mainly includes two stages: the first stage is that the relative rotation between the bolt and the nut has not yet occurred, but the pretightening force has begun to decrease; the second stage is that the relative rotation between the bolt and the nut occurs, and the pretightening force rapidly decreases until complete loosening. The main reasons for bolt loosening include the following points: (1) alternating load, vibration and impact lead to reduced friction; (2) stress redistribution at the threaded contact interface causes pretightening force decay; (3) plastic deformation of the threaded connection pair and the connected parts causes the pretightening force to decrease.

[0004] Traditional threaded anti-loose technology mainly includes static anti-loose methods such as double nuts, self-locking nuts, spring washers, anti-loose washers and adhesive coating. Although these methods can improve the anti-loose performance of threaded connections to some extent, they have the following significant limitations. For example, the anti-loose effect is severely dependent on the initial pretightening force and fixed mechanical structure, which is difficult to adapt to the changing vibration environment and working conditions during high-speed rail operation; the stability of the anti-loose performance is poor, especially under low pretightening force or high vibration conditions. Research shows that in a vibration environment with an amplitude of more than ±2mm, the effectiveness of traditional anti-loose methods will decrease by more than 50%; the loosening state of threaded connections cannot be monitored and adjusted in real time, which makes maintenance personnel unable to discover and handle potential loosening risks in time; lack of intelligence and self-adaptive ability, unable to automatically adjust the anti-loose strategy according to different working conditions.

[0005] With the continuous improvement of high-speed rail running speed and the increasing complexity of running environment, traditional anti-loose methods have been difficult to meet the strict requirements of high-speed rail systems for fastener reliability and safety. SUMMARY

[0006] The main purpose of the present application is to provide a high-speed rail fastener intelligent dynamic anti-loose control method and system, which aims to solve the problems that the current high-speed rail fastener anti-loose setting cannot adapt to dynamic changing working conditions, the anti-loose performance stability is poor, and the real-time monitoring and adjustment capability is lacking.

[0007] In order to achieve the above object, the present application provides a high-speed rail fastener intelligent dynamic anti-loose control method, comprising:

[0008] S1, receiving the working parameters of the high-speed rail fastener, wherein the working parameters include vibration data and pre-tightening force data;

[0009] S2, analyzing the working parameters, if the working parameters exceed the preset range, judging the type of loosening according to the working parameters and generating a pre-tightening force adjustment scheme;

[0010] S3, executing the pre-tightening force adjustment scheme;

[0011] S4, continuously receiving the working parameters of the high-speed rail fastener within a preset time period, and repeating the steps of S2 to S3.

[0012] Further, the step of S4 comprises:

[0013] S5, according to all the pre-tightening force adjustment schemes and the working parameters of the high-speed rail fastener after executing the pre-tightening force adjustment scheme, optimizing the generation mode of the pre-tightening force adjustment scheme.

[0014] Further, the step of S5 comprises:

[0015] S5.1: defining an optimization target, wherein the optimization target includes maximizing anti-loose performance and minimizing energy consumption;

[0016] S5.2: defining decision variables, wherein the decision variables include pre-tightening force adjustment amount, pre-tightening force adjustment frequency and pre-tightening force adjustment timing;

[0017] S5.3: according to all the pre-tightening force adjustment schemes and the working parameters of the high-speed rail fastener after executing the pre-tightening force adjustment scheme, executing non-dominated sorting genetic algorithm II to solve the optimal anti-loose strategy.

[0018] Further, the step of S2 comprises:

[0019] extracting the features of the working parameters and analyzing them, if the working parameters exceed the preset range, judging the type of loosening through a machine learning module, and generating a pre-tightening force adjustment scheme, wherein the machine learning module is trained according to historical vibration data and loosening state.

[0020] Further, in the step of S2, the pre-tightening force adjustment scheme includes pre-tightening force adjustment amount, pre-tightening force adjustment rate and pre-tightening force adjustment timing.

[0021] Furthermore, in step S2, the preload adjustment scheme is generated using a multi-objective optimization algorithm, wherein the optimization objectives include maximizing the anti-loosening performance of the high-speed rail fasteners, minimizing the energy consumption in the preload adjustment scheme, and maintaining the long-term stability of the high-speed rail fasteners.

[0022] Furthermore, in step S1, the working parameters include preload data and vibration data in the three-axis directions;

[0023] Step S2 includes:

[0024] When the vibration data shows that the vibration amplitude of the high-speed rail fastener exceeds the set amplitude threshold, the preload data shows that the preload of the high-speed rail fastener decreases by more than the set preload percentage, or the energy in a specific frequency range in the power spectral density increases by more than the set value, the type of loosening is determined based on the working parameters and a preload adjustment scheme is generated.

[0025] Furthermore, the increase in energy within a specific frequency range in the power spectral density is achieved through vibration analysis based on the power spectral density. Frequency domain analysis is performed on multi-axis vibration data, and frequency domain characteristic parameters are calculated. The mathematical expression of the power spectral density analysis is: P(f)=lim(T→∞)[|X(f)|2 / T] where X(f) is the Fourier transform of the signal x(t), and T is the sampling time.

[0026] The present invention also provides a system for implementing the above-described intelligent dynamic anti-loosening control method for high-speed rail fasteners, comprising:

[0027] The parameter monitoring module is used to obtain and analyze the vibration data and the preload data. The parameter monitoring module includes a vibration monitoring module and a preload monitoring module. The vibration monitoring module is used to monitor the vibration acceleration and vibration frequency of the high-speed rail fastener in multiple directions, and the preload monitoring module is used to monitor the preload at the connection of the high-speed rail fastener.

[0028] The intelligent control module is used to determine the type of loosening based on the vibration data and the preload data and to generate a preload adjustment scheme.

[0029] The preload adjustment module is used to execute the preload adjustment scheme and adjust the preload of the high-speed rail fastener;

[0030] Furthermore, after executing the preload adjustment scheme, the intelligent control module continues to receive the working parameters of the high-speed rail fasteners within a preset time period, continues to generate preload adjustment schemes, and optimizes the generation method of the preload adjustment scheme based on all the preload adjustment schemes.

[0031] The intelligent dynamic anti-loosening control method and system for high-speed rail fasteners provided by this invention obtains working parameters in multiple dimensions and determines the type of loosening of the fasteners based on these parameters, thereby generating a preload adjustment scheme to correct the fastening status and significantly improve the accuracy and timeliness of risk identification. Furthermore, it continues to receive working parameters of the high-speed rail fasteners within a preset time period and continuously optimizes and corrects them. Experimental verification shows that it can maintain the preload of high-speed rail fasteners at more than 95% of the initial value, reducing the loosening rate by 80%. In cyclic vibration testing, the fatigue life of high-speed rail fasteners is extended by 40%. Compared with traditional methods, the anti-loosening performance is improved by more than 60% under low preload and high vibration conditions. Besides the high-speed rail field, this control method can also be applied to bridges, aerospace, and other fields with high requirements for fastener anti-loosening performance. Attached Figure Description

[0032] Figure 1 This is a schematic diagram of an embodiment of the intelligent dynamic anti-loosening control method for high-speed rail fasteners according to the present invention;

[0033] Figure 2 This is a comparative analysis diagram of the preload retention of the intelligent dynamic anti-loosening system for high-speed rail fasteners according to an embodiment of the present invention during a lateral vibration test;

[0034] Figure 3 This is an analysis diagram of the loosening prediction of a high-speed rail fastener intelligent dynamic anti-loosening system according to an embodiment of the present invention.

[0035] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0036] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0037] Those skilled in the art will understand that, unless specifically stated otherwise, the singular forms “a,” “an,” “the,” “the,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the term “comprising” as used in this specification means the presence of the stated features, integers, steps, operations, elements, units, modules, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, units, modules, components, and / or groups thereof. It should be understood that when we say an element is “connected” or “coupled” to another element, it can be directly connected or coupled to the other element, or there may be intermediate elements. Furthermore, “connected” or “coupled” as used herein can include wireless connection or wireless coupling. The term “and / or” as used herein includes all or any of the units and all combinations of one or more associated listed items.

[0038] It will be understood by those skilled in the art that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. It should also be understood that terms such as those defined in general dictionaries should be understood to have the same meaning as in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless specifically defined as herein.

[0039] Reference Figure 1 In one embodiment of the present invention, a method for intelligent dynamic anti-loosening control of high-speed rail fasteners includes:

[0040] S1. Receive the working parameters of the high-speed rail fasteners, including vibration data and preload data;

[0041] S2. Analyze the working parameters. If the working parameters exceed the preset range, determine the type of loosening based on the working parameters and generate a pre-tightening force adjustment plan.

[0042] S3. Execute the preload adjustment scheme;

[0043] S4. Continue to receive the working parameters of the high-speed rail fasteners within the preset time period, and repeat steps S2 to S3.

[0044] In the existing technology, with the continuous increase in the operating speed of high-speed rail and the increasing complexity of the operating environment, traditional anti-loosening methods are no longer able to meet the strict requirements of high-speed rail systems for the reliability and safety of fasteners.

[0045] In this invention, the intelligent dynamic anti-loosening control method for high-speed rail fasteners includes:

[0046] In step S1, the operating parameters of the high-speed rail fastener are received, including vibration data and preload data. These operating parameters can be obtained by a parameter monitoring module, which acquires and analyzes the vibration data and preload data. The parameter monitoring module includes a vibration monitoring module and a preload monitoring module. The vibration monitoring module monitors the vibration acceleration and frequency of the high-speed rail fastener in multiple directions, while the preload monitoring module monitors the preload at the fastener connection. The vibration monitoring module includes an integrated acceleration sensor to monitor the vibration acceleration and frequency of the high-speed rail fastener in multiple directions. The preload monitoring module includes a strain gauge to monitor the strain state at the fastener connection and further calculate the preload.

[0047] In step S2, the working parameters are analyzed. If the working parameters exceed the preset range, the type of loosening is determined based on the working parameters, and a preload adjustment plan is generated.

[0048] For example, vibration data includes energy within a specific frequency range in vibration amplitude and power spectral density, while preload data includes preload. Therefore, if at least one of the following three criteria deviates from the normal range, it is determined that there is a problem with the operation of the high-speed rail fasteners.

[0049] The three judgment conditions are as follows:

[0050] A. The vibration amplitude of the high-speed rail fasteners in the vibration data exceeds the set amplitude threshold;

[0051] B. The preload of high-speed rail fasteners in the preload data decreased by more than the set preload percentage;

[0052] C. When the energy in the power spectral density of vibration data increases beyond a set value within a specific frequency range.

[0053] If the working parameters exceed the preset range, the type of loosening is determined based on the type and magnitude of the deviation. Specifically, there are at least three types of loosening: mechanical loosening, preload reduction caused by abnormal friction interface stress, and preload reduction caused by plastic deformation.

[0054] When both A and B show large deviations, the machinery becomes loose; when C shows a small deviation and B shows a large deviation, it is determined that plastic deformation has caused a decrease in preload; when C shows a large deviation and B shows a small deviation, but A does not, it is determined that the decrease in preload is caused by abnormal stress at the friction interface.

[0055] If plastic deformation is detected as causing preload decay, in addition to adjusting the preload, an alarm should be issued; if mechanical loosening is detected, the preload should be adjusted at a faster speed; if abnormal stress at the friction interface is detected as causing preload decay, the preload should be adjusted in a more rigorous manner.

[0056] The steps in S2 can be implemented through an intelligent control module, which includes an embedded processing system for executing data processing and control algorithms; a vibration analysis module for identifying vibration characteristics through frequency and time domain analysis; a machine learning module for adaptively learning loosening patterns under different working conditions; and a preload control module for generating actuator control signals. The machine learning module, trained based on historical vibration data and loosening states, can predict loosening trends under different vibration conditions, with a prediction accuracy of no less than 85%. Furthermore, the machine learning module can identify three preload decay modes caused by stress redistribution, plastic deformation, and rotational loosening.

[0057] A brief introduction to several basic mechanisms of loosening of threaded fasteners:

[0058] Rotational loosening (self-relaxation): This is mainly caused by relative rotation between threaded pairs due to external cyclic loads (such as vibration and impact). This is one of the main causes of fastener loosening in dynamic environments such as high-speed railways.

[0059] Non-rotational loosening (relaxation): refers to the loss of preload that occurs when there is no relative rotation between the threaded parts. Specific mechanisms include flattening (embedding), creep, and stress relaxation.

[0060] The specific types of loosening targeted by this invention

[0061] The following will describe in detail the main types of loosening identified by this invention and their main diagnostic features (based on working parameters) in conjunction with the actual working conditions of high-speed railway fasteners.

[0062] 1. Mechanical loosening (mechanical loosening)

[0063] Description: This type most likely corresponds to rotational loosening. Under vibration and lateral loads, the nut or bolt physically rotates, directly and rapidly causing a decrease in preload. Junker's earlier research has confirmed that lateral vibration is a major factor leading to bolt loosening.

[0064] Diagnostic feature: "When large deviations occur in both A and B simultaneously."

[0065] Condition A: Vibration data shows that the vibration amplitude of the high-speed rail fastener exceeds the set amplitude threshold (e.g., the vibration acceleration amplitude exceeds 0.5g).

[0066] Condition B: Preload data shows that the preload of the high-speed rail fasteners has decreased by more than the set preload percentage (e.g., the preload has decreased by more than 10% of the initial value).

[0067] Characteristic principle: Significant vibration (condition A) provides energy for the rotation of the thread, while a significant decrease in preload (condition B) is a direct consequence of the rotation.

[0068] 2. Plastic deformation causes preload decay (plastic deformation causes preload decay)

[0069] Description: This refers to permanent deformation of bolts, nuts, or connected parts. This may be due to excessive initial preload, material yielding under excessive working load, or settlement / embedding caused by the flattening of surface roughness peaks.

[0070] Diagnostic feature: "When C shows a small deviation and B shows a large deviation".

[0071] Condition B: The preload decreases by more than the set preload percentage (e.g., >10%).

[0072] Condition C: The energy increase in a specific frequency range of the power spectral density (PSD) in the vibration data exceeds a set value (e.g., >30%), but this deviation is described as a "small deviation". The interpretation of "small deviation" should be cautious—it may mean that although a PSD change is detected, its pattern or magnitude differs from the PSD characteristics of other types of loosening (such as stress redistribution), or it may specifically refer to a PSD pattern that points to deformation rather than free rotation.

[0073] Characteristic principle: A significant drop in preload (condition B) is a typical characteristic of material yielding or significant flattening. The "small deviation" of PSD in condition C may reflect a change in connection stiffness due to material deformation, but this change may be different from the free vibration signal generated during severe mechanical loosening.

[0074] 3. Preload decay / stress redistribution caused by abnormal stress at the friction interface (Preload decay / stress redistribution caused by abnormal stress at the friction interface)

[0075] Description: This type is closely related to stress relaxation. Preload gradually decreases over time (usually at the microscopic level) as internal material stresses redistribute, even though the overall dimensions of the connection may not change significantly. It can also involve creep in softer materials such as gaskets or coatings. Preload loss due to stress relaxation is difficult to detect using conventional methods because the change in clamping length is not obvious.

[0076] Diagnostic features: "When C shows a large deviation and B shows a small deviation, while A does not show any deviation."

[0077] Condition A: The vibration amplitude does not exceed the threshold.

[0078] Condition B: Preload decreases, but is a “small deviation” (e.g., below the threshold for mechanical loosening, but still detectable).

[0079] Condition C: The energy increase in a specific frequency range of the PSD exhibits a “large deviation” (e.g., >30%).

[0080] Characteristic principle: The absence of severe vibration (condition A) rules out immediate rotational loosening. A slight decrease in preload (condition B) is consistent with the characteristics of stress relaxation. A significant change in PSD (condition C) is a key diagnostic criterion: PSD analysis can determine whether the resonant frequency has changed due to loosening or deformation. For stress redistribution, changes in interface material properties or subtle changes in contact stiffness can cause significant changes in vibration characteristics (PSD), even when the overall amplitude (condition A) is not high and the decrease in preload (condition B) is not large.

[0081] A detailed interpretation of the power spectral density (PSD) characteristics (condition C) is crucial. Changes in PSD energy within condition C are used to distinguish between plastic deformation (small deviation in C) and stress redistribution (large deviation in C). This suggests that the nature or magnitude of the PSD changes themselves carry important diagnostic information beyond simple threshold judgments. The selection of a “specific frequency range” (1) is central. For example:

[0082] Stress redistribution may cause a slight shift in the resonant frequency or a change in damping characteristics, thereby producing significant energy changes in a specific frequency band, even if the overall vibration amplitude is not large.

[0083] Plastic deformation can lead to localized changes in joint stiffness, and the overall magnitude of the affected frequency bands or PSD energy changes may differ from the characteristics of stress redistribution. Different physical changes (plastic deformation vs. interfacial stress relaxation) will affect the joint's dynamic characteristics (mass, stiffness, damping) in different ways. These different effects will manifest as different features on the PSD spectrum (e.g., shifts in peak values ​​at different frequencies, changes in peak amplitude, peak broadening, etc.). Therefore, the "small deviation" and "large deviation" in condition C likely refer to quantitative differences in these PSD features, and a machine learning module will be trained to recognize these complex patterns.

[0084] While rule-based logic (combinations of conditions A, B, and C) provides a basic judgment framework, this invention introduces a machine learning module. This module, trained on historical vibration data and relaxation states, can identify more complex patterns and potentially distinguish with higher precision the three preload decay modes caused by stress redistribution, plastic deformation, and rotational loosening. This allows the system to go beyond fixed threshold judgments, achieving adaptive and more accurate relaxation diagnosis. The use of machine learning (e.g., decision trees, convolutional neural networks) combined with PSD data for bolt loosening state classification has been supported by existing research.

[0085] Table 1: Diagnostic Matrix for Loosening Types

[0086] To clearly and concisely summarize the identification logic for different types of loosening, the table below summarizes the different combinations of working parameters.

[0087]

[0088]

[0089] The diagnostic matrix is ​​constructed according to the following logic: First, the user's core need is to differentiate between different types of loosening and apply specific solutions. Second, it provides logic based on three conditions, A, B, and C, to distinguish between the three main types.

[0090] Customized preload adjustment strategies for specific slack phenomena

[0091] A. Basic principles of preload adjustment (force, speed, timing)

[0092] The core components of the preload adjustment scheme of this invention include "preload adjustment amount, preload adjustment rate, and preload adjustment sequence." These three key parameters collaboratively construct a complete system of basic principles for preload adjustment. The preload adjustment amount refers to the target preload value that the system needs to restore or the incremental change range of the preload. This parameter can be expressed as a percentage of the initial preload, a fixed absolute value, or a variable value calculated in real time through a dynamic algorithm. The preload adjustment rate defines the time gradient of the applied adjustment force (e.g., kN / second). This parameter directly affects the dynamic response characteristics of the material, and is particularly important when dealing with dynamic operating conditions or sensitive material systems. The preload adjustment sequence covers the initiation timing of the adjustment operation (including immediate or delayed execution options), the specific application method (single-step, multi-step, or continuous adjustment mode), and the time interval control between steps in multi-step operations. The precise coordination of these three core elements determines the technical effectiveness and engineering adaptability of the entire preload adjustment strategy.

[0093] B. Specific adjustment strategies for each type of loosening

[0094] Based on the characteristics of different loosening types identified by the relaxation phenomenon identification system, this invention has developed corresponding differentiated preload adjustment technology solutions.

[0095] 1. Strategies for dealing with mechanical loosening

[0096] For mechanical loosening, the system identifies it based on the characteristic combination of a large deviation in vibration amplitude parameter (A) and a simultaneous large deviation in preload parameter (B). The adjustment goal for this type of loosening is to quickly and effectively restore the preload to the design level, thereby preventing further loosening and potential connection integrity failure under continuous dynamic loads. Based on the technical principle of "requiring rapid preload adjustment," the system formulates the following adjustment strategy: Regarding the amount of preload adjustment, the system restores the preload to the initial design target value, or makes a moderate enhancement adjustment based on the optimization results of the system learning module (step S5) (e.g., increasing the original target value by 5-10% within the safety margin). The specific value is dynamically determined through a multi-objective optimization algorithm. Regarding the preload adjustment rate, the system adopts a high-speed adjustment mode. Its technical rationale lies in minimizing the operating time of the fastener in a significantly loose state, thereby effectively reducing the risk of accelerated wear or secondary damage caused by excessive component movement. The specific adjustment rate (e.g., X kN / s) should reach the maximum value within the actuator's capability range, while strictly avoiding impact damage. Regarding the timing of preload adjustment, the system adopts an immediate response, single-step execution mode, meaning that the correction procedure should be initiated immediately upon detection and confirmation. A deeper understanding of the "faster speed" technical parameter and its engineering rationality analysis indicate that mechanical rotational loosening signifies that the fastener is in a dynamic process of active loosening. Rapid retightening can minimize the time window for further rotational loosening of the fastener and effectively limit the duration of excessive vibration and fretting wear or impact between components. This strategy achieves both effective recovery of preload and preventative damage limitation; therefore, the "faster speed" parameter possesses clear functional rationality and technical necessity.

[0097] 2. Strategies for Addressing Preload Reduction Caused by Plastic Deformation

[0098] For preload decay caused by plastic deformation, the system identifies it based on a significant deviation in the preload parameter (B) and a small deviation or specific pattern characteristic in the power spectral density (PSD) parameter (C). The primary goal of adjusting for this type of loosening is to promptly issue a system alarm, clearly indicating that the fastener may have suffered structural damage. Any subsequent preload adjustment should only be considered a temporary safety measure before a comprehensive inspection and possible component replacement. Based on the technical principle that "in addition to preload adjustment, an alarm should also be issued," the system adopts the following adjustment strategy: Regarding the amount of preload adjustment, the system adopts a cautious and conservative adjustment mode, considering restoring the preload to a lower percentage of the original design target value (e.g., 70-80%), or adjusting it only to the minimum safe level sufficient to temporarily stabilize the joint. Simultaneously, the system must confirm that the relevant materials may have weakened. The main focus of the operation should be on the alarm function. If the material has already yielded, the adjustment operation should not attempt to reach the full preload level of the original design, as this may lead to further structural damage. Regarding the preload adjustment rate, the system employs a medium to slow adjustment speed, as applying excessively rapid external force to components that have undergone plastic deformation may exacerbate crack propagation or structural damage. Regarding the timing of preload adjustment, the system should immediately activate the alarm function. If preload adjustment is required, it should be performed after the alarm procedure is completed, using a single, cautious force application method. Simultaneously, the system should mark this fastener as a priority inspection item. The priority of the alarm function is critically important for plastic deformation phenomena, technically acknowledging that plastic deformation is usually an irreversible and permanent change, and implying a potential weakening of structural load-bearing capacity. The patent embodiments should explicitly emphasize that the fundamental purpose of the adjustment operation is to maintain a temporary safe operating state, rather than to "completely repair" permanently damaged components. This technical concept helps to rationally manage system performance expectations and highlights the system's inherent safety protection characteristics. Since plastic deformation means that the material has yielded beyond its elastic limit and its mechanical properties have changed, simply retightening to the original preload level may exceed the new, lower yield strength threshold of the damaged component or lead to further crack propagation. Therefore, the alarm function constitutes the most critical system output. Any retightening operation should only be regarded as a temporary expedient, not a fundamental solution to fully restore the original structural integrity. The corresponding adjustment parameters (force and speed) must fully reflect this technical caution.

[0099] 3. Strategies for addressing preload decay / stress redistribution caused by abnormal stress at the friction interface

[0100] To address the preload decay / stress redistribution phenomenon caused by abnormal stress at the friction interface, the system identifies it based on the following criteria: no significant deviation in vibration amplitude parameter (A), a small deviation in preload parameter (B), and a large deviation or specific pattern characteristics in power spectral density (PSD) parameter (C). The adjustment goal for this type of relaxation is to effectively compensate for the gradual preload loss caused by material stress relaxation or embedding effects, thereby maintaining the long-term structural integrity and functional stability of the fastened joint. Based on the technical principle that "preload adjustment needs to be performed in a relatively rigorous manner," the system formulates the following adjustment strategy: In terms of the preload adjustment amount, the system precisely restores the preload to the initial design target value. In terms of the preload adjustment rate, the system adopts a slow to medium-speed adjustment mode. The technical basis for this is that stress relaxation and material embedding phenomena usually exhibit time-dependent material response processes. Applying preload too quickly may not provide sufficient adjustment and adaptation time for the contact interface, or may induce new stress concentration phenomena within the material. Regarding the timing of preload adjustment, this is a key aspect embodying the technical essence of "rigorous adjustment method." The system can adopt the following strategy options: a delayed or phased application strategy, where the system first applies a partial adjustment for initial correction, evaluates the system response through real-time monitoring, and then applies the remaining adjustment, rather than using a simplified mode of immediate full retightening; a delayed re-tightening strategy, drawing on the engineering practice of "relaxation pass," where after performing an initial small adjustment or only detecting a loose phenomenon, the system waits for a preset time interval (e.g., minutes to hours, the specific duration determined by the learning algorithm of optimization module S5) to allow further embedding / loosening to fully occur before performing the main adjustment operation. For example, for certain gasket material types, waiting 1 hour before re-tightening has proven to be effective; and an iterative adjustment strategy, where the system applies a partial increment of the required preload, monitors the system response characteristics (including preload stability and PSD characteristic changes) in real time, and then applies subsequent incremental adjustments, repeating this process until the target value is stably reached. This strategy is similar to the engineering practice mode of "tightening - short wait - reapplying torque (which can be repeated in multiple steps)." For stress redistribution / relaxation phenomena, the technical definition of the concept of "rigorous / careful adjustment" is of critical importance. Since such relaxation phenomena (stress relaxation, embedding / flattening) are usually a gradual time-dependent process, a simple one-time retightening strategy is often difficult to achieve the best results. "Rigorous" adjustment can be understood as a technical method of multi-stage or timed control.Stress relaxation and material embedding are typical time-dependent material response phenomena. Applying torque force rapidly and completely may not allow the material contact interface to reach an optimal stable equilibrium state. Engineering strategies such as "tightening-waiting-re-tightening" or "relaxation-re-tightening passes" involve specific timing control to effectively counteract these physical mechanisms. Therefore, the "rigorous" adjustment technique of this invention for such relaxation phenomena may include setting a programmed delay time before the final tightening operation, or adopting a step-by-step application method with real-time monitoring, to better adapt to and match the intrinsic behavior characteristics of the material. Precise control of timing parameters becomes crucial, and optimized feedback loop systems are very suitable for mastering the optimal delay time or step-by-step strategy parameters of such "rigorous" adjustment methods through machine learning.

[0101] C. The core role of multi-objective optimization in improving and adjusting the plan

[0102] This invention employs a multi-objective optimization algorithm (such as the NSGA-II algorithm) to dynamically generate preload adjustment schemes. This optimization system comprehensively considers multiple key technical objectives: maximizing anti-loosening performance, minimizing energy consumption during adjustment, and maintaining long-term stability of the fastener (achieved by reducing adjustment frequency). This technical feature means that the specific adjustment force, speed, and timing parameters are not preset fixed values, but rather optimal solutions obtained through dynamic optimization calculations. For example, if a slightly slower adjustment speed can achieve a comparable anti-loosening effect with lower energy consumption and simultaneously promote longer-term system stability, the optimization algorithm will tend to select this technical solution. This design concept further reinforces the technical characteristic that the "optimal solution" of this invention is not a static preset, but rather continuously evolves and improves through a continuous learning and optimization process. This constitutes an innovative feature with strong patent value. Multi-objective optimization technology plays a core driving role in perfecting the adjustment scheme, ensuring that the preload adjustment strategy not only achieves the expected goals in anti-loosening performance but also achieves an optimal balance in energy efficiency and long-term operational stability, thus providing a scientifically rigorous and engineering-practical technical foundation for the entire intelligent preload adjustment system.

[0103] D. Table 2: Matrix of Specific Adjustment Strategies for Loosening Types

[0104] The table below clearly links each type of loosening diagnosed to its unique, optimized adjustment strategy (including force, velocity, and timing parameters and their underlying principles).

[0105]

[0106]

[0107]

[0108] In step S3, the preload adjustment scheme is executed. The preload adjustment scheme is executed through a preload adjustment module, which can be at least one of a hydraulic actuator and an electric actuator. The hydraulic actuator and the electric actuator support a preload adjustment range of 10kN-100kN with an adjustment accuracy of ±1kN.

[0109] In step S4, the working parameters of the high-speed rail fasteners are received again within a preset time period, and steps S2 to S3 are repeated. In specific application scenarios, a single execution of the pre-tightening force adjustment scheme cannot effectively resolve the loosening situation, resulting in multiple executions of the pre-tightening force adjustment scheme; or, after a single execution of the pre-tightening force adjustment scheme, subsequent monitoring of working parameters shows that the loosening situation has been over-optimized. Based on the analysis of working parameters, if the high-speed rail fasteners show initial loosening, a pre-tightening force adjustment scheme is generated for correction and improvement. After executing the pre-tightening force adjustment scheme, the status of the high-speed rail fasteners is monitored again within a preset time period. If it is determined that the working parameters monitored within this preset time period still fall within the scope of the current loosening situation, a new pre-tightening force adjustment scheme is generated based on the working parameters.

[0110] In summary, this method obtains multiple dimensions of operating parameters and determines the type of loosening in high-speed rail fasteners based on these parameters, thereby generating a preload adjustment scheme to correct the fastening status and significantly improve the accuracy and timeliness of risk identification. Furthermore, it continues to receive operating parameters of the high-speed rail fasteners within a preset time period and continuously optimizes and corrects them. Experimental verification shows that it can maintain the preload of high-speed rail fasteners at more than 95% of the initial value, reducing the loosening rate by 80%. In cyclic vibration testing, the fatigue life of high-speed rail fasteners is extended by 40%. Compared with traditional methods, under low preload and high vibration conditions, the anti-loosening performance is improved by more than 60%. Besides the high-speed rail field, this control method can also be applied to bridges, aerospace, and other fields with high requirements for fastener anti-loosening performance.

[0111] In one embodiment, step S4 is followed by:

[0112] S5. Based on all the preload adjustment schemes and the working parameters of the high-speed rail fasteners after implementing the preload adjustment schemes, optimize the generation method of the preload adjustment scheme.

[0113] In this embodiment, based on the analysis of working parameters, an initial loosening of the high-speed rail fasteners is detected, and a pre-tightening force adjustment scheme is generated for correction and improvement. After executing the pre-tightening force adjustment scheme, the status of the high-speed rail fasteners continues to be monitored within a preset time period. If the working parameters monitored within this preset time period are still within the scope of the current loosening situation, a new pre-tightening force adjustment scheme is generated based on the working parameters. The pre-tightening force adjustment scheme is generated based on the working parameters using a specific algorithm. This algorithm can largely solve the loosening situation, but in specific application scenarios, a single execution of the pre-tightening force adjustment scheme cannot achieve a high-quality solution to the loosening situation. Therefore, multiple executions of the pre-tightening force adjustment scheme have occurred; or, after a single execution of the pre-tightening force adjustment scheme, subsequent monitoring of working parameters shows that the loosening situation has been over-optimized. The first pre-tightening force adjustment scheme and all pre-tightening force adjustment schemes within the subsequent preset time periods form a scheme group. When the original algorithm fails to effectively resolve loosening issues in a single execution, the generation method (algorithm) of the original preload adjustment scheme is optimized based on the preload adjustment schemes obtained from multiple executions or an overly optimized preload adjustment scheme in a single execution. For example, if multiple preload adjustment schemes were executed before the loosening issue was resolved, the preload adjustment scheme could be a synthesis of the adjustment values ​​from multiple schemes, either numerically or temporally, so that the adjustment is completed in the first preload adjustment scheme. Alternatively, if a single preload adjustment scheme is over-optimized, the preload adjustment scheme could reduce the adjustment value of the preload adjustment scheme.

[0114] In one embodiment, step S5 includes:

[0115] S5.1: Define optimization objectives, wherein the optimization objectives include maximizing anti-loosening performance and minimizing energy consumption;

[0116] S5.2: Define decision variables, wherein the decision variables include preload adjustment amount, preload adjustment frequency, and preload adjustment timing;

[0117] S5.3: Based on all the preload adjustment schemes and the working parameters of the high-speed rail fasteners after the preload adjustment schemes are implemented, execute the non-dominated sorting genetic algorithm II to solve for the optimal anti-loosening strategy.

[0118] In this embodiment, a method for optimizing the preload adjustment scheme is presented. The variables are the preload adjustment amount, the preload adjustment frequency, and the preload adjustment timing. The optimization objectives are to maximize anti-loosening performance and minimize energy consumption. Based on the preload adjustment scheme and the working parameters of the high-speed rail fasteners after implementing the preload adjustment scheme, a non-dominated sorting genetic algorithm II is used to solve for the optimal anti-loosening strategy.

[0119] In one embodiment, step S2 includes:

[0120] The features of the working parameters are extracted and analyzed. If the working parameters exceed the preset range, the machine learning module is used to determine the type of loosening and generate a pre-tightening force adjustment scheme. The machine learning module is trained based on historical vibration data and loosening status.

[0121] In this embodiment, as mentioned in the previous embodiments, the type of loosening can be determined to a certain extent based on the three judgment conditions (A, B, and C). Therefore, a model is introduced, and a dataset is used to train the machine learning module. Ultimately, the machine learning module can accurately determine the type of loosening. Through the machine learning module, the loosening patterns under different working conditions can be adaptively learned, continuously optimizing the anti-loosening performance.

[0122] In one embodiment, in step S2, the preload adjustment scheme includes preload adjustment amount, preload adjustment rate, and preload adjustment timing.

[0123] In this embodiment, the preload adjustment scheme includes a preload application procedure. The decision variables of the preload adjustment scheme specifically include the numerical value, the addition speed, and the addition regime, which are combined to form an optimal adjustment method.

[0124] In one embodiment, in step S2, the preload adjustment scheme is generated using a multi-objective optimization algorithm, wherein the optimization objectives include maximizing the anti-loosening performance of the high-speed rail fasteners, minimizing the energy consumption in the preload adjustment scheme, and maintaining the long-term stability of the high-speed rail fasteners.

[0125] In this embodiment, the multi-objective optimization algorithm is a mathematical method for solving optimization problems with multiple objective functions. It helps decision-makers find a reasonable balance between multiple conflicting objectives. By simulating evolution or group behavior in nature, a wider range of regions can be explored in the search space, thus avoiding getting trapped in local optima. By optimizing multiple objectives, the model can not only improve its performance on a single metric but also maintain good performance on other metrics, thereby enhancing overall robustness. Anti-loosening performance refers to the preload retention ratio, energy consumption refers to the energy consumed in the preload adjustment scheme, and long-term stability represents the ability to avoid multiple adjustments over a long period. First, at the operational level, the multi-objective optimization algorithm is used to dynamically generate the optimal preload adjustment scheme in real-time or near real-time. This process aims to balance multiple conflicting objectives, mainly including: maximizing the anti-loosening performance of the fastener (e.g., maintaining the preload near its target value), minimizing the energy consumed by the preload adjustment mechanism, and maximizing the long-term stability and reliability of the fastening connection (e.g., minimizing the adjustment frequency or component stress). Secondly, at the strategic or meta-level, a multi-objective optimization algorithm (specifically, Non-Dominated Sorting Genetic Algorithm II, NSGA-II) is employed to optimize the generation method or methodology of the preload adjustment scheme. This higher-level optimization process considers objectives such as maximizing overall (cumulative) anti-loosening performance and minimizing cumulative energy consumption, with decisions based on historical data and system performance feedback. This two-level, multi-objective adaptive approach enables the system to find a set of Pareto optimal solutions, providing a flexible and powerful means of managing fastener preload, thereby significantly improving the safety of critical fastening components, reducing maintenance, and extending their service life. The system is closed-loop not only for control actions but also for improvements to the optimization strategy itself. In step S5, optimizing the generation method of the preload adjustment scheme means that the multi-objective optimization algorithm in S5 uses historical data (inputs, decisions made by S2-MOO, and results) to adjust the operation of the multi-objective optimization algorithm in S2. This is a higher level of adaptation, making the system more robust to unforeseen situations or dynamic changes in the system's lifecycle.

[0126] Balanced performance: Multi-objective optimization algorithms, particularly those using NSGA-II, enable the system to find solutions that represent the optimal trade-offs between conflicting objectives (anti-loosening, energy consumption, stability), rather than excelling in one objective at the expense of others. This results in more comprehensive optimization of fastener performance.

[0127] Adaptability and robustness: In step S2, real-time adaptation to changing conditions is provided. In step S5, the system is enabled to learn from past performance and adjust its strategies over the long term, making the system robust to constantly changing operating patterns or wear characteristics. This is a key advantage over static or purely rule-based systems.

[0128] Scientific and objective decision-making: Using mathematically formed objective functions and well-defined optimization algorithms (such as NSGA-II) ensures that preload adjustments are determined based on objective standards and scientific principles, rather than just heuristic rules.

[0129] Enhanced reliability and safety: By maintaining optimal preload and accommodating loosening conditions, this system helps improve the overall reliability and safety of structures containing fasteners, such as high-speed railways. The "long-term stability" objective directly supports this.

[0130] Maintenance efficiency: Optimizing "long-term stability" (reducing the number of adjustments) and potential "energy minimization" can reduce maintenance interventions and lower operating costs. The abstract mentions "80% reduction in loosening rate" and "40% extension of fatigue life." These quantifiable improvements are strong indicators of the technical effectiveness and innovative steps achieved by the entire system (with MOO as its core intelligent component). Linking MOO's capabilities (balancing objectives, adaptability) to these quantifiable results strengthens the argument for MOO's partial contribution and non-obviousness.

[0131] Superior to single-objective optimization: Multi-objective optimization algorithms outperform sequential optimization objectives or those using simple weighted sums, especially in finding diverse, high-quality non-dominated solution sets.

[0132] Handling complexity: MOEAs such as NSGA-II can effectively handle complex optimization problems with multiple decision variables and constraints.

[0133] In one embodiment, the working parameters in step S1 include preload data and vibration data in the three-axis directions;

[0134] Step S2 includes:

[0135] When the vibration data shows that the vibration amplitude of the high-speed rail fastener exceeds the set amplitude threshold, the preload data shows that the preload of the high-speed rail fastener decreases by more than the set preload percentage, or the energy in a specific frequency range in the power spectral density increases by more than the set value, the type of loosening is determined based on the working parameters and a preload adjustment scheme is generated.

[0136] In this embodiment, in step S2, the meaning of the working parameters exceeding the preset range is that one of the following conditions is triggered: the vibration acceleration amplitude exceeds 0.5g; the preload decreases by more than 10% of the initial value; or the power spectral density analysis shows an energy increase of more than 30% within a specific frequency range. In particular, the sudden increase in energy within a specific frequency range is often due to the loosening or deformation of high-speed rail fasteners, thus determining whether the resonant frequency has changed. Combining the vibration acceleration amplitude data can eliminate the loosening characteristic.

[0137] In one embodiment, the increase in energy within a specific frequency range in the power spectral density is achieved through vibration analysis based on the power spectral density. Frequency domain analysis is performed on multi-axis vibration data, and frequency domain characteristic parameters are calculated. The mathematical expression of the power spectral density analysis is: P(f)=lim(T→∞)[|X(f)|2 / T] where X(f) is the Fourier transform of the signal x(t), and T is the sampling time.

[0138] This embodiment presents the usage and analysis method of power spectral density, enabling status monitoring of high-speed rail fasteners on the XYZ axes.

[0139] In one embodiment, steps S1 and S2 include:

[0140] Receive ambient temperature change data and correct the vibration data and preload data.

[0141] In this embodiment, the ambient temperature is monitored by a temperature compensation module. After obtaining the temperature data, the vibration data and preload data are corrected, thereby enabling more accurate monitoring of the status of the high-speed rail fasteners and allowing the system to maintain working accuracy within the range of -40℃ to +80℃.

[0142] The present invention also provides a system for implementing the above-described intelligent dynamic anti-loosening control method for high-speed rail fasteners, comprising:

[0143] The parameter monitoring module is used to obtain and analyze the vibration data and the preload data. The parameter monitoring module includes a vibration monitoring module and a preload monitoring module. The vibration monitoring module is used to monitor the vibration acceleration and vibration frequency of the high-speed rail fastener in multiple directions, and the preload monitoring module is used to monitor the preload at the connection of the high-speed rail fastener.

[0144] The intelligent control module is used to determine the type of loosening based on the vibration data and the preload data and to generate a preload adjustment scheme.

[0145] The preload adjustment module is used to execute the preload adjustment scheme and adjust the preload of the high-speed rail fastener;

[0146] In this embodiment, the key components of the preload adjustment module are made of GH4169 high-temperature alloy material and undergo heat treatment, enabling them to operate normally at a high temperature of 650℃. The vibration monitoring module includes an integrated acceleration sensor to monitor the vibration acceleration and vibration frequency of the high-speed rail fastener in multiple directions. The preload monitoring module includes a strain gauge section to monitor the strain state at the connection of the high-speed rail fastener. The system's operation is the same as described in the aforementioned method embodiment and will not be repeated here.

[0147] In one embodiment, after the preload adjustment scheme is executed, the intelligent control module continues to receive the working parameters of the high-speed rail fasteners within a preset time period, continues to generate preload adjustment schemes, and optimizes the generation method of the preload adjustment scheme based on all the preload adjustment schemes.

[0148] In this embodiment, the system operates in the same manner as in the aforementioned method embodiments, and will not be repeated here.

[0149] In one embodiment, the parameter monitoring module further includes a temperature compensation module;

[0150] The temperature compensation module is used to monitor changes in ambient temperature and correct the vibration data and the preload data.

[0151] In this embodiment, the ambient temperature is monitored by a temperature compensation module. After obtaining the temperature data, the vibration data and preload data are corrected, thereby enabling more accurate monitoring of the status of the high-speed rail fasteners and allowing the system to maintain working accuracy within the range of -40℃ to +80℃.

[0152] In one embodiment, the preload adjustment module includes at least one of a hydraulic actuator and an electric actuator;

[0153] The hydraulic and electric actuators support a preload adjustment range of 10kN-100kN with an adjustment accuracy of ±1kN.

[0154] Example 1: Intelligent Dynamic Anti-Loosening Control Method for High-Speed ​​Railway Fasteners

[0155] The intelligent dynamic anti-loosening control method for high-speed rail fasteners provided by this invention includes the following steps:

[0156] Step S1: Data Acquisition and Analysis. The vibration monitoring module acquires real-time vibration acceleration and strain data at the high-speed rail fasteners at a frequency of 100Hz, transmitting the data to the intelligent control module via the data acquisition unit. The intelligent control module filters and normalizes the raw data, then performs dual-dimensional analysis in the time and frequency domains. In the time domain analysis, the system calculates parameters such as the root mean square value, peak value, and peak factor of the vibration acceleration. In the frequency domain analysis, the system uses Fast Fourier Transform (FFT) to calculate the power spectral density (PSD) and extracts features such as the dominant frequency, amplitude, and frequency band energy distribution.

[0157] Step S2: The dynamic adjustment decision-making intelligent control module determines whether preload adjustment is needed based on the vibration analysis results. The adjustment decision is based on a comprehensive evaluation of the following conditions:

[0158] The vibration acceleration RMS value exceeds the set threshold (0.5g laterally, 0.8g axially);

[0159] Strain data shows that the preload has decreased by more than 10% of its initial value;

[0160] PSD analysis showed that the energy increased by more than 30% within the bolt's natural frequency range;

[0161] Other abnormal vibration characteristics.

[0162] When any of the above conditions are met, the system enters the preload adjustment process. Simultaneously, the machine learning module calculates the loosening risk level (level 1-5) based on the current vibration feature vector and historical data to assist the decision-making process.

[0163] Step S3: Perform Preload Adjustment. The preload control module generates a preload adjustment strategy based on the loosening risk level and vibration characteristics, including the adjustment amount, adjustment rate, and timing. The system uses different adjustment amounts according to the risk level, from minor risk (5% increase in preload) to severe risk (20% increase in preload). The preload adjustment command is converted into a control signal by the drive control circuit, driving the actuator to perform precise preload adjustment.

[0164] Step S4: Effect Feedback and Verification After the preload adjustment is completed, the system continues to monitor the vibration and strain state of the bolt to verify the adjustment effect. Verification indicators include the reduction in vibration RMS value and the degree of improvement in preload stability. If the adjustment effect is not ideal, the system will initiate a secondary adjustment, using a Bayesian optimization method to gradually approach the optimal preload value.

[0165] Step S5: The anti-loosening strategy optimization system continuously learns the loosening patterns and anti-loosening effects under different working conditions through a machine learning module, and continuously optimizes the anti-loosening strategy. The optimization process adopts a multi-objective optimization method, simultaneously considering anti-loosening effect, energy consumption, and system lifespan. Through continuous learning and optimization, the system can adapt to various complex working conditions and continuously improve anti-loosening performance.

[0166] Example 2: Intelligent Dynamic Anti-Loosening System for High-Speed ​​Railway Fasteners

[0167] The intelligent dynamic anti-loosening system for high-speed rail fasteners provided by this invention includes a parameter monitoring module, a preload adjustment module, and an intelligent control module.

[0168] The parameter monitoring module includes a multi-axis accelerometer, strain gauges, and a data acquisition unit. The multi-axis accelerometer is used to monitor the vibration acceleration of high-speed rail fasteners (bolt assemblies) in the X, Y, and Z directions, with a measurement range of ±10g and an accuracy of 0.01g. The strain gauges are foil resistance strain gauges, attached to the bolt surface, used to monitor the strain state of the bolt. The data acquisition unit uses a high-speed ADC chip with a sampling frequency of 100Hz to convert the sensor signals into digital signals and transmit them to the intelligent control module.

[0169] The preload adjustment module includes a miniature hydraulic actuator (or electric actuator) and a drive control circuit. The miniature hydraulic actuator is installed at the high-speed rail fastener and can apply axial force to adjust the bolt preload, with an adjustment range of 10kN-100kN. The drive control circuit receives control signals from the intelligent control module and drives the actuator to perform precise preload adjustment.

[0170] The intelligent control module, based on an embedded processor, includes a data processing module, a vibration analysis module, a machine learning module, and a preload control module. The data processing module receives and preprocesses sensor data; the vibration analysis module performs time-domain and frequency-domain analysis of vibration data; the machine learning module learns loosening patterns under different working conditions based on historical data and predicts loosening trends; and the preload control module generates preload adjustment commands based on the analysis results, controlling the actions of the preload adjustment module.

[0171] To verify the effectiveness of this intelligent dynamic anti-loosening system for high-speed rail fasteners, several experimental tests were conducted:

[0172] 1. Transverse vibration test verification experimental setup: electromagnetic vibration test bench, frequency range 5-2000Hz; test sample: M20×2.5 high-strength alloy steel bolt, initial preload value is 85% of standard preload; control group: three traditional anti-loosening methods are used: double nut, spring washer, and nylon insert anti-loosening washer; experimental group: equipped with the system of this invention.

[0173] Experimental results: After 20 hours of continuous vibration testing, the preload retention rate of traditional anti-loosening methods was 68.7%-77.6%, while the preload retention rate of the system of this invention reached 94.8%-96.2%. Figure 2 As shown in the figure, especially in the 150-200Hz frequency band (the resonant frequency region of the high-speed rail wheel-track system), the retention rate of traditional anti-loosening methods decreased significantly, with a maximum decrease of 32%, while the system of this invention only decreased by 4.6%. In the figure, the curves from bottom to top represent the double nut, spring washer, nylon insert anti-loosening washer, and experimental group, respectively.

[0174] 2. Fatigue life testing experimental apparatus: hydraulic servo fatigue testing machine; test sample: M16×2.0 high-strength alloy steel bolt; control group: high-speed rail fasteners with nylon insert self-locking nuts and coated with anaerobic adhesive; experimental group: equipped with the system of this invention.

[0175] Experimental results: After 10^7 cycles, the bolt preload in the control group decreased by more than 40% on average, and some samples showed fatigue cracks; while the bolt preload in the experimental group remained above 89%, with no loosening or cracking, and the fatigue life was extended by an average of 40.4%. This system exhibited excellent stability under different preload conditions.

[0176] 3. Anti-loosening performance test under different preload conditions: Under low preload (30kN) conditions, the preload retention rate of the control group bolts was only 43.2% (room temperature) and 38.7% (high temperature); while the preload retention rate of the experimental group bolts reached 84.8% and 81.5% respectively, with an anti-loosening performance improvement of over 100%. Figure 3 As shown, this system can accurately predict the risk of loosening under different conditions, with a prediction accuracy of over 85%.

[0177] Application Cases

[0178] The intelligent dynamic anti-loosening system for high-speed rail fasteners has been applied and verified in high-speed rail track fasteners. A six-month experimental deployment was conducted on a test section of the Beijing-Shanghai High-Speed ​​Railway. Fasteners equipped with this intelligent dynamic anti-loosening system showed an average preload reduction of only 4.3%, far lower than the 25.7% reduction of traditional anti-loosening fasteners; the track geometry was significantly improved, with a 36.8% reduction in gauge variation and a 42.3% increase in track smoothness.

[0179] The intelligent dynamic anti-loosening system for high-speed rail fasteners has been applied and verified in the suspension system of high-speed trains. This intelligent dynamic anti-loosening system was applied to key bolts in the suspension system of high-speed trains, with a cumulative test distance of 50,000 kilometers. Bogies equipped with this system showed a 15.3% reduction in suspension system vibration, a 17.9% improvement in ride comfort, and bolt preload fluctuations controlled within ±7%, while the preload of traditional anti-loosening bolts decreased by up to 32.8%.

[0180] The intelligent dynamic anti-loosening system for high-speed rail fasteners has been validated in brake disc bolts. This system was applied to the fixing bolts of high-speed train brake discs. Under emergency braking conditions, the preload of traditional bolts decreased by an average of 21.6%, while the preload of bolts using this intelligent dynamic anti-loosening system decreased by only 6.8%. The radial runout of the brake disc was reduced by 34.7%, extending the service life of the brake disc and brake pads.

[0181] In summary, the intelligent dynamic anti-loosening control method and system for high-speed rail fasteners provided by this invention obtains working parameters in multiple dimensions and determines the type of loosening of the fasteners based on these parameters, thereby generating a preload adjustment scheme and correcting the fastening state, significantly improving the accuracy and timeliness of risk identification. Furthermore, it continues to receive working parameters of the high-speed rail fasteners within a preset time period and continuously optimizes and corrects them. Experimental verification shows that it can maintain the preload of high-speed rail fasteners at more than 95% of the initial value, reducing the loosening rate by 80%. In cyclic vibration testing, the fatigue life of high-speed rail fasteners is extended by 40%. Compared with traditional methods, the anti-loosening performance is improved by more than 60% under low preload and high vibration conditions. Besides the high-speed rail field, this control method can also be applied to bridges, aerospace, and other fields with high requirements for fastener anti-loosening performance.

[0182] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A high-iron fastener intelligent dynamic anti-loose control method, characterized in that, Comprising: S1, receiving working parameters of high-speed rail fasteners, wherein the working parameters include vibration data and pretightening force data; S2, analyzing the working parameters, if the working parameters exceed a preset range, judging the type of loosening according to the working parameters and generating a pretightening force adjustment scheme; S3, executing the pretightening force adjustment scheme; S4, continuously receiving the working parameters of the high-speed rail fasteners within a preset time length, and repeating the steps of S2 to S3.

2. The high rail fastener intelligent dynamic anti-loose control method according to claim 1, characterized in that, The step of S4 comprises: S5, optimizing the generation mode of the pretightening force adjustment scheme according to all the pretightening force adjustment schemes and the working parameters of the high-speed rail fasteners after executing the pretightening force adjustment schemes.

3. The high rail fastener intelligent dynamic anti-loose control method according to claim 2, characterized in that, The step of S5 comprises: S5.1, defining optimization objectives, wherein the optimization objectives include maximizing the anti-loosening performance and minimizing the energy consumption; S5.2, defining decision variables, wherein the decision variables include the pretightening force adjustment amount, the pretightening force adjustment frequency and the pretightening force adjustment timing; S5.3, solving the optimal anti-loosening strategy by executing the non-dominated sorting genetic algorithm II according to all the pretightening force adjustment schemes and the working parameters of the high-speed rail fasteners after executing the pretightening force adjustment schemes.

4. The high rail fastener intelligent dynamic anti-loose control method according to claim 1, characterized in that, The step of S2 comprises: extracting features of the working parameters and analyzing them, if the working parameters exceed a preset range, judging the type of loosening by a machine learning module, and generating a pretightening force adjustment scheme, wherein the machine learning module is trained according to historical vibration data and loosening states.

5. The high rail fastener intelligent dynamic anti-loose control method according to any one of claims 1 to 4, characterized in that, In the step of S2, the pretightening force adjustment scheme includes the pretightening force adjustment amount, the pretightening force adjustment rate and the pretightening force adjustment timing.

6. The high rail fastener intelligent dynamic anti-loose control method according to any one of claims 1 to 4, characterized in that, In the step of S2, the generation of the pretightening force adjustment scheme adopts a multi-objective optimization algorithm, wherein the optimization objectives include maximizing the anti-loosening performance of the high-speed rail fasteners, minimizing the energy consumption in the pretightening force adjustment scheme and maintaining the long-term stability of the high-speed rail fasteners.

7. The high rail fastener intelligent dynamic anti-loose control method according to any one of claims 1 to 4, characterized in that, In the step of S1, the working parameters include the pretightening force data and the vibration data in three-axis directions; The step of S2 comprises: when the vibration data shows that the vibration amplitude of the high-speed rail fastener exceeds a set amplitude threshold, the pretightening force data shows that the pretightening force of the high-speed rail fastener decreases by more than a set percentage, or the energy in a specific frequency range of the power spectral density increases by more than a set value, then judging the type of loosening according to the working parameters and generating a pretightening force adjustment scheme.

8. The high rail fastener intelligent dynamic anti-loose control method according to claim 7, characterized in that, The increase of energy in a specific frequency range of the power spectral density is realized by vibration analysis based on the power spectral density, frequency domain analysis is performed on multi-axis vibration data, and frequency domain feature parameters are calculated, wherein the mathematical representation of the power spectral density analysis is: P(f) = lim(T→∞)[|X(f)|2 / T] wherein X(f) is the Fourier transform of the signal x(t) and T is the sampling time.

9. A system for performing the method of intelligent dynamic anti-loosening control of high-strength fasteners according to any one of claims 2 to 3, characterized in that Comprising: a parameter monitoring module for obtaining and analyzing the vibration data and the pretightening force data, the parameter monitoring module comprising a vibration monitoring module and a pretightening force monitoring module, the vibration monitoring module being used to monitor the vibration acceleration and vibration frequency of the high-speed rail fastener in multiple directions, and the pretightening force monitoring module being used to monitor the pretightening force at the connection of the high-speed rail fastener; The intelligent control module is configured to determine the type of looseness and generate a pre-tightening force adjustment scheme according to the vibration data and the pre-tightening force data. The pre-tightening force adjustment module is configured to execute the pre-tightening force adjustment scheme and adjust the pre-tightening force of the high-speed rail fastener.

10. The system of claim 9, wherein, After the pre-tightening force adjustment scheme is executed, the intelligent control module is further configured to continue to receive the working parameters of the high-speed rail fastener within a preset time length, continue to generate the pre-tightening force adjustment scheme, and optimize the generation mode of the pre-tightening force adjustment scheme according to all the pre-tightening force adjustment schemes.

Images