A fatigue life prediction method for ultra-high performance concrete

Abstract

The present invention discloses a fatigue life prediction method for ultra-high performance concrete, which relates to the technical field of concrete fatigue life prediction and includes the following steps: deploying high-frequency response sensors at key load input positions of high-performance concrete components, continuously collecting load time series signals of the dynamic response of high-performance concrete under load, and providing basic data for the entire fatigue life prediction process; inputting the collected load time series signals into the fast Fourier transform algorithm to complete the conversion from time domain to frequency domain. The present invention extracts high-frequency load characteristics through high-frequency response sensors and fast Fourier transform, constructs a load frequency intensity index based on the tail tailing phenomenon of the spectrum, and introduces a micro-damage threshold to identify fatigue degradation risks in real time. The fatigue life reduction is dynamically applied according to the frequency intensity exceeding the amplitude, thereby realizing the dynamic evolution of fatigue life prediction and significantly improving the safety and reliability of high-performance concrete components in high-frequency load environments.

Description

Technical Field

[0001] The present invention relates to the technical field of concrete fatigue life prediction, and in particular to a fatigue life prediction method for ultra-high performance concrete. Background Art

[0002] Ultra-high-performance concrete (UHPC) fatigue life prediction involves calculating the number of cycles or service life that a UHPC component can withstand under cyclic loading at a specific stress level and load frequency, based on the material's internal damage evolution characteristics, microcrack propagation patterns, and mechanical property degradation trends. This method uses experimental data, numerical modeling, or machine learning methods to estimate the number of cycles or service life that a UHPC component can withstand at a specific stress level and load frequency. This method provides a scientific basis for engineering structural design, service life safety assessment, and maintenance decision-making, thereby preventing sudden failures caused by accumulated fatigue damage and improving the reliability and structural durability of UHPC in key areas such as bridges and super-high-rise buildings, thereby extending the project's service life and reducing maintenance costs.

[0003] The existing technology has the following deficiencies:

[0004] In existing technologies, under the repeated action of cyclic loads (such as vehicle loads and wind vibration loads), micro cracks will gradually form inside ultra-high performance concrete (UHPC). The cracks will expand and connect with the load cycle, eventually leading to macroscopic structural damage. Fatigue life prediction technology aims to quantify the above-mentioned evolution process from initial micro damage to final fracture. However, some UHPC materials exhibit the so-called "fatigue limit" phenomenon under low stress amplitude conditions, that is, they have the characteristics of ultra-long life; but under extremely high load frequency environments, even small load amplitudes may cause accelerated accumulation of micro damage, resulting in rapid deterioration of actual fatigue performance. Existing fatigue life prediction methods based on SN curves usually fail to fully capture such high-frequency micro damage evolution effects, and are prone to seriously underestimate the failure risk of UHPC components in high-frequency load environments such as high-speed railway bridges and wind turbine towers, which may eventually lead to sudden fracture accidents at low stress levels.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0006] The purpose of the present invention is to provide a fatigue life prediction method for ultra-high performance concrete to solve the problems in the above-mentioned background technology.

[0007] In order to achieve the above object, the present invention provides the following technical solution: a method for predicting fatigue life of ultra-high performance concrete, comprising the following steps:

[0008] High-frequency response sensors are deployed at key load input locations (such as stress concentration areas, support points, and connection nodes) of high-performance concrete components to continuously collect load time series signals of the dynamic response of high-performance concrete under load, providing basic data for the entire fatigue life prediction process.

[0009] The collected load time series signals are input into the Fast Fourier Transform (FFT) algorithm to complete the conversion from the time domain to the frequency domain, so that the high-frequency load components can be identified and quantified;

[0010] In the obtained load spectrum, identify the main frequency peak with the highest energy, determine its corresponding frequency, and mark it as the load main frequency;

[0011] Extract high-frequency load characteristics based on the identified load dominant frequency. Perform in-depth feature engineering analysis on the rolling extracted high-frequency load characteristic data within a set short-term sliding monitoring window. Build a load frequency intensity index based on the analyzed data to quantitatively characterize the high-frequency characteristics and intensity level of the current load.

[0012] Set a micro-damage threshold (determined based on material test data or empirical calculations) and compare the current load frequency intensity with the micro-damage threshold in real time. If the load frequency intensity exceeds the micro-damage threshold, it is determined that the load has an accelerating effect on the accumulation of micro-damage in concrete, triggering the subsequent life adjustment mechanism. The fatigue life reduction is dynamically applied according to the excess amplitude of the load frequency intensity, and the original fatigue life prediction results are intelligently adjusted.

[0013] Preferably, the frequency response range of the high-frequency response sensor needs to cover at least twice the target load frequency to meet the requirements of the Nyquist sampling theorem and ensure the accuracy of subsequent frequency analysis.

[0014] Preferably, the collected load time series signal is input into the Fast Fourier Transform (FFT) algorithm to complete the conversion from the time domain to the frequency domain. The specific steps are as follows:

[0015] First, the collected load time series signals are preliminarily processed, including removing DC offset, filtering and denoising, and unifying the sampling frequency to ensure that the signal quality meets the requirements of frequency domain analysis.

[0016] Then, select an appropriate window function (such as Hanning window or Hamming window) to perform windowing on the time series to reduce the spectrum leakage effect;

[0017] Next, the windowed time series signal is input into the Fast Fourier Transform (FFT) algorithm to perform a discrete Fourier transform operation to obtain a complex frequency domain output.

[0018] Then, the amplitude spectrum of the complex output is calculated to extract the energy corresponding to each frequency component;

[0019] Finally, the frequency axis is calibrated, the frequency resolution is calculated based on the sampling rate and data length, and a frequency-amplitude spectrum is generated, providing a basis for subsequent main frequency extraction and load high-frequency characteristic analysis.

[0020] Preferably, in the obtained load spectrum, the main frequency peak with the highest energy is identified, and its corresponding frequency is determined and marked as the load main frequency. The specific steps are as follows:

[0021] First, data extraction is performed on the load spectrum obtained by fast Fourier transform (FFT), and frequency-amplitude pairs are extracted to form a frequency component array;

[0022] Next, a peak detection algorithm, such as a method for finding local maxima, is applied to traverse the frequency component array and identify the frequency point with the largest amplitude. To avoid misidentifying noise or false peaks, an energy threshold is set to retain only valid peaks exceeding the threshold. Neighborhood smoothing or multi-scale peak recognition techniques can be combined to further improve accuracy.

[0023] Finally, the frequency point with the highest energy and meeting the effectiveness screening conditions is extracted, its corresponding frequency value is recorded, and it is officially marked as the main frequency of the load, which serves as an important basis for subsequent load high-frequency feature extraction and fatigue life prediction analysis.

[0024] Preferably, high-frequency load features are extracted around the identified load main frequency, wherein the specific features extracted are: the degree of energy tailing formed by the slow decay of spectrum energy in the high-frequency tail (away from the main frequency area); within the set short-time sliding monitoring window, deep feature engineering analysis is performed on the rolling extracted high-frequency load feature data to generate a spectrum tail drag reference value; based on the spectrum tail drag reference value, a load frequency intensity index is constructed to quantitatively characterize the high-frequency characteristics and intensity level of the current load.

[0025] Preferably, within a set short-time sliding monitoring window, after performing a deep feature engineering analysis on the energy tailing degree caused by the slow decay of spectrum energy in the high-frequency tail (away from the main frequency area), a spectrum tail drag reference value is generated. The specific steps are as follows:

[0026] In the set short-time sliding monitoring window, the load spectrum data is first segmented and the high-frequency tail frequency range away from the main frequency area is selected. , extract the amplitude spectrum corresponding to each frequency point in the high-frequency tail frequency range, and calculate the energy cumulative growth curve in ascending order of frequency based on the amplitude spectrum. The calculation expression is: ,in: Indicates the starting frequency from the high-frequency tail The accumulated energy between λ and frequency f; It is the spectrum amplitude at frequency point x, and the square represents the energy density; dx represents the frequency interval element;

[0027] The spectrum tail drag reference value is calculated based on the energy accumulation curve. The spectrum tail drag reference value is the weighted normalized inverse form of the high-frequency tail cumulative energy growth rate. The specific calculation expression is: ,in: is the reference value of spectrum tail drag, is the growth rate of the energy accumulation curve at frequency f; is a nonlinear enhancement exponential parameter used to emphasize the slow energy growth in the high-frequency tail region ( weighted amplification of the slow-growth part); is the starting frequency of the high-frequency tail; is the high-frequency tail end frequency, by introducing The power transformation and integration can effectively amplify the slow growth trend of high-frequency tail energy and make the dragging phenomenon more prominent. The power is used to make the reference value of the spectrum tail drag have a stable scale, which is convenient for comparison across samples.

[0028] Preferably, the current spectrum tail drag reference value is compared with the micro-damage threshold to determine whether the load has an accelerating effect on the accumulation of concrete micro-damage. The specific steps are as follows:

[0029] If the spectrum tail drag reference value is greater than the micro-damage threshold, it is determined that the load has an accelerating effect on the accumulation of micro-damage in concrete, triggering the subsequent life adjustment mechanism;

[0030] If the spectrum tail drag reference value is less than or equal to the micro-damage threshold, it is determined that the load has no accelerating effect on the accumulation of micro-damage in concrete, and the original fatigue life prediction is maintained.

[0031] Preferably, when it is determined that the load has an accelerating effect on the accumulation of micro-damage in concrete, fatigue life reduction is dynamically applied according to the excess amplitude of the load frequency intensity, and the original fatigue life prediction result is intelligently adjusted. The specific steps are as follows:

[0032] After determining that the load has an accelerating effect on the accumulation of micro-damage in concrete, the excess amplitude of the spectrum tail drag reference value relative to the micro-damage threshold is first calculated, and then a nonlinear normalization process is performed on it to obtain the dynamic weighting factor. The specific calculation formula is as follows: ,in: is the dynamic weighting factor, is the spectrum tail drag reference value obtained by real-time monitoring of the current short-term sliding monitoring window, is the preset micro-damage frequency threshold, The upper limit of the reasonable maximum spectrum tail drag reference value set by the system, is the nonlinear enhancement factor ( Amplify the excess amplitude when processing sensitive information);

[0033] Based on the dynamic weighting factor obtained in the first step , adaptive compression is applied to the original fatigue life prediction value to generate an intelligently adjusted fatigue life prediction value. The specific compression formula is as follows: ,in: is the original fatigue life prediction value obtained based on the traditional method, is the fatigue life prediction value after adaptive compression adjustment, is the life reduction sensitivity coefficient (which can be set according to specific material properties or safety level requirements), is the life reduction acceleration index, which is used to further adjust the response curve shape between the compression amplitude and the frequency intensity exceeding degree ( The reduction curve is steeper).

[0034] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0035] This invention deploys high-frequency response sensors to capture dynamic load changes in real time, combines fast Fourier transforms to extract the main frequency and high-frequency load characteristics, constructs a load frequency intensity index based on the energy tailing phenomenon at the end of the spectrum, and introduces a micro-damage threshold to identify the accelerating effect of load on micro-damage accumulation in real time. This allows for accurate identification of potential fatigue degradation risks under extremely high-frequency load conditions. Furthermore, fatigue life reduction is dynamically applied based on the excess amplitude of load frequency intensity, and life prediction results are intelligently adjusted, transforming fatigue life prediction from static inference to dynamic evolutionary response. This significantly improves the safety, reliability, and scientific life management of high-performance concrete components in high-frequency load service scenarios such as high-speed railway bridges, wind turbine towers, and high-rise building wind-resistant structures. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction to the drawings required for use in the embodiments will be given below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0037] Figure 1 The present invention is a method flow chart of a fatigue life prediction method for ultra-high performance concrete. DETAILED DESCRIPTION

[0038] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art.

[0039] The present invention provides Figure 1 A fatigue life prediction method for ultra-high performance concrete is shown, comprising the following steps:

[0040] High-frequency response sensors are deployed at key load input locations (such as stress concentration areas, support points, and connection nodes) of high-performance concrete components to continuously collect load time series signals of the dynamic response of high-performance concrete under load, providing basic data for the entire fatigue life prediction process.

[0041] High-frequency response sensors, such as strain gauges, accelerometers, or dynamic pressure sensors. The frequency response range of high-frequency response sensors must cover at least twice the target load frequency to meet the requirements of the Nyquist sampling theorem and ensure the accuracy of subsequent frequency analysis. The sensor needs to have high sensitivity and low noise characteristics, and be equipped with a synchronous acquisition system (multi-channel DAQ device) to ensure data timing alignment and avoid phase distortion. Through this high-precision real-time monitoring, it is possible to continuously collect complete time series signals reflecting the dynamic response of high-performance concrete under load, providing a reliable and accurate raw data foundation for subsequent frequency domain analysis, providing basic data for the entire fatigue life prediction process, and ensuring the accuracy and completeness of subsequent frequency analysis.

[0042] The collected load time series signals are input into the Fast Fourier Transform (FFT) algorithm to complete the conversion from the time domain to the frequency domain, so that the high-frequency load components can be identified and quantified;

[0043] The collected load time series signal is input into the Fast Fourier Transform (FFT) algorithm to complete the conversion from time domain to frequency domain. The specific steps are as follows:

[0044] First, the collected load time series signals are preliminarily processed, including removing DC offset, filtering and denoising, and unifying the sampling frequency to ensure that the signal quality meets the requirements of frequency domain analysis.

[0045] Then, select an appropriate window function (such as Hanning window or Hamming window) to perform windowing on the time series to reduce the spectrum leakage effect;

[0046] Next, the windowed time series signal is input into the Fast Fourier Transform (FFT) algorithm to perform a discrete Fourier transform operation to obtain a complex frequency domain output.

[0047] Then, the amplitude spectrum of the complex output is calculated to extract the energy corresponding to each frequency component;

[0048] Finally, the frequency axis is calibrated, the frequency resolution is calculated based on the sampling rate and data length, and a frequency-amplitude spectrum is generated, providing a basis for subsequent main frequency extraction and load high-frequency characteristic analysis.

[0049] Through these steps, the collected time domain load signal is successfully converted into a frequency domain signal, which can provide valuable frequency component data for subsequent frequency analysis and fatigue life prediction.

[0050] In the obtained load spectrum, identify the main frequency peak with the highest energy, determine its corresponding frequency, and mark it as the load main frequency;

[0051] The specific steps are as follows:

[0052] First, data extraction is performed on the load spectrum obtained by fast Fourier transform (FFT), and frequency-amplitude pairs are extracted to form a frequency component array;

[0053] Next, a peak detection algorithm, such as a method for finding local maxima, is applied to traverse the frequency component array and identify the frequency point with the largest amplitude. To avoid misidentifying noise or false peaks, an energy threshold is set to retain only valid peaks exceeding the threshold. Neighborhood smoothing or multi-scale peak recognition techniques can be combined to further improve accuracy.

[0054] Finally, the frequency point with the highest energy and meeting the effectiveness screening conditions is extracted, its corresponding frequency value is recorded, and it is officially marked as the main frequency of the load, which serves as an important basis for subsequent load high-frequency feature extraction and fatigue life prediction analysis.

[0055] The main frequency extraction result is used to characterize the dominant cyclic frequency under load. It is the core input parameter for dynamic adjustment of fatigue life. Accurately defining the dominant frequency of fatigue action can provide a basic basis for subsequent fatigue damage rate correction.

[0056] Extract high-frequency load characteristics based on the identified load dominant frequency. Perform in-depth feature engineering analysis on the rolling extracted high-frequency load characteristic data within a set short-term sliding monitoring window. Build a load frequency intensity index based on the analyzed data to quantitatively characterize the high-frequency characteristics and intensity level of the current load.

[0057] High-frequency load features are extracted based on the identified load main frequency. The specific features extracted are: the degree of energy tailing formed by the slow decay of spectrum energy in the high-frequency tail (away from the main frequency area). Within the set short-term sliding monitoring window, deep feature engineering analysis is performed on the rolling extracted high-frequency load feature data to generate a spectrum tail drag reference value. Based on the spectrum tail drag reference value, a load frequency intensity index is constructed to quantitatively characterize the high-frequency characteristics and intensity level of the current load.

[0058] When performing frequency domain analysis on a load signal, the spectrum reflects the energy distribution corresponding to different frequency components. For typical low-frequency loads (such as slow traffic and temperature loads), energy is highly concentrated in the low-frequency region and rapidly decays upon entering the high-frequency region, resulting in a steep drop in the spectrum. However, when the load frequency is very high, due to the rapid rate of change and short cycle period, the signal contains a large amount of high-frequency components, causing energy to persist in the high-frequency region and decay more slowly, resulting in a pronounced high-frequency tail. A larger tail indicates a richer high-frequency component in the signal and a longer-lasting high-frequency energy retention, reflecting the high-speed alternation and wide-frequency distribution of the load system. Therefore, it serves as an important characteristic for determining whether the load frequency is very high. This phenomenon is particularly pronounced when ultra-high-performance concrete components are subjected to complex environmental loads such as high-speed trains and high-speed wind vibrations from wind turbine towers.

[0059] Within the set short-term sliding monitoring window, after performing in-depth feature engineering analysis on the energy tailing caused by the slow decay of spectrum energy in the high-frequency tail (away from the main frequency area), a spectrum tail tail drag reference value is generated. The specific steps are as follows:

[0060] In the set short-time sliding monitoring window, the load spectrum data is first segmented and the high-frequency tail frequency range away from the main frequency area is selected. , extract the amplitude spectrum corresponding to each frequency point in the high-frequency tail frequency range, and calculate the energy cumulative growth curve in ascending order of frequency based on the amplitude spectrum. The calculation expression is: ,in: Indicates the starting frequency from the high-frequency tail The accumulated energy between λ and frequency f; It is the spectrum amplitude at frequency point x, and the square represents the energy density; dx represents the frequency interval element;

[0061] This integral accumulation accurately depicts the accumulation rate of energy at the tail of the spectrum as frequency increases, avoiding the impact of energy sparseness or local mutations. This step provides basic data for subsequent quantification of the tailing degree by constructing the energy evolution trajectory of the high-frequency tail of the spectrum.

[0062] The spectrum tail drag reference value is calculated based on the energy accumulation curve. The spectrum tail drag reference value is the weighted normalized inverse form of the high-frequency tail cumulative energy growth rate. The specific calculation expression is: ,in: is the reference value of spectrum tail drag, is the growth rate of the energy accumulation curve at frequency f; is a nonlinear enhancement exponential parameter used to emphasize the slow energy growth in the high-frequency tail region ( weighted amplification of the slow-growth part); is the starting frequency of the high-frequency tail; is the high-frequency tail end frequency, by introducing The power transformation and integration can effectively amplify the slow growth trend of high-frequency tail energy and make the dragging phenomenon more prominent. The power,makes the reference value of spectrum tail drag have a stable scale, which is convenient for,comparison across samples;

[0063] This step quantifies the degree of tailing of energy accumulation in the high-frequency tail of the spectrum. The larger the value, the richer the high-frequency components and the higher the load frequency.

[0064] The Spectral Tail Drag reference value, generated through deep feature engineering analysis of the degree of energy smearing caused by the slow decay of spectral energy in the high-frequency tail (away from the main frequency region) within a set short-term sliding monitoring window, indicates that a larger value indicates a higher load frequency, while a smaller value indicates a lower load frequency. This is because the Spectral Tail Drag reference value essentially quantifies the accumulation of high-frequency energy in the tail of the payload signal spectrum as frequency increases within the set short-term sliding monitoring window. When the load frequency is high, the signal contains more high-frequency components. Therefore, energy decay becomes slower in the high-frequency tail away from the main frequency region, resulting in a noticeable smearing phenomenon. The Spectral Tail Drag reference value, generated through deep feature engineering analysis, quantifies the degree of this slow energy decay. Therefore, a larger spectral tail drag reference value indicates a more significant accumulation of high-frequency tail energy and a higher proportion of high-frequency components in the load, indicating a higher overall load frequency. Conversely, a smaller spectral tail drag reference value indicates a rapid attenuation of high-frequency component energy and a sparse high-frequency tail energy, reflecting a lower load frequency. This spectral tail drag reference value can accurately capture the changing trends of the load frequency environment from the perspective of microscopic energy distribution characteristics, making it particularly suitable for quantitative analysis of high-frequency characteristics under complex dynamic load conditions.

[0065] Set a micro-damage threshold (determined based on material test data or empirical calculations) and compare the current load frequency intensity with the micro-damage threshold in real time. If the load frequency intensity exceeds the micro-damage threshold, it is determined that the load has an accelerating effect on the accumulation of micro-damage in concrete, triggering the subsequent life adjustment mechanism. The fatigue life reduction is dynamically applied according to the excess magnitude of the load frequency intensity, and the original fatigue life prediction results are intelligently adjusted.

[0066] Compare the current spectrum tail drag reference value with the micro-damage threshold to determine whether the load has an accelerating effect on the accumulation of concrete micro-damage. The specific steps are as follows:

[0067] If the spectrum tail drag reference value is greater than the micro-damage threshold, it is determined that the load has an accelerating effect on the accumulation of micro-damage in concrete, triggering the subsequent life adjustment mechanism;

[0068] If the spectrum tail drag reference value is less than or equal to the micro-damage threshold, it is determined that the load has no accelerating effect on the accumulation of micro-damage in concrete, and the original fatigue life prediction is maintained.

[0069] When it is determined that the load has an accelerating effect on the accumulation of micro-damage in concrete, fatigue life reduction is dynamically applied according to the excess amplitude of the load frequency intensity, and the original fatigue life prediction results are intelligently adjusted. The specific steps are as follows:

[0070] After determining that the load has an accelerating effect on the accumulation of micro-damage in concrete, the excess amplitude of the spectrum tail drag reference value relative to the micro-damage threshold is first calculated, and then a nonlinear normalization process is performed on it to obtain the dynamic weighting factor. The specific calculation formula is as follows: ,in: is the dynamic weighting factor, is the spectrum tail drag reference value obtained by real-time monitoring of the current short-term sliding monitoring window, is the preset micro-damage frequency threshold, The upper limit of the reasonable maximum spectrum tail drag reference value set by the system, is the nonlinear enhancement factor ( Amplify the excess amplitude when processing sensitive information);

[0071] Through this normalization and nonlinear adjustment process, the degree to which the current load frequency exceeds the threshold is accurately quantified, ensuring that the greater the frequency intensity exceeds, the higher the weight of the impact on the life adjustment, while effectively avoiding interference from extreme values.

[0072] Based on the dynamic weighting factor obtained in the first step , adaptive compression is applied to the original fatigue life prediction value to generate an intelligently adjusted fatigue life prediction value. The specific compression formula is as follows: ,in: is the original fatigue life prediction value obtained based on the traditional method, is the fatigue life prediction value after adaptive compression adjustment, is the life reduction sensitivity coefficient (which can be set according to specific material properties or safety level requirements), is the life reduction acceleration index, which is used to further adjust the response curve shape between the compression amplitude and the frequency intensity exceeding degree ( The reduction curve is steeper when

[0073] Through exponential compression (rather than nonlinear or linear direct multiplication and contraction), the sensitivity of fatigue life to changes in load frequency intensity can be finely controlled, and dynamic adaptive adjustment of life prediction can be achieved, ensuring that the life is quickly reduced when the frequency intensity exceeds a large limit, and only mild adjustment is made when it exceeds slightly, avoiding excessive conservatism or excessive radicalism.

[0074] In the fatigue life prediction process for ultra-high performance concrete (UHPC), a micro-damage threshold is set and compared with the load frequency intensity obtained from real-time monitoring. The core purpose is to achieve intelligent identification and dynamic response to potential high-frequency damage acceleration effects within the loading environment, thereby ensuring that fatigue life prediction results more accurately reflect the damage evolution trend of concrete under actual service conditions. By setting the micro-damage threshold based on material test data or long-term empirical calculations, a physically meaningful discriminant criterion can be introduced into fatigue analysis, effectively distinguishing normal load frequency states from abnormal high-frequency damage accumulation states. If the load frequency intensity exceeds this threshold, that is, the reference value of the spectrum tail drag exceeds the micro-damage threshold, it means that the microcrack propagation rate within the concrete has significantly accelerated, and micro-damage accumulation has entered an accelerated stage. If traditional static fatigue life prediction models are still used at this time, problems such as life overestimation and inaccurate risk assessment are likely to occur, leading to structural safety hazards. Therefore, timely triggering of subsequent life adjustment mechanisms when the frequency intensity exceeds the threshold becomes a key component of the fatigue life management system to dynamically perceive and rapidly respond to environmental changes.

[0075] Specifically, the dynamic application of fatigue life reduction intelligently adjusts the original life prediction based on the extent to which the spectral tail drag reference value exceeds the micro-damage threshold. This ensures that the life prediction not only relies on initial material properties and design load levels but also reflects in real time the impact of changing load characteristics on damage rates during actual service. This adaptive compression mechanism breaks the static limitations of traditional fatigue life prediction, which relies on "fixed curve inference of life," transforming life prediction into an intelligent decision-making process that dynamically evolves with the load environment. When the load frequency intensity slightly exceeds the threshold, only a small life reduction is applied, avoiding overly conservative resource waste. When the load frequency intensity significantly exceeds the threshold, the life prediction value is rapidly reduced, providing timely warning of potential failure risks. This dynamic life adjustment not only improves the accuracy and sensitivity of fatigue life predictions but also facilitates the development of more appropriate maintenance strategies, inspection frequencies, and structural health management plans, maximizing the safe service life of structures while controlling maintenance costs. Therefore, this process of setting a micro-damage threshold and dynamically reducing fatigue life based on frequency intensity is both a core technical step in improving the scientific accuracy of fatigue prediction and a key mechanism for ensuring the long-term reliability of high-performance concrete structures, playing an indispensable and important role.

[0076] The above-mentioned solution significantly improves the fatigue life prediction accuracy of ultra-high-performance concrete (UHPC) under high-frequency loading conditions, enabling dynamic perception and adaptive correction of fatigue damage evolution, overcoming the limitations of traditional static SN curve-based methods that overestimate life under high-frequency, micro-damage-accelerated conditions. By deploying high-frequency response sensors to capture dynamic load changes in real time, combining fast Fourier transform (FFT) to extract the main frequency and high-frequency load characteristics, constructing a load frequency intensity index based on the energy tailing phenomenon at the end of the spectrum, and introducing a micro-damage threshold to identify the accelerating effect of load on micro-damage accumulation in real time, the method can accurately identify potential fatigue degradation risks under extremely high-frequency loading conditions. Furthermore, fatigue life reduction is dynamically applied based on the magnitude of the load frequency intensity excess, and life prediction results are intelligently adjusted, transforming fatigue life prediction from static inference to dynamic evolutionary response. This significantly improves the safety, reliability, and scientific life management of high-performance concrete components in high-frequency loading scenarios, such as high-speed railway bridges, wind turbine towers, and high-rise building wind-resistant structures.

[0077] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0078] The above description is merely illustrative of certain exemplary embodiments of the present invention. It goes without saying that those skilled in the art will be able to modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and description are illustrative in nature and should not be construed as limiting the scope of protection of the claims.

[0079] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0080] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0081] In the several embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection of some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0082] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0083] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0084] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program code, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0085] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A method for predicting fatigue life of ultra-high performance concrete, characterized in that: The following steps are involved: High-frequency response sensors are deployed at key load input locations on high-performance concrete components to continuously collect load time series signals of the dynamic response of high-performance concrete under load, providing basic data for the entire fatigue life prediction process. The collected load time series signals are input into the fast Fourier transform algorithm to complete the conversion from time domain to frequency domain, so that the high-frequency load components can be identified and quantified; In the obtained load spectrum, identify the main frequency peak with the highest energy, determine its corresponding frequency, and mark it as the load main frequency; Extract high-frequency load characteristics based on the identified load dominant frequency. Perform in-depth feature engineering analysis on the rolling extracted high-frequency load characteristic data within a set short-term sliding monitoring window. Build a load frequency intensity index based on the analyzed data to quantitatively characterize the high-frequency characteristics and intensity level of the current load. Set a micro-damage threshold and compare the current load frequency intensity with the micro-damage threshold in real time. If the load frequency intensity exceeds the micro-damage threshold, it is determined that the load has an accelerating effect on the accumulation of micro-damage in concrete, triggering the subsequent life adjustment mechanism. The fatigue life reduction is dynamically applied according to the excess amplitude of the load frequency intensity, and the original fatigue life prediction results are intelligently adjusted. High-frequency load features are extracted based on the identified load's dominant frequency. Specifically, the extracted features include the degree of energy tailing, caused by the slow decay of spectrum energy at the high-frequency tail. Within a set short-term sliding monitoring window, in-depth feature engineering analysis is performed on the rolling extracted high-frequency load feature data to generate a spectrum tail drag reference value. Based on this spectrum tail drag reference value, a load frequency intensity index is constructed to quantitatively characterize the high-frequency characteristics and intensity level of the current load. Within the set short-term sliding monitoring window, after performing deep feature engineering analysis on the energy tailing degree caused by the slow decay of spectrum energy at the high-frequency tail, a spectrum tail tail drag reference value is generated. The specific steps are as follows: In the set short-time sliding monitoring window, the load spectrum data is first segmented and the high-frequency tail frequency range away from the main frequency area is selected. , extract the amplitude spectrum corresponding to each frequency point in the high-frequency tail frequency range, and calculate the energy cumulative growth curve in ascending order of frequency based on the amplitude spectrum. The calculation expression is: ,in: Indicates the starting frequency from the high-frequency tail The accumulated energy between λ and frequency f; It is the spectrum amplitude at frequency point x, and the square represents the energy density; dx represents the frequency interval element; The spectrum tail drag reference value is calculated based on the energy accumulation curve. The specific calculation expression is: ,in: is the reference value of spectrum tail drag, is the growth rate of the energy accumulation curve at frequency f; It is a nonlinear enhancement exponential parameter, which is used to emphasize the slow energy growth in the high-frequency tailing area. is the starting frequency of the high-frequency tail; is the end frequency of the high-frequency tail.

2. The fatigue life prediction method of ultra-high performance concrete according to claim 1, characterized in that: The frequency response range of the high-frequency response sensor must cover at least twice the target load frequency to meet the requirements of the Nyquist sampling theorem.

3. The fatigue life prediction method of ultra-high performance concrete according to claim 1, characterized in that: The collected load time series signal is input into the fast Fourier transform algorithm to complete the conversion from time domain to frequency domain. The specific steps are as follows: Perform preliminary processing on the collected load time series signals to ensure that the signal quality meets the requirements of frequency domain analysis; Select the window function to perform windowing on the time series to reduce the spectrum leakage effect; The windowed time series signal is input into the fast Fourier transform algorithm, and the discrete Fourier transform operation is performed to obtain the frequency domain output in complex form; Calculate the amplitude spectrum of the complex output and extract the energy corresponding to each frequency component; The frequency axis is calibrated, the frequency resolution is calculated based on the sampling rate and data length, and a frequency-amplitude spectrum is generated.

4. The fatigue life prediction method of ultra-high performance concrete according to claim 1, characterized in that: In the obtained load spectrum, identify the main frequency peak with the highest energy, determine its corresponding frequency, and mark it as the load main frequency. The specific steps are as follows: Perform data extraction on the load spectrum obtained by fast Fourier transform, extract the frequency-amplitude pairs, and form a frequency component array; Apply the peak detection algorithm to traverse the frequency component array and identify the frequency point with the largest amplitude; The frequency point with the highest energy and meeting the validity screening conditions is extracted, its corresponding frequency value is recorded, and it is officially marked as the load main frequency.

5. The fatigue life prediction method of ultra-high performance concrete according to claim 1, characterized in that: Compare the current spectrum tail drag reference value with the micro-damage threshold to determine whether the load has an accelerating effect on the accumulation of concrete micro-damage. The specific steps are as follows: If the spectrum tail drag reference value is greater than the micro-damage threshold, it is determined that the load has an accelerating effect on the accumulation of micro-damage in concrete, triggering the subsequent life adjustment mechanism; If the spectrum tail drag reference value is less than or equal to the micro-damage threshold, it is determined that the load has no accelerating effect on the accumulation of micro-damage in concrete, and the original fatigue life prediction is maintained.

6. The fatigue life prediction method of ultra-high performance concrete according to claim 5, characterized in that: When it is determined that the load has an accelerating effect on the accumulation of micro-damage in concrete, fatigue life reduction is dynamically applied according to the excess amplitude of the load frequency intensity, and the original fatigue life prediction results are intelligently adjusted. The specific steps are as follows: After determining that the load has an accelerating effect on the accumulation of micro-damage in concrete, the excess amplitude of the spectrum tail drag reference value relative to the micro-damage threshold is calculated and subjected to nonlinear normalization to obtain the dynamic weighting factor. The specific calculation formula is as follows: ,in: is the dynamic weighting factor, is the spectrum tail drag reference value obtained by real-time monitoring of the current short-term sliding monitoring window, is the preset micro-damage frequency threshold, The upper limit of the reasonable maximum spectrum tail drag reference value set by the system, is the nonlinear enhancement factor; Based on the dynamic weighting factor obtained , adaptive compression is applied to the original fatigue life prediction value to generate an intelligently adjusted fatigue life prediction value. The specific compression formula is as follows: ,in: is the original fatigue life prediction value obtained based on the traditional method, is the fatigue life prediction value after adaptive compression adjustment, is the life reduction sensitivity coefficient, is the life reduction acceleration index, which is used to further control the response curve shape between the compression amplitude and the degree of frequency intensity excess.

Images