Railway fastener spring load-stress spectrum transmission and fatigue risk combined evaluation method

Abstract

The present application provides a railway fastener spring strip load-stress spectrum transmission and fatigue risk combined evaluation method. The method comprises: obtaining the state information of the railway fastener spring strip under multiple evaluation conditions; establishing a first analysis time window and a second analysis time window, constructing a regional evaluation position set according to the evaluation purpose, calculating the load-stress apparent spectrum transmission index and the input-output coherence index in the target frequency band based on the de-meaned target scalar load and the de-meaned fatigue evaluation stress parameter in the second analysis time window; based on the fatigue evaluation stress parameter time history in the first analysis time window, the material fatigue relationship is calculated and the damage accumulation is accumulated to obtain the position-level fatigue damage index, determine the fatigue risk ranking of each condition, and output the combined evaluation result of the railway fastener spring strip under each condition. The present application can realize the combined evaluation of the spectrum transmission characteristics and the fatigue risk of the railway fastener spring strip.

Description

Technical Field

[0001] This invention relates to the field of dynamic analysis and fatigue risk assessment of railway track structures, and in particular to a method for joint assessment of load-stress spectrum transfer and fatigue risk of railway fastener elastic bars. Background Technology

[0002] Railway fastener elastic clips serve functions such as rail restraint, gauge maintenance, and transmission of wheel-rail forces to the substructure. Influenced by installation pre-tensioning, curve geometry, local contact constraints, and the dynamic action of the wheel and rail, stress concentration and complex amplitude cycles easily occur at the root of the elastic clip and in the bending transition area. Its fatigue state has a significant impact on the service reliability of the fastener system.

[0003] Existing fatigue evaluation methods for fastener spring clips typically start from the local stress time history, calculating fatigue life or damage indices through cycle counting, mean stress correction, material fatigue relationships, and damage accumulation models. Other methods use power spectrum, frequency response, or transfer characteristics to describe the transmission of wheel-rail excitation to the local response of the fastener. While these existing methods can provide stress response, spectral characteristics, or fatigue damage, they generally lack a unified approach to the relationship between spectral transfer and cyclic damage.

[0004] Spectral analysis and fatigue analysis have different requirements for time history processing. Spectral transfer analysis typically requires removing static components and avoiding initial transients to extract stable dynamic responses; fatigue evaluation, on the other hand, requires retaining the complete stress time history and its mean information to obtain the cyclic mean after cycle counting and to reflect the average stress effects caused by preload and assembly conditions. Mean removal processing of stress parameters in fatigue evaluation alters the absolute level of the cyclic mean, thus affecting the equivalent cyclic amplitude after mean stress correction. Furthermore, directly using the complete time history, including initial contact adjustment or loading transients, for spectral analysis may subject the target frequency band spectral energy and transfer parameters to transient broadband components. Therefore, using the same time window and time history preprocessing method for both types of evaluation makes it difficult to simultaneously meet the processing requirements of spectral transfer analysis and cyclic fatigue evaluation.

[0005] Different operating conditions may activate different local high-response regions. For evaluation tasks that identify high-response regions for each operating condition, it is necessary to allow the region location to change with the operating condition; for evaluation tasks that compare the same spatial region with changes in operating parameters, it is necessary to maintain consistency of the evaluation region across different operating conditions. Using a single representative location or a region selection method that is not adjusted according to the evaluation purpose can easily make the results susceptible to the influence of local mesh, individual response locations, or spatial region drift. At the same time, high stress spectral energy or high apparent transfer level does not necessarily mean that it is stably and coherently transferred by the input load, nor does it necessarily correspond to the highest fatigue damage; judging the dangerous operating condition solely based on modal approximation, spectral peaks, or single stress energy may lead to misjudgment.

[0006] Field wheel-rail loads are random. Determining fatigue risk ranking based solely on a single load segment makes it difficult to assess whether the resulting ranking remains stable across load sample variations. Therefore, a universal evaluation method is needed that can consider factors such as physical division of time windows, adaptive region selection for evaluation purposes, load-stress spectrum transfer, input-output coherence, cyclic damage accumulation, and ranking stability within a given load sample set. Summary of the Invention

[0007] The embodiments of the present invention provide a method for jointly assessing the load-stress spectrum transfer and fatigue risk of railway fastener elastic clips, so as to achieve a joint assessment of the spectrum transfer characteristics and fatigue risk of railway fastener elastic clips.

[0008] To achieve the above objectives, the present invention adopts the following technical solution.

[0009] A method for jointly assessing the load-stress spectrum transfer and fatigue risk of railway fastener elastic bars includes:

[0010] Obtain the status information of railway fastener elastic strips under multiple working conditions to be evaluated;

[0011] A first and second analysis time window are established to describe the effective load process of railway fastener elastic strips. The fatigue evaluation stress parameters in the first analysis time window retain the cyclic mean and are used for fatigue evaluation. The target scalar load and fatigue evaluation stress parameters in the second analysis time window are respectively processed to remove the mean and used for spectral transfer and coherence evaluation.

[0012] Construct a set of regional evaluation locations based on the evaluation purpose; construct an independent set of high-response locations for each working condition when identifying its own high-response areas; and construct a set of common high-response locations across working conditions when comparing the response differences of the same spatial region with changes in working condition parameters.

[0013] Based on the state information of the railway fastener elastic bar, the load-stress apparent spectrum transfer index and input-output coherence index in the target frequency band are calculated using the mean-reduced target scalar load and mean-reduced fatigue evaluation stress parameters within the second analysis time window.

[0014] Based on the fatigue evaluation stress parameter time history within the first analysis time window, cyclic counting, mean stress correction, material fatigue relationship calculation, and damage accumulation are performed to obtain location-level fatigue damage indicators. The location-level fatigue damage indicators that meet the fatigue evaluation validity conditions in the evaluation location set of the region are statistically analyzed, and the fatigue risk ranking of each working condition is determined based on the effective location-level fatigue damage statistics.

[0015] The fatigue risk ranking of each working condition of the railway fastener elastic bar is compared with the ranking of the apparent spectral transfer index to determine the degree of mismatch between the spectral transfer and fatigue risk ranking of each working condition. The joint evaluation result of each working condition of the railway fastener elastic bar is output in combination with the coherence index.

[0016] Preferably, the acquisition of status information of railway fastener elastic strips under multiple working conditions to be evaluated includes:

[0017] The system acquires the status information of railway fastener elastic bars under multiple evaluation conditions. This status information includes the target input load time history for each condition, the fatigue evaluation stress parameter time history for multiple evaluation locations in the target area of ​​the fastener elastic bar, and the condition parameters. It identifies the time series, load components, and evaluation location identifiers, and completes finite value checks, time sorting, duplicate time point processing, and time alignment within a common effective time range.

[0018] Preferably, the first and second analysis time windows for establishing the effective load action process of the railway fastener elastic strip are as follows: the fatigue evaluation stress parameters within the first analysis time window retain the cyclic mean and are used for fatigue evaluation; the target scalar load and fatigue evaluation stress parameters within the second analysis time window are respectively subjected to mean-reduction processing and used for spectral transfer and coherence evaluation, including:

[0019] A first analysis time window covering the effective load action process is established, and a second analysis time window is set within the first analysis time window. When performing time window sensitivity verification, a reference candidate time window is selected from multiple candidate time windows located within the first analysis time window. A pre-selected evaluation location set for sensitivity verification is constructed based on the dynamic response index of each evaluation location within the reference candidate time window. Under the condition of keeping the pre-selected evaluation location set and the spectrum estimation parameters consistent, the regional mid-dynamic response index and the preset frequency band stress spectrum energy corresponding to each candidate time window are calculated based on the state information of the railway fastener elastic bar under the evaluation condition. The second analysis time window is determined based on the degree of change of the index relative to the reference candidate time window.

[0020] The pre-selected evaluation location set is used for time window sensitivity verification. The fatigue evaluation stress parameters in the first analysis time window retain the cyclic mean and are used for fatigue evaluation. The target scalar load and fatigue evaluation stress parameters in the second analysis time window are respectively processed to remove the mean and used for spectral transfer and coherence evaluation. The same second analysis time window is used for each working condition in the same comparison group.

[0021] Preferably, the construction of a regional evaluation location set according to the evaluation purpose, the construction of an independent high-response location set for each working condition when identifying its own high-response area, and the construction of a cross-working-condition common high-response location set when comparing the response differences of the same spatial region with changes in working condition parameters, includes:

[0022] When the evaluation objective is to identify the high-response region for each working condition, the independent high-response location set for each working condition consists of the top N evaluation locations ranked by the dynamic stress response index within that working condition. When the evaluation objective is to compare the response differences of the same spatial region with variations in working condition parameters, the common high-response location set is selected from the valid evaluation locations shared by all working conditions to be compared, and the top N locations are selected based on the cross-working-condition statistics of the normalized dynamic stress response index for each working condition. Evaluation position; The determination is based on the stability of the regional response statistics under the number of candidate locations, the stability of the evaluation location set under repeated samples or perturbation analysis, and the spatial coverage requirements of the target area.

[0023] Preferably, the calculation of the load-stress apparent spectrum transfer index and input-output coherence index within the target frequency band based on the state information of the railway fastener elastic bar, using the mean-reduced target scalar load and mean-reduced fatigue evaluation stress parameters within the second analysis time window, includes:

[0024] For a predetermined target scalar load channel, the power spectrum of the target scalar load and the power spectrum of the fatigue evaluation stress parameters within the second analysis time window are calculated using consistent spectral estimation parameters based on the state information of the railway fastener elastic bar under the working condition to be evaluated. The cross power spectrum between the target scalar load and the fatigue evaluation stress parameters is also calculated. A target frequency band is constructed around the target frequency, and the position-level load-stress apparent spectrum transfer index and input-output coherence index within the target frequency band are calculated.

[0025] The target frequency is any one of the following: a preset excitation frequency, a fastener spring characteristic frequency, or a load spectrum characteristic frequency of the target scalar load channel; or it can be determined based on the operating speed v and the spatial excitation scale. according to Confirmed; the target frequency band is ;No. The evaluation position is relative to the first Apparent spectral transfer index of each target scalar load channel The energy of the power spectrum of the fatigue evaluation stress parameter within the target frequency band and the first The square root of the ratio of the power spectrum energies of the target scalar load channels is determined.

[0026] Preferably, the process involves performing cyclic counting, mean stress correction, material fatigue relationship calculation, and damage accumulation based on the fatigue evaluation stress parameter time history within the first analysis time window to obtain location-level fatigue damage indices. This includes statistically analyzing the location-level fatigue damage indices that meet the fatigue evaluation validity conditions for the concentrated evaluation locations in the region, and determining the fatigue risk ranking of each working condition of the railway fastener elastic bar based on the effective location-level fatigue damage statistics.

[0027] Based on the fatigue evaluation stress parameter time history within the first analysis time window, cycle counting, mean stress correction, material fatigue relationship calculation, and damage accumulation are performed. Evaluation locations within the regional evaluation location set that have a Goodman mean stress correction denominator not greater than zero are marked, and the Goodman corrected damage results of these evaluation locations are not included in the regional fatigue damage quantitative statistics. The remaining evaluation locations with effective fatigue calculation results are combined into a fatigue evaluation effective location set. The location-level fatigue damage indicators within the fatigue evaluation effective location set are statistically analyzed, and the fatigue risk ranking for each working condition is determined by the regional median damage rate or the regional damage statistics that monotonically correspond to it.

[0028] The cycle count uses rainflow counting; the mean stress correction uses Goodman correction; the material fatigue relationship uses stress-life relationship; the damage accumulation uses Miner linear damage accumulation criterion; the fatigue damage index includes at least the damage rate per unit time, and may further include at least one of cumulative damage, equivalent cycle index and reference cycle requirement pass rate.

[0029] Preferably, the step of comparing the fatigue risk ranking of each working condition of the railway fastener elastic bar with the ranking of the apparent spectral transfer index to determine the degree of mismatch between the spectral transfer and fatigue risk ranking of each working condition, and outputting the joint evaluation result of each working condition of the railway fastener elastic bar in combination with the coherence index, includes:

[0030] For K ≥ 2 working conditions to be evaluated, based on the common effective fatigue evaluation location set formed by the intersection of the effective fatigue evaluation location sets of each working condition, the regional apparent spectrum transfer index and the regional fatigue damage index are statistically analyzed to determine the ranking of the apparent spectrum transfer index. And fatigue risk ranking and in accordance with | - | Determine the first The normalized ranking mismatch degree of each working condition to be evaluated is determined; the apparent spectral transfer index ranking, fatigue risk ranking, input spectrum weighted coherence, and normalized ranking mismatch degree of each working condition to be evaluated are output to characterize the consistency or mismatch degree between spectral transfer and fatigue risk, and a joint evaluation result of each working condition to be evaluated is formed. The joint evaluation result includes spectral transfer ranking, fatigue risk ranking, ranking mismatch degree, and related auxiliary interpretation indicators.

[0031] Preferably, the method further includes: for no less than three independent load samples, normalizing and ranking each working condition according to the regional fatigue damage index within each load sample; and statistically analyzing the frequency of each working condition becoming the highest-risk working condition. and interquartile range of cross-sample ranking The stability of fatigue risk ranking within a given load sample set is characterized by the frequency of occurrence of the highest-risk working condition and the interquartile range of the cross-sample ranking. When multiple working conditions have the same fatigue damage index in the same area, they are ranked in parallel. When a load sample is missing any working condition to be compared, the load sample is not included in the ranking stability statistics based on the complete set of working conditions. Alternatively, a subset of working conditions commonly included in each load sample is predetermined, and all valid load samples are normalized and ranked according to the same subset of working conditions.

[0032] A joint assessment system for load-stress spectrum transfer and fatigue risk of railway fastener elastic bars includes a processor and a memory. The memory stores a computer program, which, when executed by the processor, causes the system to perform the following operations: acquire and align the time history of the target input load and the time history of fatigue evaluation stress parameters at multiple evaluation locations in the target area; construct a first analysis time window and a second analysis time window; construct an independent high-response location set or a common high-response location set across working conditions according to the evaluation purpose; calculate the apparent spectrum transfer index and input-output coherence index of the target frequency band; calculate the fatigue damage index and cyclic damage interval contribution; verify the effectiveness of fatigue calculation and construct an effective fatigue evaluation location set; determine the fatigue risk ranking and the degree of mismatch between spectrum transfer and fatigue risk ranking; when there are no less than three independent load samples, complete the normalization and ranking of working conditions within each load sample, summarize and statistically analyze the frequency of occurrence of each working condition becoming the highest-risk working condition and the interquartile range of the cross-sample ranking, and output the fatigue risk ranking stability result within the given load sample set.

[0033] A computer-readable storage medium storing a computer program, which, when executed by a processor, implements the method for joint assessment of load-stress spectrum transfer and fatigue risk of railway fastener elastic clips.

[0034] As can be seen from the technical solutions provided by the embodiments of the present invention described above, the method of the present invention can identify situations where the spectral transfer peak value is inconsistent with the highest fatigue risk condition by jointly ranking and calculating the degree of mismatch. The present invention can simultaneously output fatigue risk ranking and the spectral transfer, coherence, and cyclic damage information that forms this ranking. The present invention can solve the problems in existing fatigue evaluation of fastener spring clips, such as the mixed use of spectral analysis and fatigue analysis time history processing, mismatch between evaluation area selection and evaluation purpose, unclear relationships between multi-component load inputs, difficulty in directly corresponding fatigue risk to spectral transfer peak values, and insufficient representativeness of single-sample risk ranking.

[0035] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and will become apparent from the description or may be learned by practice of the invention. Attached Figure Description

[0036] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 The above is a flowchart of a method for jointly assessing the load-stress spectrum transfer and fatigue risk of railway fastener elastic bars, provided in an embodiment of the present invention.

[0038] Figure 2 This invention provides a flowchart for constructing an adaptive region evaluation location set for dual-time-window processing and evaluation purposes.

[0039] Figure 3 A flowchart for load-stress spectrum transfer and coherence calculation is provided for an embodiment of the present invention.

[0040] Figure 4 A flowchart for calculating cyclic fatigue index and damage contribution is provided for an embodiment of the present invention.

[0041] Figure 5 This is a flowchart of a joint sorting, mismatch identification, and stability evaluation within a given load sample set provided in an embodiment of the present invention.

[0042] Figure 6 This is a schematic diagram of the module composition of a fatigue risk joint assessment system provided in an embodiment of the present invention. Detailed Implementation

[0043] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0044] Those skilled in the art will understand that, unless specifically stated otherwise, the singular forms “a,” “an,” “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, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, 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 connections or couplings. The term “and / or” as used herein includes any and all combinations of one or more of the associated listed items.

[0045] 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 defined as herein.

[0046] To facilitate understanding of the embodiments of the present invention, the following will provide further explanation and description with reference to the accompanying drawings and several specific embodiments. These embodiments do not constitute a limitation on the embodiments of the present invention.

[0047] The present invention will be further described below with reference to the accompanying drawings and embodiments. The fastener types, load sources, operating parameters, time windows, number of evaluation locations, target frequency bands, fatigue parameters, and output formats in the embodiments are only used to illustrate possible implementations of the present invention and do not constitute a limitation on the scope of protection.

[0048] The processing flow of a joint assessment method for load-stress spectrum transfer and fatigue risk of railway fastener elastic bars provided in this embodiment of the invention is as follows: Figure 1 As shown, it includes the following steps:

[0049] Step S1: Obtain and verify the input data.

[0050] The system acquires the status information of railway fastener elastic clips under multiple evaluation conditions. This status information includes the target input load time history of the condition, the fatigue evaluation stress parameter time history of multiple evaluation locations in the target area of ​​the fastener elastic clip, and the condition parameters. It identifies time series, load components, and evaluation location identifiers, and performs finite value checks, time sorting, duplicate time point processing, and time alignment within a common valid time range. When multiple load components exist, target scalar load channels are determined separately. When the sampling frequencies of load and stress data are inconsistent, resampling is performed after anti-aliasing filtering, avoiding extrapolation outside the original data's valid time range. When there is a time delay in acquisition or transmission, delay compensation is performed based on synchronization information or cross-correlation analysis.

[0051] The aforementioned time alignment is completed within the common effective time range of the target input load and the fatigue evaluation stress parameters. When the sampling frequencies of the two are inconsistent, anti-aliasing filtering is performed before resampling, and extrapolation is not performed outside the effective time range of the original data. When there is a time delay in the acquisition or transmission of the target input load and the fatigue evaluation stress parameters, time delay compensation is performed based on synchronization trigger information or cross-correlation analysis. The amplitude squared coherence is determined based on the cross power spectrum between the target scalar load and the fatigue evaluation stress parameters, the power spectrum of the target scalar load, and the power spectrum of the fatigue evaluation stress parameters. Weighted coherence is calculated within the target frequency band using the power spectrum of the target scalar load as the weight. Before calculation, it is verified that the upper limit of the target frequency band does not exceed the Nyquist frequency, the number of independent frequency components within the target frequency band meets the preset requirements, the energy of the input load spectrum is greater than the preset lower limit, and the spectrum data is a finite value.

[0052] Step S2: Establish the first and second analysis time windows for the effective load action process of the railway fastener elastic strip.

[0053] A first analysis time window covering the effective load action process is established, and a second analysis time window is set within the first analysis time window. The second analysis time window can be predetermined based on the initial transient decay stage and the duration of the main dynamic response, or it can be determined through time window sensitivity verification. When performing time window sensitivity verification, a reference candidate time window is first selected from multiple candidate time windows located within the first analysis time window, and a pre-selected evaluation location set for sensitivity verification is constructed based on the dynamic response indices of each evaluation location within the reference candidate time window. Subsequently, while maintaining consistency between the pre-selected evaluation location set and the spectral estimation parameters, the regional median dynamic response index and the preset frequency band stress spectrum energy corresponding to each candidate time window are calculated respectively, and the final second analysis time window is determined based on the degree of change of the indices relative to the reference candidate time window. The pre-selected evaluation location set is only used for time window sensitivity verification. After the final second analysis time window is determined, the formal regional evaluation location set is determined according to the subsequent regional evaluation location set construction steps. The fatigue evaluation stress parameters within the first analysis time window are not subject to mean removal to retain cyclic mean information for cyclic fatigue evaluation. Within the second analysis time window, mean removal is performed on the target scalar load and the fatigue evaluation stress parameters at each evaluation location, and this is used for dynamic response, spectral transfer, and coherence evaluation. The same second analysis time window is used for all operating conditions within the same comparison group.

[0054] Step S3: Construct an adaptive set of regional evaluation locations for the evaluation purpose.

[0055] When the evaluation objective is to identify high-response areas for each operating condition, a set of independent high-response locations for each operating condition is constructed based on the dynamic response indicators of each condition. When the evaluation objective is to compare the response differences of the same spatial region with changes in operating condition parameters, a set of common high-response locations across operating conditions is constructed at the valid evaluation locations that coexist in all operating conditions. The number of high-response locations, N, is determined based on the stability of the regional response statistics under the number of candidate locations, the stability of the evaluation location set under repeated samples or perturbation analysis, and the spatial coverage requirements of the target area.

[0056] Step S4: Calculate the load-stress spectrum transfer and coherence.

[0057] For a predetermined target scalar load channel, consistent spectral estimation parameters are used to calculate the target scalar load power spectrum, fatigue evaluation stress parameter power spectrum, and cross-power spectrum between the two within the second analysis time window. A target frequency band is constructed around the target frequency, and the position-level apparent spectral transfer index and input spectrum weighted coherence are calculated. The overall root mean square (RMS) type transfer index is calculated according to the evaluation requirements. At the same time, time synchronization, frequency range, number of independent frequency components, input spectrum energy, and finite value verification are performed.

[0058] Step S5: Calculate the cyclic fatigue index.

[0059] Based on the state information of the railway fastener elastic bar under the above-mentioned working conditions, rainflow counting is performed on the fatigue evaluation stress parameter time history within the first analysis time window. Goodman mean stress correction is performed based on the cycle amplitude, cycle mean, and cycle number. The cumulative damage, damage rate, equivalent cycle index, and reference cycle requirement pass rate are obtained by combining the material S-N relationship and Miner linear damage accumulation criterion. At the same time, the proportion of high-amplitude cycles and the damage contribution of each cycle amplitude interval are calculated.

[0060] Step S6: Conduct regional statistics and determine the fatigue risk ranking.

[0061] Regional statistics were performed on the location-level apparent spectral transfer index, input spectral weighted coherence, and stress spectral energy within the regional evaluation location set. Evaluation locations exhibiting Goodman mean stress correction denominators not greater than zero cycles were marked, and their Goodman-corrected damage results were excluded from the regional fatigue damage quantitative statistics. The remaining evaluation locations with valid fatigue calculation results were combined into a fatigue evaluation valid location set ε. v Based on the effective location set of fatigue evaluation, fatigue damage indicators are statistically analyzed, and the fatigue risk ranking of each working condition is determined by the regional median damage rate or the regional damage statistics that monotonically correspond to it. For the mismatch evaluation between the spectral transfer ranking and the fatigue risk ranking, both the apparent spectral transfer indicators and the fatigue damage indicators are based on the common effective location set of fatigue evaluation for each working condition to be evaluated; the spectral indicators obtained based on the complete regional evaluation location set are output as auxiliary results of the overall regional response.

[0062] Step S7: Identify the consistency or mismatch between spectral transfer and fatigue risk.

[0063] The apparent spectral transfer index ranking is compared with the fatigue risk ranking. The normalized ranking mismatch is calculated, and combined with the input spectrum weighted coherence and cyclic damage redistribution results, the spectral transfer ranking, fatigue risk ranking, ranking mismatch, and related auxiliary interpretation indicators for each working condition to be evaluated are output.

[0064] Optionally, in step S8, the stability of cross-sample ranking is evaluated. When there are at least three independent load samples, the regional fatigue damage index of each working condition within each load sample is normalized and ranked. The frequency of each working condition becoming the highest risk working condition and the interquartile range of the cross-sample ranking are statistically analyzed, and the relative fatigue risk ranking stability within a given load sample set is output. The stability index is not interpreted as a population probability or statistical confidence interval.

[0065] This invention also provides a joint assessment system for load-stress spectrum transfer and fatigue risk of railway fastener elastic bars. The system includes a processor and a memory. The memory stores a computer program. When the computer program is executed by the processor, it performs data acquisition and quality verification, dual time window processing, construction of regional evaluation location sets, spectrum transfer and coherence calculation, cyclic fatigue, effectiveness verification and damage decomposition, regional statistics and joint ranking, cross-sample stability evaluation when there are no less than three independent load samples, and standardized result output.

[0066] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described method for joint assessment of load-stress spectrum transfer and fatigue risk of railway fastener elastic clips.

[0067] Example 1: Input Data, Quality Verification, and Dual Time Window Processing

[0068] The input data includes the time histories of target input loads for multiple evaluation conditions, the time histories of fatigue evaluation stress parameters at multiple evaluation locations within the target area of ​​the fastener spring clip, and the condition identifier. The input load can be vertical force, lateral force, longitudinal force, or other load components with clear physical meaning. When multiple load components exist, each load component is used as a target scalar load channel for spectral transfer evaluation, or a target scalar load channel is formed according to a preset and physically based mapping relationship. The fatigue evaluation stress parameters can be signed principal stresses, equivalent stresses determined according to preset sign rules, hot spot stresses, critical surface fatigue parameters, or scalar stress parameters calibrated experimentally and matching the fatigue relationship of the material used. This embodiment uses the wheel-rail vertical force channel and the equivalent stress of finite element elements to illustrate the method, and interprets the obtained fatigue results as a comparative risk indicator.

[0069] The input data undergoes quality checks, including checks for time and response value finiteness, ascending time sorting, handling of repeated time points, length checks of load and stress data, and checks of the common effective time range. When load and stress time points are inconsistent, they are resampled to a unified time grid within the common effective time range. When the sampling frequency is reduced, anti-aliasing filtering is performed first. Extrapolation is not performed outside the valid range of the original data. If there are acquisition trigger errors or transmission time delays, time delay compensation is performed based on the synchronization trigger signal or cross-correlation peak value.

[0070] The first analysis time window covers the complete process of the input load generating an effective response to the target fastener, and is used for cycle counting and fatigue damage calculation. The second analysis time window is located within the first analysis time window and is used for dynamic response, spectral transfer, and coherence evaluation. The start and end positions of the second analysis time window can be preset according to the initial transient end time and the duration of the main dynamic response, or it can be determined by sensitivity verification of multiple candidate time windows. When performing the time window sensitivity verification, a reference candidate time window is selected from multiple candidate time windows, and a pre-selected evaluation position set is constructed based on the dynamic response index of each evaluation position within the time window. Under the condition that the pre-selected evaluation position set, window function, segment length, overlap rate, fast Fourier transform length, and target frequency band remain unchanged, the regional median dynamic response index and the regional median preset frequency band stress spectrum energy corresponding to each candidate time window are calculated respectively, and the stability of the time window is judged based on the relative changes of the two with respect to the reference candidate time window, so as to avoid the simultaneous influence of time window changes and evaluation area changes on the verification results. When the relative changes in the regional dynamic response index and the regional preset frequency band stress spectrum energy both do not exceed the corresponding preset thresholds, the corresponding candidate time window is considered to meet the stability requirements. Combining the initial transient exclusion degree and the degree of retention of the main dynamic response, the final second analysis time window is determined. Subsequently, the regional evaluation location set is reconstructed based on the dynamic response index of each evaluation location within the final second analysis time window. When the final second analysis time window is consistent with the reference candidate time window, the pre-selected evaluation location set can be directly used. Within the second analysis time window, the time average of the target scalar load and the fatigue evaluation stress parameters of each evaluation location is calculated and removed. The relative changes in the regional dynamic response index and the relative changes in the regional frequency band stress spectrum energy of the (q)th candidate time window relative to the reference candidate time window are calculated according to the following formula:

[0071]

[0072]

[0073] In the formula, q is the candidate time window number; the superscript (0) is the reference candidate time window; This is the regional median dynamic response indicator under the q-th candidate time window; The stress spectrum energy in the mid-frequency band of the region under the q-th candidate time window; and These represent the relative changes of the corresponding indicators; and All should be greater than the corresponding preset effective lower limit.

[0074] when At that time, it was believed that the first Each candidate time window meets the stability requirements. The threshold for relative change in regional dynamic response; This is the threshold for the relative change in energy within the regional frequency band. Among the candidate time windows that meet the stability requirements, the time window that can exclude the initial transient and fully preserve the main dynamic response is selected; when multiple candidate time windows meet the above requirements, the candidate time window with the shorter duration is preferentially selected as the final second analysis time window.

[0075] The relative change thresholds for regional dynamic response and regional frequency band energy can be predetermined based on the repeatability of the evaluation data, the accuracy of numerical calculation, and the allowable degree of variation in the evaluation results, and should remain consistent in the same evaluation task.

[0076] This invention is applicable to fatigue risk screening for different types of fastener springs, different external excitation conditions, and different assembly or structural states. The input load can come from field measurements, experimental loading, vehicle-track dynamics calculations, finite element simulations, or combinations thereof. When the input load contains multiple directions or multiple physical components, each load component is evaluated as a target scalar load channel, or a target scalar load channel is formed according to a preset physical mapping. The evaluation location can be a finite element element, integration point, node, strain measurement point, or a discrete response location determined by preset rules.

[0077] Example 2: Evaluation Objective Adaptive Regional Evaluation Location Set

[0078] Figure 2 This is a flowchart illustrating the construction of an adaptive region evaluation location set for dual-time-window processing and evaluation purposes, as provided in an embodiment of the present invention. For the first... At each evaluation location, the dynamic stress response index is calculated within the second analysis time window. This embodiment uses the root mean square of dynamic stress, calculated as follows: [Formula omitted for brevity] This refers to the time interval corresponding to the second analysis time window. The duration of the second analysis time window. For the first Mean-free fatigue evaluation stress parameters at each evaluation location It is a time variable.

[0079]

[0080] When the evaluation objective is to identify the high-response region for each operating condition, each operating condition is evaluated separately according to... Sort in descending order and select the first few. Each evaluation location forms an independent set of high-response locations for that operating condition. When the evaluation objective is to compare the response differences of the same spatial region with variations in operating condition parameters, first determine the effective evaluation locations that coexist across all operating conditions; within each operating condition, evaluate R... eNormalize, then calculate the average, median or other preset statistics for each location across operating conditions, and select the top N evaluation locations to form a common high-response location set.

[0081] Determined through sensitivity testing of multiple candidate values. For each candidate... The value is used to calculate the deviation of the regional median dynamic response index relative to the reference N value. When there are repeated load samples, grid disturbances, or resampling results, the overlap rate and Jaccard similarity coefficient of the high-response location set under the same evaluation purpose are further calculated. For the operating condition-independent high-response location set, the stability of the set is evaluated between different samples or disturbance results of the same operating condition, without using the degree of location overlap between different operating conditions as a criterion. When the regional dynamic response statistics and location set stability meet the preset requirements, N is determined in combination with the spatial coverage requirements of the target area; when there are no additional spatial coverage requirements, a smaller value is preferred. The number of public valid locations is less than When this happens, all public valid locations are used and an insufficient quantity indicator is output.

[0082]

[0083]

[0084] In the formula, , They respectively represent the first evaluation goal and the location set construction rule, under the same evaluation purpose and location set construction rule. , The location set consists of the top N high-response locations corresponding to each repeated load sample, grid perturbation, or resampling result; The overlap rate; The Jaccard similarity coefficient; For reference position quantity; For regional dynamic response statistics relative to the reference Deviation of values; , , This represents the corresponding allowable threshold.

[0085] The threshold values ​​for regional overlap rate, Jaccard similarity coefficient, and relative deviation of regional response statistics can be predetermined based on the repeatability of the evaluation data, the accuracy of numerical calculation, and the allowable variation in the evaluation results, and should remain consistent throughout the same evaluation task. These threshold values ​​are used to verify the stability of the number of evaluation locations N, and are not used as indicators for ranking fatigue risk.

[0086] Example 3: Load-Stress Spectrum Transfer and Coherence

[0087] Figure 3A flowchart for load-stress spectrum transfer and coherence calculation is provided for an embodiment of the present invention. Within the second analysis time window, using... Indicates the first The demeaned load of each target scalar load channel, to Indicates the first The mean-free fatigue evaluation stress parameters are calculated for each evaluation location. The apparent spectral transfer index and coherence index are calculated separately for multiple load components; loads of different dimensions or directions are not directly and undefinedly merged into a single input spectrum. The target scalar load power spectrum is calculated using the same window function, piece length, overlap rate, Fast Fourier Transform (FFT) length, and sampling frequency. Power spectrum of stress parameters for fatigue evaluation and cross-power spectrum The target frequency can be determined by a preset excitation frequency, the fastener characteristic frequency, or the input spectrum characteristic frequency; for an excitation with a spatial scale λ and passing through at a velocity v, the target frequency can be determined by the following formula. λ represents the operating speed, in m / s; λ represents the spatial excitation scale. The target frequency is in Hz.

[0088]

[0089] Let the target frequency band be Then the first The apparent spectral transfer index for each evaluation location can be determined by the following formula. Where, For frequency variables, For the target frequency band half-width, For the first The input channel to the first Apparent spectral transfer index at each location.

[0090]

[0091] Indicates the first The target scalar load channel to the first The apparent spectral transfer index of an evaluation location within the target frequency band is used for comparisons between different operating conditions and is not equivalent to the complete frequency response function of a linear time-invariant system. The amplitude squared coherence and its input spectrum weighted average frequency band value can be determined by the following formula:

[0092]

[0093]

[0094] Since the amplitude squared coherence ranges from [0,1] and the target scalar load power spectrum has non-negative weights, the input spectrum weighted coherence ranges from [0,1] even when the input spectrum energy in the target frequency band is greater than zero. The input spectrum weighted coherence is used to reduce the influence of weaker frequency points in the input spectrum on the results. It can also be used to calculate the overall RMS transfer index. This describes the overall dynamic transfer level within the second analysis time window.

[0095]

[0096]

[0097] Before spectral calculation, the upper limit of the target frequency band must not exceed the Nyquist frequency, the number of independent frequency components within the target frequency band must meet preset requirements, the input spectral energy must be greater than the preset lower limit, and the frequency vectors of the power spectrum and cross-power spectrum must be consistent. When using piecewise average spectrum estimation, the number of effective average segments or equivalent degrees of freedom can be further checked to see if they meet preset requirements; interpolated frequency points obtained by zero-padding are not included in the number of independent frequency components. If the check fails, the target frequency band or spectrum estimation parameters are adjusted, or an invalid flag is output. When the input load or response data has been resampled, the spectrum calculation uses the unified sampling frequency after resampling; when time delay compensation is performed between input and output, coherence and cross-power spectrum are calculated based on the compensated synchronous data. Each evaluation position is in the preset frequency band. The internal stress spectrum energy is calculated according to the following formula:

[0098]

[0099] In the formula, , These are the lower and upper limits of the preset frequency band, respectively; For the first The stress spectrum energy of each evaluation position within the preset frequency band.

[0100] The lower limit of the input spectral energy can be predetermined based on the background noise level of the load measurement or the effective accuracy of numerical calculation. This is to prevent the input spectral energy within the target frequency band from being too low, which would cause the apparent spectral transfer index and the weighted coherence of the input spectrum to lose numerical stability. The preset requirements for the number of effective average segments, equivalent degrees of freedom, and the number of independent frequency components can be determined based on the spectral estimation method, data duration, frequency resolution, and the stability of the evaluation results, and should be kept consistent within the same evaluation task.

[0101] Example 4: Cyclic Fatigue Indices and Cyclic Damage Decomposition

[0102] Figure 4This invention provides a flowchart for calculating cyclic fatigue indices and damage contribution. Rainflow counting is performed on the fatigue evaluation stress parameter time history at the e-th evaluation location within the first analysis time window to obtain the cyclic amplitude of the i-th cycle group. Circular mean and number of cycles This embodiment uses Goodman mean stress correction, material S-N relationship, and Miner linear damage accumulation criterion for calculation. The specific formula is as follows:

[0103]

[0104]

[0105]

[0106] In the formula, For the first Number of loop iterations; For cumulative damage; Damage rate per unit time; The duration of the first analysis time window; The equivalent number of loops when the current load segment is applied repeatedly. , For material SN parameters; The allowed number of cycles corresponding to the equivalent cycle amplitude.

[0107] in, The ultimate strength of the material, To correct the allowable number of iterations corresponding to the equivalent loop amplitude, The duration of the first analysis time window, This represents the total number of rainflow count cycles. When the Goodman mean stress correction denominator is greater than zero and the corrected equivalent cycle amplitude is a finite positive value, the allowable number of cycles and fatigue damage for the corresponding cycle are calculated; when the correction denominator is not greater than zero, a status flag indicating that the mean stress correction exceeds its applicable range is output, and the cycle is not treated as zero damage. If there are no valid cycles, the corresponding status flag is output; cumulative damage... When the value is zero, the zero-damage result is retained; in both cases above, the equivalent cyclic index is not calculated. When the equivalent stress, which is always non-negative, is used as the fatigue evaluation parameter, the resulting mean stress correction result is mainly used for comparative risk ranking under the same stress characterization method and the same material parameters, and is not directly used as the absolute service life prediction result.

[0108] When a cycle exists at an evaluation location where the denominator of the mean stress correction is not greater than zero, that evaluation location is marked as having a mean stress correction outside its applicable range. The Goodman-corrected damage result for that location is not included in the quantitative statistics of regional fatigue damage, and the number of evaluation locations in this state and their proportion of the regional evaluation location set are counted separately. This state indicates that the current stress conditions exceed the applicable range of the adopted Goodman mean stress correction relationship and is not equivalent to zero fatigue damage. Evaluation locations that are not marked as having a mean stress correction outside its applicable range and have valid fatigue calculation results are grouped into a valid fatigue evaluation location set. The reference cycle requirement pass rate is calculated solely based on the effective set of fatigue evaluation locations.

[0109] Regardless of whether the proportion of valid fatigue assessment locations is lower than the preset threshold, if there are assessment locations where the mean stress correction exceeds the applicable range, their spatial location, quantity, and proportion should be output separately, along with a corresponding risk warning. Damage ranking within the set of valid fatigue assessment locations is not used to prove that the fatigue risk of working conditions containing the aforementioned locations exceeding the applicable range is low. When the set of valid fatigue assessment locations... When the proportion of valid evaluation locations in the regional evaluation location set is lower than a preset threshold, a warning message "Insufficient validity of regional fatigue evaluation" is output. The preset threshold can be determined based on the sensitivity of evaluation location reduction, the stability of regional damage statistics, or the results of resampling analysis, and should remain consistent within the same evaluation task. This threshold is only used to trigger the regional fatigue evaluation validity warning and is not used as a threshold for fatigue risk classification or ranking.

[0110]

[0111] In the formula, For the effective location set of fatigue evaluation, To effectively evaluate the number of locations; For reference, the loop requirement is as follows; is the regional pass rate; 1(·) is the indicator function.

[0112] To describe the redistribution of cycles towards higher amplitudes, a high amplitude threshold is set. And calculate the high-amplitude cycle ratio. The loop is divided into multiple intervals b according to the original loop amplitude, but the damage of each loop is calculated using the corrected equivalent loop amplitude. The damage contribution of each interval is... Determine using the following formula:

[0113]

[0114]

[0115] The high-amplitude threshold and cyclic amplitude interval can be set according to the material fatigue characteristics, sample distribution quantiles, or evaluation objectives. Fatigue risk ranking is primarily determined based on the regional median damage rate or the regional damage statistic that monotonically corresponds to it; equivalent cycles, pass rate, high-amplitude cycle proportion, and interval damage contribution are used to assist in verification and interpretation. The same high-amplitude threshold and cyclic amplitude interval division rules are used for all working conditions within the same comparison group.

[0116] Example 5: Joint Ranking and Spectral Transfer—Fatigue Risk Mismatch Identification

[0117] In the same evaluation task, for each working condition to be evaluated, the apparent spectral transfer index, input spectral weighted coherence, and preset frequency band stress spectral energy within the regional evaluation location set are statistically analyzed to characterize the spectral response characteristics of the complete target region. The intersection of the effective fatigue evaluation location sets for each working condition to be evaluated is taken to form a common effective fatigue evaluation location set. Within this common effective fatigue evaluation location set, the apparent spectral transfer index and fatigue damage rate are statistically analyzed, and the apparent spectral transfer ranking r is determined from high to low according to the regional median apparent spectral transfer index and regional median fatigue damage rate. H,j And fatigue risk ranking r D,j Both ranking categories use the same tie-ranking rules, with ranking 1 indicating the highest corresponding indicator. Spectral indicators obtained from the complete regional evaluation location set are output as auxiliary results; when the number of valid locations for common fatigue evaluation is lower than a preset lower limit, an indicator of insufficient evaluation validity due to ranking mismatch is output. The preset lower limit can be determined based on the stability of regional statistics and ranking results under changes in the number of valid locations for common fatigue evaluation, and should remain consistent within the same evaluation task.

[0118] For K ≥ 2 working conditions to be evaluated, the normalized ranking mismatch degree of the j-th working condition can be calculated using the following formula. A ranking of 1 indicates the highest indicator or the highest risk of fatigue. ∈[0,1].

[0119]

[0120] The ranking of apparent spectral transfer indicators, fatigue risk ranking, input spectral weighted coherence, and normalized ranking mismatch for each working condition are summarized to form a joint evaluation matrix. This joint evaluation matrix displays the correspondence between spectral transfer strength, input-output coherence, and fatigue risk ranking. Depending on specific engineering needs, additional thresholds for transfer ranking quantiles, fatigue risk ranking quantiles, and coherence can be set to form explanatory labels; these explanatory labels are not necessary conditions for determining fatigue risk ranking in this invention, and fatigue risk ranking is still determined by regional fatigue damage statistics.

[0121] Example 6: Evaluation of Stability and Assembly Status of Samples Across Loads

[0122] Figure 5 This invention provides a flowchart for joint sorting, mismatch identification, and stability evaluation within a given load sample set. For at least three independent load samples s, the regional fatigue damage index D for each working condition j is calculated within each sample. s,j And perform in-sample normalization based on the maximum value within the sample:

[0123]

[0124] When the denominator is zero, the load sample is marked as a sample without effective damage and is not included in the cross-sample statistics.

[0125] The normalization is used to compare the relative trends of different samples without changing the risk ranking within a sample. All load samples use the same set of working conditions, regional evaluation purpose, number of evaluation locations, analysis time window, spectral estimation parameters, target frequency band, and fatigue material parameters. The ranking within a sample is determined based on the original regional fatigue damage index; when the same index exists, a tie-breaking ranking is used. The frequency P of each working condition becoming the highest-risk working condition within the sample is calculated. 1,j and cross-sample ranking interquartile range The calculation formula is as follows:

[0126]

[0127] in, Let j be the sample set containing the valid rankings of the j-th working condition. =| | represents the number of valid samples, and I(·) is an indicator function that takes the value 1 when the condition in parentheses is true, and 0 otherwise.

[0128]

[0129] In the formula, Let be the in-sample risk ranking of working condition j in the s-th sample.

[0130] When a load sample lacks any of the comparison conditions, that load sample is not included in the ranking stability statistics based on the complete set of conditions; alternatively, a subset of conditions commonly included in all load samples is pre-determined, and all effective load samples are re-normalized and re-ranked according to the same subset of conditions. The indices are used to describe relative stability within a given sample set and are not interpreted as overall probability of occurrence or statistical confidence intervals. When the number of effective load samples is small, the ability to distinguish the frequency of highest risk occurrence and the quartile range of ranking is limited by the sample size; increasing the number of mutually independent load samples can improve the stability indices' ability to characterize cross-sample differences, but does not change the stability evaluation method.

[0131] When comparing different installation preload, clamping force, contact state, support stiffness, or other assembly states, a common high-response location set is used to calculate spectral transfer, overall dynamic transfer, and fatigue damage indices. Further comparisons can be made of damage trends with and without mean stress correction to help identify the relative impact of dynamic cyclic amplitude variations and mean stress variations on fatigue risk trends; this comparison should not be interpreted as two strictly independent and additive damage components.

[0132] Example 7: Verification Application of a Specific Fastener Spring Strip

[0133] To verify the feasibility of this invention, a high-speed railway elastic clip fastener was used as an example. The on-site wheel-rail load background and multiple short-spatial-scale excitation conditions were input, and multiple installation torque conditions were set. The finite element model output the equivalent stress time history of multiple elements in the root region of the elastic clip. The first analysis time window was set to 0~40ms; the number of evaluation locations in the region was N=200, the target frequency band half-width was 30Hz, and sensitivity checks were performed on the candidate N values ​​and half-width.

[0134] To determine the second analysis time window, three candidate time windows—6–31 ms, 6–36 ms, and 6–40 ms—were set, with 6–36 ms serving as the baseline time window. For representative spatial scale conditions, an independent high-response evaluation location set was determined within the baseline time window, and this evaluation location set remained unchanged within the remaining candidate time windows to avoid the simultaneous impact of time window changes and evaluation area changes on the sensitivity verification results. The root mean square of the regional median dynamic stress and the energy of the preset frequency band stress spectrum were calculated for each candidate time window. The results show that the change in the root mean square of the regional median dynamic stress relative to the baseline time window does not exceed 2.0% within each candidate time window; the energy of the preset frequency band stress spectrum remains stable under most conditions, but a significant deviation occurs in the 100 mm condition within the 6–31 ms time window, indicating that this time window is highly sensitive to the truncation of the subsequent dynamic response. Considering the initial transient exclusion, retention of major vibration responses, and stability of evaluation indicators, 6–36 ms was selected as the second analysis time window for subsequent dynamic response, spectral transfer, and coherence analysis.

[0135] In one load sample of this embodiment, the median fatigue damage rate was highest in the 100mm spatial scale condition. The median apparent spectral transfer indices for the 120mm and 130mm conditions were approximately 0.735 MPa / kN and 0.774 MPa / kN, respectively. The weighted coherence of the median input spectrum in the 120mm condition was approximately 0.164, while that in the 130mm condition was approximately 0.680. Neither of these conditions resulted in the highest fatigue damage rate. This demonstrates that a high apparent spectral transfer index in the target frequency band does not necessarily correspond to the highest fatigue risk, and a low coherence may indicate that the enhancement of the stress spectrum in the target frequency band is not entirely driven by the stable and coherent target input load. Through the joint sorting and mismatch calculation of this invention, situations where the spectral transfer peak value is inconsistent with the condition of highest fatigue risk can be identified.

[0136] When three independent field load segments were used for repeated evaluation, the spatial scales of the highest fatigue risk in the three load samples were 70mm, 100mm, and 110mm, respectively, indicating that the highest risk condition was affected by the random load background. However, each sample showed a non-monotonic risk ranking, and the condition corresponding to the apparent spectral transfer peak was not necessarily the highest fatigue risk condition. The frequency of occurrence of the highest risk and the interquartile range of the ranking can be used to distinguish between conditions that repeatedly exhibit high risk in a given sample set and conditions that are more sensitive to load samples. When comparing multiple installation torques, the overall dynamic transfer level and the proportion of high-amplitude cycles generally decreased with increasing torque, but the mean stress change, after Goodman correction, partially weakened the damage reduction trend caused by the decrease in dynamic cycle amplitude. This embodiment illustrates that the present invention can simultaneously output fatigue risk ranking and the spectral transfer, coherence, and cyclic damage information that forms this ranking.

[0137] The fastener type, operating speed, spatial dimensions, torque range, time window, number of regional evaluation locations N, target frequency band, and material parameters in the above embodiments are all verification parameters and do not constitute a limitation on the scope of application of the present invention. For other types of fastener springs or fastener systems with elastic clamping components, the fatigue evaluation stress parameter time history of the target area can be re-acquired according to its structural form, material properties, and assembly state, and the corresponding analysis time window, number of regional evaluation locations, target frequency, target frequency band, and fatigue material parameters can be determined. A comparative fatigue risk assessment can then be performed according to the steps described in this invention.

[0138] Example 8: System and Standardized Output

[0139] Figure 6 This is a schematic diagram illustrating the module composition of a fatigue risk joint assessment system provided in an embodiment of the present invention. Figure 6As shown, the evaluation system includes a processor and a memory, with the memory storing a computer program. When executed by the processor, the computer program performs data acquisition and quality verification, dual-time-window processing, construction of regional evaluation location sets, spectral transfer and coherence calculation, cyclic fatigue analysis, effectiveness verification and damage decomposition, regional statistics and joint ranking, cross-sample stability evaluation when at least three independent load samples exist, and standardized result output. The processor can be a single processor or multiple interconnected computing units.

[0140] The system output should include at least: a table of operating parameters; a list of evaluation locations; tables of location-level spectral transfer, coherence, and fatigue indices; a summary table of regional medians and quantiles; a fatigue risk ranking table; results of spectral transfer and risk ranking mismatch; contribution of cyclic damage intervals; frequency of highest risk occurrence and interquartile range of ranking within a given load sample set; and the status of input data, spectral calculations, and fatigue evaluation validity verification. Output data can be in the form of CSV, spreadsheets, database records, graphical reports, or a combination thereof.

[0141] This invention yields comparative fatigue risk results under given loads, models, material fatigue parameters, and evaluation windows. If used for absolute service life prediction, factors such as load frequency, material dispersion, environmental impact, manufacturing residual stress, wear degradation, and test calibration should also be considered.

[0142] In summary, the embodiments of the present invention use a first analysis time window covering the effective load action process and a second analysis time window excluding the initial transient state for fatigue evaluation and spectral transfer evaluation, respectively, thereby reducing the interference of the initial transient and static components on the dynamic spectral index while retaining the cyclic mean stress information.

[0143] By selecting either an independent high-response location set or a cross-condition common high-response location set based on the evaluation objective, the independent high-response location set is used to track the spatial changes of high-response areas under different operating conditions, reducing the impact of missing local high-response areas at fixed evaluation locations. The cross-condition common high-response location set is used to compare changes in operating condition parameters based on a unified spatial location, reducing the interference of evaluation area drift on the comparison of response trends. Further reducing the impact of location set size selection on regional statistical results can be achieved by combining the stability of regional response statistics, the stability of location sets under repeated samples or disturbance conditions, and the spatial coverage requirements of the target area to determine the number of evaluation locations.

[0144] By combining the apparent spectral transfer index with the input spectrum weighted coherence, it is possible to help determine whether the enhancement of the stress spectrum within the target frequency band is mainly driven by the target input load in a stable and coherent manner, thus avoiding directly equating the enhancement of the target frequency band stress spectrum with high fatigue risk.

[0145] By determining the risk ranking based on regional fatigue damage statistics within the effective location set, and combining the high-amplitude cyclic proportion and cyclic amplitude interval damage contribution to explain the risk formation process, it is possible to reveal the cyclic redistribution in the process of spectrum transfer transforming into fatigue damage.

[0146] By comparing the spectral transmission ranking with the fatigue risk ranking and quantifying the ranking differences, we can identify situations where the peak value of the spectral index is inconsistent with the fatigue hazard conditions.

[0147] By ranking the risk within a given load sample set, the frequency of occurrence of the highest risk, and the interquartile range of the ranking using at least three independent load samples, the relative stability of the risk ranking with respect to random load background changes can be described. This method is suitable for comparative fatigue risk screening under multiple fastener types, multiple excitation conditions, and multiple assembly states.

[0148] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of one embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing the present invention.

[0149] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of the present invention.

[0150] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for apparatus or system embodiments, since they are basically similar to method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments. The apparatus and system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. 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 the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0151] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for jointly assessing the load-stress spectrum transfer and fatigue risk of railway fastener elastic bars, characterized in that, include: Obtain the status information of railway fastener elastic strips under multiple working conditions to be evaluated; A first and second analysis time window are established to describe the effective load process of railway fastener elastic strips. The fatigue evaluation stress parameters in the first analysis time window retain the cyclic mean and are used for fatigue evaluation. The target scalar load and fatigue evaluation stress parameters in the second analysis time window are respectively processed to remove the mean and used for spectral transfer and coherence evaluation. Construct a set of regional evaluation locations based on the evaluation purpose; construct an independent set of high-response locations for each working condition when identifying its own high-response areas; and construct a set of common high-response locations across working conditions when comparing the response differences of the same spatial region with changes in working condition parameters. Based on the state information of the railway fastener elastic bar, the load-stress apparent spectrum transfer index and input-output coherence index in the target frequency band are calculated using the mean-reduced target scalar load and mean-reduced fatigue evaluation stress parameters within the second analysis time window. Based on the fatigue evaluation stress parameter time history within the first analysis time window, cyclic counting, mean stress correction, material fatigue relationship calculation, and damage accumulation are performed to obtain location-level fatigue damage indicators. The location-level fatigue damage indicators that meet the fatigue evaluation validity conditions in the evaluation location set of the region are statistically analyzed, and the fatigue risk ranking of each working condition is determined based on the effective location-level fatigue damage statistics. The fatigue risk ranking of each working condition of the railway fastener elastic bar is compared with the ranking of the apparent spectral transfer index to determine the degree of mismatch between the spectral transfer and fatigue risk ranking of each working condition. The joint evaluation result of each working condition of the railway fastener elastic bar is output in combination with the coherence index.

2. The method according to claim 1, characterized in that, The acquisition of status information of railway fastener elastic bars under multiple working conditions to be evaluated includes: The system acquires the status information of railway fastener elastic bars under multiple evaluation conditions. This status information includes the target input load time history for each condition, the fatigue evaluation stress parameter time history for multiple evaluation locations in the target area of ​​the fastener elastic bar, and the condition parameters. It identifies the time series, load components, and evaluation location identifiers, and completes finite value checks, time sorting, duplicate time point processing, and time alignment within a common effective time range.

3. The method according to claim 2, characterized in that, The establishment of the first and second analysis time windows for the effective load action process of the railway fastener elastic strip includes the following: In the first analysis time window, the fatigue evaluation stress parameters retain their cyclic mean and are used for fatigue evaluation; in the second analysis time window, the target scalar load and fatigue evaluation stress parameters are respectively subjected to mean-reduction processing and used for spectral transfer and coherence evaluation, including: A first analysis time window covering the effective load action process is established, and a second analysis time window is set within the first analysis time window. When performing time window sensitivity verification, a reference candidate time window is selected from multiple candidate time windows located within the first analysis time window. A pre-selected evaluation location set for sensitivity verification is constructed based on the dynamic response index of each evaluation location within the reference candidate time window. Under the condition of keeping the pre-selected evaluation location set and the spectrum estimation parameters consistent, the regional mid-dynamic response index and the preset frequency band stress spectrum energy corresponding to each candidate time window are calculated based on the state information of the railway fastener elastic bar under the evaluation condition. The second analysis time window is determined based on the degree of change of the index relative to the reference candidate time window. The pre-selected evaluation location set is used for time window sensitivity verification. The fatigue evaluation stress parameters in the first analysis time window retain the cyclic mean and are used for fatigue evaluation. The target scalar load and fatigue evaluation stress parameters in the second analysis time window are respectively processed to remove the mean and used for spectral transfer and coherence evaluation. The same second analysis time window is used for each working condition in the same comparison group.

4. The method according to claim 2, characterized in that, The construction of a regional evaluation location set according to the evaluation purpose, the construction of an independent high-response location set for each working condition when identifying its own high-response area, and the construction of a cross-working-condition common high-response location set when comparing the response differences of the same spatial region with changes in working condition parameters include: When the evaluation objective is to identify the high-response region for each working condition, the independent high-response location set for each working condition consists of the top N evaluation locations ranked by the dynamic stress response index within that working condition. When the evaluation objective is to compare the response differences of the same spatial region with variations in working condition parameters, the common high-response location set is selected from the valid evaluation locations shared by all working conditions to be compared, and the top N locations are selected based on the cross-working-condition statistics of the normalized dynamic stress response index for each working condition. Evaluation position; The determination is based on the stability of the regional response statistics under the number of candidate locations, the stability of the evaluation location set under repeated samples or perturbation analysis, and the spatial coverage requirements of the target area.

5. The method according to claim 4, characterized in that, Based on the state information of the railway fastener elastic bar, and using the mean-reduced target scalar load and mean-reduced fatigue evaluation stress parameters within the second analysis time window, the load-stress apparent spectrum transfer index and input-output coherence index within the target frequency band are calculated, including: For a predetermined target scalar load channel, the power spectrum of the target scalar load and the power spectrum of the fatigue evaluation stress parameters within the second analysis time window are calculated using consistent spectral estimation parameters based on the state information of the railway fastener elastic bar under the working condition to be evaluated. The cross power spectrum between the target scalar load and the fatigue evaluation stress parameters is also calculated. A target frequency band is constructed around the target frequency, and the position-level load-stress apparent spectrum transfer index and input-output coherence index within the target frequency band are calculated. The target frequency is any one of the following: a preset excitation frequency, a fastener spring characteristic frequency, or a load spectrum characteristic frequency of the target scalar load channel; or it can be determined based on the operating speed v and the spatial excitation scale. according to Confirmed; the target frequency band is ;No. The evaluation position is relative to the first Apparent spectral transfer index of each target scalar load channel The energy of the power spectrum of the fatigue evaluation stress parameter within the target frequency band and the first The square root of the ratio of the power spectrum energies of the target scalar load channels is determined.

6. The method according to claim 5, characterized in that, The method involves performing cyclic counting, mean stress correction, material fatigue relationship calculation, and damage accumulation based on the fatigue evaluation stress parameter time history within the first analysis time window to obtain location-level fatigue damage indices. It then statistically analyzes the location-level fatigue damage indices that meet the fatigue evaluation validity conditions for the concentrated evaluation locations in the region, and determines the fatigue risk ranking of each working condition of the railway fastener elastic bar based on the effective location-level fatigue damage statistics, including: Based on the fatigue evaluation stress parameter time history within the first analysis time window, cycle counting, mean stress correction, material fatigue relationship calculation, and damage accumulation are performed. Evaluation locations within the regional evaluation location set that have a Goodman mean stress correction denominator not greater than zero are marked, and the Goodman corrected damage results of these evaluation locations are not included in the regional fatigue damage quantitative statistics. The remaining evaluation locations with effective fatigue calculation results are combined into a fatigue evaluation effective location set. The location-level fatigue damage indicators within the fatigue evaluation effective location set are statistically analyzed, and the fatigue risk ranking for each working condition is determined by the regional median damage rate or the regional damage statistics that monotonically correspond to it. The cycle count uses rainflow counting; the mean stress correction uses Goodman correction; the material fatigue relationship uses stress-life relationship; the damage accumulation uses Miner linear damage accumulation criterion; the fatigue damage index includes at least the damage rate per unit time, and may further include at least one of cumulative damage, equivalent cycle index and reference cycle requirement pass rate.

7. The method according to claim 6, characterized in that, The method of comparing the fatigue risk ranking of the railway fastener elastic clip under various working conditions with the ranking of apparent spectral transfer indices to determine the degree of mismatch between spectral transfer and fatigue risk ranking for each working condition, and outputting the joint evaluation results of the railway fastener elastic clip under various working conditions in conjunction with coherence indices, includes: For K ≥ 2 working conditions to be evaluated, based on the common effective fatigue evaluation location set formed by the intersection of the effective fatigue evaluation location sets of each working condition, the regional apparent spectrum transfer index and the regional fatigue damage index are statistically analyzed to determine the ranking of the apparent spectrum transfer index. And fatigue risk ranking and in accordance with | - | Determine the first The normalized ranking mismatch degree of each working condition to be evaluated is determined; the apparent spectral transfer index ranking, fatigue risk ranking, input spectrum weighted coherence, and normalized ranking mismatch degree of each working condition to be evaluated are output to characterize the consistency or mismatch degree between spectral transfer and fatigue risk, and a joint evaluation result of each working condition to be evaluated is formed. The joint evaluation result includes spectral transfer ranking, fatigue risk ranking, ranking mismatch degree, and related auxiliary interpretation indicators.

8. The method according to claim 1, characterized in that, The method further includes: for no less than three independent load samples, normalizing and ranking each working condition according to the regional fatigue damage index within each load sample; and statistically analyzing the frequency of each working condition becoming the highest-risk working condition. and interquartile range of cross-sample ranking The stability of fatigue risk ranking within a given load sample set is characterized by the frequency of occurrence of the highest-risk working condition and the interquartile range of the cross-sample ranking. When multiple working conditions have the same fatigue damage index in the same area, they are ranked in parallel. When a load sample is missing any working condition to be compared, the load sample is not included in the ranking stability statistics based on the complete set of working conditions. Alternatively, a subset of working conditions commonly included in each load sample is predetermined, and all valid load samples are normalized and ranked according to the same subset of working conditions.

9. A joint assessment system for load-stress spectrum transfer and fatigue risk of railway fastener elastic bars, characterized in that, The system includes a processor and a memory, the memory storing a computer program that, when executed by the processor, causes the system to perform the following operations: acquire and align the time history of the target input load and the time history of fatigue evaluation stress parameters at multiple evaluation locations in the target area; Construct the first and second analysis time windows; construct an independent high-response location set or a common high-response location set across operating conditions according to the evaluation purpose; calculate the target frequency band apparent spectrum transfer index and input-output coherence index; calculate the fatigue damage index and cyclic damage interval contribution. Verify the effectiveness of fatigue calculation and construct a set of effective locations for fatigue evaluation; determine the fatigue risk ranking and the degree of mismatch between spectrum transfer and fatigue risk ranking; when there are no less than three independent load samples, complete the normalization and ranking of the working conditions in each load sample, and summarize and statistically analyze the frequency of occurrence of each working condition as the highest risk working condition and the interquartile range of the cross-sample ranking, and output the fatigue risk ranking stability results in the given load sample set.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the method for joint assessment of load-stress spectrum transfer and fatigue risk of railway fastener elastic bars as described in any one of claims 1 to 8.

Images