Method, device, and apparatus for predicting fatigue damage of structural member, and storage medium

Abstract

To solve the problem that, in the prediction of fatigue damage of a structural member, conventional methods have limitations in accuracy and practicality, and in particular, when relying on artificial judgment, variability in judgment, and uncertainty in prediction.SOLUTION: A method and a device for predicting fatigue damage of a structural member, and a storage medium are disclosed. A crack state of a target structural member is judged on the basis of a voiceprint signal, and its result is compared with artificial judgment and has high accuracy. operation state parameters of second predetermined time are determined on the basis of history synchronization data, so that the method has high reliability. Also, a second finite element structure model is a simulation model conforming to the current actual condition of a target member, stress data applied to the model is the data close to the actual operation state, and a prediction result based on both the model and the stress data has high accuracy and authenticity. Thus, a strong basis is provided for safe operation and maintenance of a generator set.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to the field of prediction technology, and more particularly to a method, apparatus, device and storage medium for predicting fatigue damage in structural members. [Background technology]

[0002] Generator-generator sets are prone to fatigue damage due to continuous and complex load bearing, and if the degree of fatigue damage is serious, it will affect the safe and stable operation of the pumped storage power plant. Therefore, timely monitoring of the health status of the structural members of the generator sets and relatively accurate prediction of the degree of fatigue damage plays a positive role in the stable operation of the pumped storage power plant.

[0003] Currently, the degree of fatigue damage to the structural components of a generator set is mainly predicted by on-site manual inspection. However, this method makes it difficult to detect initial or small cracks in the equipment in a timely manner, and it is also difficult to predict the expansion of cracks and the instability of structural components through inspection, which is detrimental to the safe and stable operation of the generator set. Summary of the Invention [Problem to be solved by the invention]

[0004] To solve the above problems, the present disclosure provides a method, an apparatus, a device, and a storage medium for predicting fatigue damage in a structural member. [Means for solving the problem]

[0005] In a first aspect, the present invention provides a method for predicting fatigue damage in a structural member, comprising the steps of: The method includes the steps of: acquiring a voiceprint signal generated by the target structural member within a first predetermined time, a first finite element structural model corresponding to the target structural member, material characteristic parameters of the target structural member, and driving situation parameters within a second predetermined time, the first predetermined time having a current time as an end time, the second predetermined time having a current time as a start time, and the driving situation parameters being determined based on history-synchronized driving situation parameters; determining a crack state of the target structural member based on the voiceprint signal; setting a crack in the first finite element structural model based on the crack state; and setting a second finite element structural model. a step of obtaining a model of the target structural member during the second predetermined time; a step of loading operating condition parameters within a second predetermined time into a first finite element structural model as boundary conditions, combining the model with material characteristic parameters, and performing transient dynamic analysis to obtain stress data corresponding to each predetermined position of the target structural member during the second predetermined time; and a step of adding the corresponding stress data at each predetermined position of the second finite element structural model during the second predetermined time, performing a simulation prediction of the fatigue damage status of the target structural member during the second predetermined time, and obtaining a first prediction result.

[0006] In this embodiment, the method for predicting fatigue damage to a structural member includes: obtaining a voiceprint signal of a target structural member within a first predetermined time period, a first finite element structural model corresponding to the target structural member, material characteristic parameters of the target structural member, and operating condition parameters within a second predetermined time period; first, judging the crack state of the target structural member at the current time based on the voiceprint signal; placing the crack in the first finite element structural model; and obtaining a second finite element structural model; the judgment result obtained based on the voiceprint signal has higher accuracy compared with the current artificial judgment form; further, loading the operating condition parameters for the second predetermined time period into the first finite element structural model and combining them with the material characteristic parameters to predict stress data corresponding to each predetermined position of the target structural member within the second predetermined time period, thereby obtaining the force-bearing state of each position of the target structural member within a certain period of time in the future; and The operating condition parameters for the second predetermined time period are determined based on historical synchronized data, so they have relatively high reference value and can make the predicted force-bearing condition closer to the actual condition. Finally, within the second predetermined time period, corresponding stress data are added to each predetermined position of the second finite element structural model to simulate and predict the fatigue damage condition of the target structural member within the second predetermined time period, thereby obtaining a first prediction result. Since the second finite element structural model is a simulation model that matches the current actual condition of the target structural member, the stress data added to the second finite element structural model is data that is closer to the actual operating condition. Therefore, the prediction result obtained by prediction based on both of them will necessarily have relatively high accuracy and truthfulness, be of relatively high reference value, and provide an important basis for the safe operation and maintenance of the generator set.

[0007] In one alternative embodiment, after acquiring a voiceprint signal generated by the target structural member within a first predetermined time period, the method includes: The method further includes the steps of performing noise reduction processing on the voiceprint signal to obtain noise-reduced signals; using an ensemble empirical mode decomposition method to decompose each noise-reduced signal into a residual component and at least two eigenmode components; reconstructing the noise-reduced signal based on the eigenmode components and the residual components to obtain a reconstructed signal; and determining the crack state of the target structural member based on the reconstructed signal.

[0008] In an alternative embodiment, the step of reconstructing the noise-reduced signal based on the eigenmode components and the residual components to obtain the reconstructed signal comprises: The method includes the steps of: determining whether or not a noise component is present in at least two eigenmode components corresponding to the noise-reduced signal; performing noise removal on the noise component if the noise component is present, and obtaining a dominant component after the noise removal; and adding together the residual component, the dominant component after the noise removal, and the eigenmode components other than the noise component to obtain a reconstructed signal.

[0009] In the method for predicting fatigue damage of a structural member according to this embodiment, after obtaining a voiceprint signal generated within a first predetermined time, the voiceprint signal is further subjected to noise reduction, decomposition, and reconstruction to remove noise from the voiceprint signal, thereby making the crack state of the target structural member determined based on the reconstructed signal more accurate and further making the final prediction result more accurate.

[0010] In one alternative embodiment, the step of determining the crack state of the target structural member based on the voiceprint signal comprises: The method includes the steps of extracting feature parameters from the voiceprint signal, performing dimension reduction processing on the feature parameters and extracting signal principal components from the dimension-reduced feature parameters, and recognizing the signal principal components to obtain the crack state of the target structural member.

[0011] In one alternative embodiment, each predetermined position of the target structural member is The target structural member is determined by obtaining operating condition parameters within a first predetermined time, loading the operating condition parameters within the first predetermined time into a first finite element structural model as boundary conditions, obtaining a stress distribution status of the target structural member within the first predetermined time, determining a position to be studied of the target structural member based on the stress distribution status, and determining the position to be studied as a predetermined position.

[0012] In an alternative embodiment, the operating situation parameters within the second predetermined time are loaded as boundary conditions into the first finite element structural model, and combined with the material characteristic parameters to perform transient dynamic analysis, and after obtaining stress data respectively corresponding to each predetermined position of the target structural member within the second predetermined time, the method includes: The method further includes predicting the remaining life of the target structural member based on the stress data corresponding to each predetermined position within a second predetermined time period to obtain a second prediction result.

[0013] In a second aspect, the present invention provides an apparatus for predicting fatigue damage in a structural member, comprising: an acquisition module for acquiring a voiceprint signal generated by the target structural member within a first predetermined time, a first finite element structural model corresponding to the target structural member, material characteristic parameters of the target structural member, and driving situation parameters within a second predetermined time, wherein the first predetermined time has a current time as an end time and the second predetermined time has a current time as a start time, and the driving situation parameters are determined based on history-synchronized driving situation parameters; and an acquisition module for determining a crack state of the target structural member based on the voiceprint signal, and locating cracks in the first finite element structural model based on the crack state, to obtain a second finite element structural model. an analysis module for loading operating condition parameters within a second predetermined time into a first finite element structural model as boundary conditions and combining them with material characteristic parameters to perform transient dynamic analysis and obtain stress data corresponding to each predetermined position of the target structural member within the second predetermined time; and a first prediction module for adding corresponding stress data at each predetermined position of the second finite element structural model within the second predetermined time, performing simulation prediction of the fatigue damage status of the target structural member within the second predetermined time, and obtaining a first prediction result.

[0014] In one alternative embodiment, after the acquisition module, the device: The system further includes a noise reduction module for performing noise reduction processing on the voiceprint signal to obtain a noise-reduced signal; a decomposition module for decomposing each noise-reduced signal into a residual component and at least two eigenmode components using an ensemble empirical mode decomposition method; and a reconstruction module for reconstructing the noise-reduced signal based on the eigenmode component and the residual component to obtain a reconstructed signal, and determining the crack state of the target structural member based on the reconstructed signal.

[0015] In an alternative embodiment, the reconstruction module: The noise reduction signal includes a determination submodule for determining whether or not a noise component is present in at least two eigenmode components corresponding to the noise-reduced signal; a noise removal submodule for removing the noise component if a noise component is present and obtaining a dominant component after the noise removal; and a reconstruction submodule for adding the residual component, the dominant component after the noise removal, and the eigenmode components other than the noise component to obtain a reconstructed signal.

[0016] In an alternative embodiment, the determination module: The system includes a first extraction submodule for extracting feature parameters from the voiceprint signal, a second extraction submodule for performing dimension reduction processing on the feature parameters and extracting signal principal components from the feature parameters after dimension reduction, and a recognition submodule for recognizing the signal principal components and obtaining the crack state of the target structural member.

[0017] In one alternative embodiment, each predetermined position of the target structural member in the analysis module is The method is determined by an acquisition submodule for acquiring operating condition parameters within a first predetermined time period, a calculation submodule for loading the operating condition parameters within the first predetermined time period into a first finite element structural model as boundary conditions and obtaining the stress distribution status of the target structural member within the first predetermined time period, and a determination submodule for determining a position to be studied of the target structural member based on the stress distribution status and determining the position to be studied as a predetermined position.

[0018] In one alternative embodiment, after the analysis module, the device: The system further includes a second prediction module for predicting the remaining life of the target structural member based on the stress data respectively corresponding to each predetermined position within a second predetermined time period to obtain a second prediction result.

[0019] In a third aspect, the present invention provides a computing device comprising: There is provided a computer device including a memory and a processor, the memory and the processor being communicatively connected, the memory storing computer instructions, the processor executing the computer instructions to perform the method for predicting fatigue damage in a structural member according to any one of the embodiments of the first aspect of the Summary of the Invention.

[0020] In a fourth aspect, the present invention provides a computer-readable storage medium having stored thereon computer instructions for causing a computer to perform the method for predicting fatigue damage in a structural member according to any of the embodiments of the first aspect of the Summary of the Invention. [Brief explanation of the drawings]

[0021] The drawings herein are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure, and together with the specification serve to explain the principles of the present disclosure. In the following, in order to more clearly explain the technical solutions in the embodiments of the present disclosure or the prior art, drawings that need to be used to explain the embodiments or the prior art are briefly introduced, and it is obvious that a person skilled in the art can derive other drawings based on these drawings without any creative effort. [Figure 1] 1 is a flowchart of a method for predicting fatigue damage in a structural member according to an embodiment of the present invention. [Figure 2] 4 is a flowchart of a method for predicting fatigue damage in a structural member according to another embodiment of the present invention. [Figure 3] 10 is a flowchart of a method for predicting fatigue damage in a structural member according to yet another embodiment of the present invention. [Figure 4] 1 is a block diagram of the structure of an apparatus for predicting fatigue damage of a structural member according to an embodiment of the present invention; [Figure 5] 1 is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention; DETAILED DESCRIPTION OF THE INVENTION

[0022] In order to make the above-mentioned objects, features and advantages of the present disclosure more clearly understood, the solutions of the present disclosure will be further described below. Furthermore, where not inconsistent, the embodiments and features in the embodiments of the present disclosure can be combined with each other.

[0023] In order to fully understand the present disclosure, many specific details are described in the following description, but it is clear that the present disclosure may be implemented in other forms different from those described herein, and the embodiments in the specification are only a part of the embodiments of the present disclosure, and are not all of the embodiments. All other embodiments that can be obtained by a person skilled in the art based on the embodiments of the present invention without any creative effort belong to the scope of protection of the present invention.

[0024] According to an embodiment of the present application, an embodiment of a method for predicting fatigue damage in a structural component is provided, and it is noted that the steps shown in the flowcharts of the drawings may be performed in a computer system as a set of computer-executable instructions, and that although a logical order is shown in the flowcharts, in some cases the steps shown or described may be performed in a different order than in this specification.

[0025] In this embodiment, a method for predicting fatigue damage of a structural member applicable to a detection device is provided, and FIG. 1 is a flowchart of the method for predicting fatigue damage of a structural member according to an embodiment of the present invention. As shown in FIG. 1, the flow includes the following steps S101 to S104.

[0026] In step S101, a voiceprint signal generated by a target structural member within a first predetermined time, a first finite element structural model corresponding to the target structural member, material characteristic parameters of the target structural member, and driving situation parameters within a second predetermined time are obtained.

[0027] Specifically, the first predetermined time is a historical time period starting from any one of the historical times and ending at the current time. The target structural member is a structural member of the generator set, and may be, in particular, an important structural member that affects the operation of the generator set, such as a runner or a generator set blade. The voiceprint signal can be obtained from a signal collected by a voiceprint sensor attached to the wall of the draft tube.

[0028] For example, in one selectable embodiment, the target structural component is a runner, and voiceprint signals collected at each collection time over the past six months can be obtained. Before implementing this method, the surface of the runner's draft tube is first cleaned of coating and oil stains, and then filled with Vaseline binder. Finally, a voiceprint sensor is attached to the wall of the runner's draft tube, and the voiceprint sensor parameters are set so that the voiceprint sensor collects voiceprint signals according to the set requirements. In this embodiment, the voiceprint sensor parameters can be set as follows: instrument frequency bandwidth 50-200 kHz, sampling frequency 1 MHz, pre-trigger time 100 μs, filtering range 5-200 kHz, threshold 40 dB, peak limit time 300 μs, collision limit time 600 μs, and collision closure time 1000 μs.

[0029] Specifically, the first finite element structural model corresponding to the target structural member is pre-constructed by finite element analysis software before implementing this method, and the finite element analysis software includes, but is not limited to, ANSYS, ABAQUS, Hyperworks, etc.

[0030] Specifically, the material characteristic parameters of the target structural member can be obtained from specific information of the structural member. For example, the target structural member is a runner, and the characteristic parameters of the runner include, but are not limited to, yield limit, strength, toughness, etc.

[0031] Specifically, the second predetermined time period starts from the current time and represents a certain period in the future. The driving situation parameters within the second predetermined time period are driving situation parameters corresponding to each time within the second predetermined time period, and the driving situation parameters for the second predetermined time period are determined based on the history-synchronized driving situation parameters. For example, the driving situation parameters for the second predetermined time period may be the average value of the synchronized driving situation parameters over the past few years, or the synchronized driving situation parameters for the year closest to the environmental conditions of the current year may be determined as the driving situation parameters for the second predetermined time period. Furthermore, the driving situation parameters for the second predetermined time period may be obtained by prediction based on the history-synchronized driving situation parameters using a neural network model. This embodiment does not specifically limit the manner in which the driving situation parameters for the second predetermined time period are determined.

[0032] In step S102, the crack state of the target structural member is determined based on the voiceprint signal, and the crack is placed in the first finite element structural model based on the crack state to obtain a second finite element structural model.

[0033] Specifically, the manner of determining the crack state based on the voiceprint signal includes, but is not limited to, determining based on the signal-to-noise ratio, determining by recognizing the characteristics of the voiceprint signal, and the like.

[0034] Specifically, the crack state includes a crack condition and a duration, the crack condition includes no crack, crack occurrence, crack extension and crack destabilization extension, and the duration is the duration of the current crack condition.

[0035] Specifically, the crack state corresponding to the target structural member in the current operating conditions is determined based on the voiceprint signal for a first predetermined time, and a crack corresponding to the current crack state is placed in the first finite element structural model based on the current crack state, thereby obtaining a second finite element structural model including the crack. When placing a crack in the first finite element structural model, a crack with a width, depth, and length corresponding to the duration of the current crack can be placed in a position where cracks are likely to occur based on the duration of the current crack. The position where cracks are likely to occur includes, but is not limited to, a welded area.

[0036] In step S103, the operating condition parameters within the second predetermined time are loaded into the first finite element structural model as boundary conditions, and combined with the material characteristic parameters to perform transient dynamic analysis, thereby obtaining stress data corresponding to each predetermined position of the target structural member within the second predetermined time.

[0037] Specifically, each predetermined position corresponds to a set of stress data within a second predetermined time, and the stress data within the second predetermined time is stress data corresponding to each time in the second predetermined time, and the stress data within the second predetermined time may be represented in an array format or in an image format. The stress data at the predetermined position a within the second predetermined time is (t1,f a1 ), (t2,f a2 ), …, (t n ,f an ), and the stress data at the predetermined position b within the second predetermined time may be (t1,f b1 ), (t2,f b2 ), …, (t n ,f bn ) may also be used.

[0038] Exemplarily, the predetermined locations include, but are not limited to, high stress areas, fatigue risk areas, and risk locations, and there is stress data corresponding to each predetermined location at each time of the second predetermined time, and stress data corresponding to different predetermined locations at the same time are not necessarily the same, and stress data corresponding to the same predetermined location at different times are not necessarily the same, and the stress data is related to the driving situation parameters and acting locations at the current time.

[0039] Specifically, each predetermined position of the target structural member is The method includes obtaining operating condition parameters within a first predetermined time period, loading the operating condition parameters within the first predetermined time period into a first finite element structural model as boundary conditions, obtaining a stress distribution condition of the target structural member within the first predetermined time period, determining a position to be studied of the target structural member based on the stress distribution condition, and determining the position to be studied as a predetermined position, where the stress distribution condition can be represented by a heat map, and recognizing the heat map determines the position to be studied of the target structural member.

[0040] In step S104, corresponding stress data is added at each predetermined position of the second finite element structural model within a second predetermined time, and a simulation prediction is performed on the fatigue damage status of the target structural member within the second predetermined time to obtain a first prediction result.

[0041] For example, at each time of the second predetermined time, stress data corresponding to the time is applied at a different predetermined position. For example, at time t1, stress f is applied at a predetermined position a of the second finite element structural model. a1 and apply a stress f at a given location b of the second finite element structural model. b1 and when the time reaches the end time of the second predetermined time, outputting the result of the simulation prediction (i.e., the first prediction result), which is the crack state corresponding to the end time of the second predetermined time.

[0042] In this embodiment, the method for predicting fatigue damage to a structural member includes: obtaining a voiceprint signal of a target structural member within a first predetermined time period, a first finite element structural model corresponding to the target structural member, material characteristic parameters of the target structural member, and operating condition parameters within a second predetermined time period; first, judging the crack state of the target structural member at the current time based on the voiceprint signal; placing the crack in the first finite element structural model; and obtaining a second finite element structural model; the judgment result obtained based on the voiceprint signal has higher accuracy compared with the current artificial judgment form; further, loading the operating condition parameters for the second predetermined time period into the first finite element structural model and combining them with the material characteristic parameters to predict stress data corresponding to each predetermined position of the target structural member within the second predetermined time period, thereby obtaining the force-bearing state of each position of the target structural member within a certain period of time in the future; and The operating condition parameters for the second predetermined time period are determined based on historical synchronized data, so they have relatively high reference value and can make the predicted force-bearing condition closer to the actual condition. Finally, within the second predetermined time period, corresponding stress data are added to each predetermined position of the second finite element structural model to simulate and predict the fatigue damage condition of the target structural member within the second predetermined time period, thereby obtaining a first prediction result. Since the second finite element structural model is a simulation model that matches the current actual condition of the target structural member, the stress data added to the second finite element structural model is data that is closer to the actual operating condition. Therefore, the prediction result obtained by prediction based on both of them will necessarily have relatively high accuracy and truthfulness, be of relatively high reference value, and provide an important basis for the safe operation and maintenance of the generator set.

[0043] In this embodiment, a method for predicting fatigue damage of a structural member applicable to a detection device is provided, and FIG. 2 is a flowchart of the method for predicting fatigue damage of a structural member according to an embodiment of the present invention. As shown in FIG. 2, the flow includes the following steps S201 to S208.

[0044] In step S201, the voiceprint signal generated by the target structural member within a first predetermined time, the first finite element structural model corresponding to the target structural member, the material characteristic parameters of the target structural member, and the driving situation parameters within a second predetermined time are acquired. For details, refer to step S101 in the embodiment shown in Figure 1, and redundant explanations will be omitted here.

[0045] In step S202, noise reduction processing is performed on the voiceprint signal to obtain a noise-reduced signal.

[0046] Specifically, noise reduction processing methods include, but are not limited to, Fourier transform, wavelet threshold noise reduction, hard threshold noise reduction, soft threshold noise reduction, and wavelet semi-soft threshold noise reduction.

[0047] For example, in this embodiment, a method of noise reduction based on wavelet semi-soft thresholding is adopted to perform noise reduction processing on the voiceprint signals, and a noise-reduced signal corresponding to each voiceprint signal can be obtained.

[0048] The noise reduction formula by wavelet semi-soft threshold is as follows: JPEG2025188020000002.jpg25143 where x(t) is the noise-reduced signal at time t, w is the voiceprint signal at time t, and λ1 and λ2 are predetermined thresholds.

[0049] In step S203, an ensemble empirical mode decomposition method is used to decompose each noise-reduced signal into a residual component and at least two intrinsic mode components.

[0050] For example, Gaussian white noise is added to the noise-reduced signal, and the noise-reduced signal is decomposed once each time white noise is added. Take one noise-reduced signal x(t) as an example. After the first white noise addition, the noise-reduced signal is decomposed into n eigenmode components and one residual component, i.e., x(t)=IMF 11 +IMF 12+…IMF 1n +r 1n and after adding the second white noise, x(t)=IMF 21 +IMF 22 +…IMF 2n +r 2n After adding the j-th white noise, x(t)=IMF j1 +IMF j2 +…IMF jn +r jn and the IMF 11 , IMF 21 , …, IMF j1 The average value of IMF is IMF1, 21 , IMF 22 , …, IMF j2 Let the average value of IMF2 be IMF2, and similarly, IMF 1n , IMF 2n , …, IMF jn The average value of IMF n Let r 1n , r 2n , …, r jn The average value of r n So far, the ensemble empirical mode decomposition method divides the noise-reduced signal x(t) into IMF1, IMF2, ..., IMF n and r n Decomposed into IMF j are the eigenmode components, and r n is the residual component.

[0051] In step S204, the noise-reduced signal is reconstructed based on the eigenmode component and the residual component to obtain a reconstructed signal, and the crack state of the target structural member is determined based on the reconstructed signal.

[0052] Specifically, the above step S204 includes the following steps S2041 to S2043.

[0053] In step S2041, it is determined whether a noise component exists in at least two eigenmode components corresponding to the noise-reduced signal.

[0054] Specifically, a method for determining whether a noise component exists in at least two eigenmode components corresponding to a noise-reduced signal may include: selecting a dominant component from the at least two eigenmode components; determining whether other components besides the dominant component exist in the at least two eigenmode components; and, if so, determining the other components as noise components. Methods for reselecting a dominant component from the at least two eigenmode components include, but are not limited to, a direct selection method based on a frequency spectrum, a selection method based on a correlation coefficient, and a selection method based on kurtosis. This embodiment does not specifically limit the method for determining a noise component, and those skilled in the art may adopt other feasible forms.

[0055] In step S2042, if a noise component is present, the noise component is subjected to noise removal to obtain a dominant component after noise removal.

[0056] Specifically, noise reduction is performed on noise-dominant components using a fuzzy threshold method. The fuzzy threshold method is already a mature technology, and therefore, redundant explanations of its processing principles will be omitted here.

[0057] In step S2043, the residual components, the dominant components after noise removal, and the eigenmode components other than the noise components are added together to obtain a reconstructed signal.

[0058] In step S205, the crack state of the target structural member is determined based on the voiceprint signal, and the crack is placed in the first finite element structural model based on the crack state to obtain a second finite element structural model.

[0059] Specifically, the crack state of the target structural member is determined based on the voiceprint signal, and the crack is placed in the first finite element structural model based on the crack state to obtain a second finite element structural model.

[0060] In step S206, the operating status parameters within the second predetermined time are loaded into the first finite element structural model as boundary conditions, and combined with the material characteristic parameters to perform transient dynamic analysis to obtain stress data corresponding to each predetermined position of the target structural member within the second predetermined time. For details, see step S103 in the embodiment shown in Figure 1, and repeated explanations will be omitted here.

[0061] In step S207, the corresponding stress data is added to each predetermined position of the second finite element structural model within a second predetermined time, and a simulation prediction is performed for the fatigue damage status of the target structural member within the second predetermined time to obtain a first prediction result. For details, see step S104 of the embodiment shown in Figure 1, and repeated explanations will be omitted here.

[0062] In step 208, the remaining life of the target structural member is predicted based on the stress data respectively corresponding to each predetermined position within a second predetermined time period to obtain a second prediction result.

[0063] Specifically, the stress data corresponding to each predetermined position within a second predetermined time period is input, and the rainflow method is used to read the stress data corresponding to each time and calculate the cyclic stress-strain deviation. Based on the two-dimensional probability Miner criterion, with a reliability of 95%, the material SN curve and stress correction such as Goodman are combined to correct the material performance equation, and the cyclic spectrum block is calculated. Based on the theory of fatigue cumulative damage, the remaining life of the runner blade is predicted.

[0064] In the method for predicting fatigue damage of a structural member according to this embodiment, after obtaining a voiceprint signal generated within a first predetermined time, the voiceprint signal is further subjected to noise reduction, decomposition, and reconstruction to remove noise from the voiceprint signal, thereby making the crack state of the target structural member determined based on the reconstructed signal more accurate and further making the final prediction result more accurate.

[0065] In this embodiment, a method for predicting fatigue damage of a structural member applicable to a detection device is provided, and FIG. 3 is a flowchart of the method for predicting fatigue damage of a structural member according to an embodiment of the present invention. As shown in FIG. 3, the flow includes the following steps S301 to S304.

[0066] In step S301, the voiceprint signal generated by the target structural member within a first predetermined time, the first finite element structural model corresponding to the target structural member, the material characteristic parameters of the target structural member, and the driving situation parameters within a second predetermined time are acquired. For details, refer to step S101 in the embodiment shown in Figure 1, and redundant explanations will be omitted here.

[0067] In step S302, the crack state of the target structural member is determined based on the voiceprint signal, and the crack is placed in the first finite element structural model based on the crack state to obtain a second finite element structural model.

[0068] Specifically, the above step S302 includes the following steps S3021 to S3023.

[0069] In step S3021, feature parameters are extracted from the voiceprint signal.

[0070] For example, a voiceprint signal is taken as input, and feature parameters are extracted from the voiceprint signal using wavelet packet energy spectrum coefficients and singular value decomposition.

[0071] In step S3022, dimension reduction processing is performed on the feature parameters, and signal principal components are extracted from the dimension-reduced feature parameters.

[0072] For example, the feature parameters extracted in step S3021 are subjected to dimension reduction by compression using kernel principal component analysis, and the signal principal components are extracted from the feature parameters after dimension reduction based on the criterion that the cumulative contribution rate of the principal components exceeds 96%, thereby determining the dominant element components of the signal.

[0073] In step S3023, the signal main components are recognized to obtain the crack state of the target structural member.

[0074] For example, a particle swarm optimization-based least squares support vector machine (PSO-LS-SVM) mode recognition method can be used to recognize the dominant element components, and finally, the crack state corresponding to the reconstructed signal, that is, the crack state corresponding to the voiceprint signal at the last time within the first predetermined time, can be output as the crack state of the target structural member.

[0075] In step S303, the operating status parameters within the second predetermined time are loaded into the first finite element structural model as boundary conditions, and combined with the material characteristic parameters to perform transient dynamic analysis to obtain stress data corresponding to each predetermined position of the target structural member within the second predetermined time. For details, refer to step S103 in the embodiment shown in Figure 1, and repeated explanations will be omitted here.

[0076] In step S304, corresponding stress data is added to each predetermined position of the second finite element structural model within a second predetermined time, and a simulation prediction is performed for the fatigue damage status of the target structural member within the second predetermined time to obtain a first prediction result. For details, see step S104 in the embodiment shown in Figure 1, and repeated explanations will be omitted here.

[0077] This embodiment further provides an apparatus for predicting fatigue damage in a target structural member, which is used to realize the above-described embodiments and preferred embodiments. The duplicated explanation of what has already been explained will be omitted. The term "module" used below refers to a combination of software and / or hardware capable of implementing a predetermined function. While the apparatus described in the following embodiment is preferably implemented using software, it is also conceivable to implement it using hardware or a combination of software and hardware.

[0078] This embodiment provides a system for predicting fatigue damage of a structural member, and as shown in FIG. 4, the system includes: an acquisition module 401 for acquiring a voiceprint signal generated by a target structural member within a first predetermined time, a first finite element structural model corresponding to the target structural member, material characteristic parameters of the target structural member, and driving situation parameters within a second predetermined time, wherein the first predetermined time is set to a current time as an end time, the second predetermined time is set to a current time as a start time, and the driving situation parameters are determined based on history-synchronized driving situation parameters; a determining module 402 for determining a crack state of the target structural member based on the voiceprint signal, and locating a crack in the first finite element structural model based on the crack state to obtain a second finite element structural model; an analysis module 403 for loading the operating status parameters within a second predetermined time period as boundary conditions into a first finite element structural model, and combining the operating status parameters with material characteristic parameters to perform transient dynamic analysis, thereby obtaining stress data corresponding to each predetermined position of the target structural member within the second predetermined time period; and a first prediction module 404 for adding corresponding stress data at each predetermined position of the second finite element structural model within a second predetermined time period, and performing a simulation prediction of the fatigue damage status of the target structural member within the second predetermined time period to obtain a first prediction result.

[0079] In some alternative embodiments, after the acquisition module, the device may: The system further includes a noise reduction module for performing noise reduction processing on the voiceprint signal to obtain a noise-reduced signal; a decomposition module for decomposing each noise-reduced signal into a residual component and at least two eigenmode components using an ensemble empirical mode decomposition method; and a reconstruction module for reconstructing the noise-reduced signal based on the eigenmode component and the residual component to obtain a reconstructed signal, and determining the crack state of the target structural member based on the reconstructed signal.

[0080] In some alternative embodiments, the reconstruction module: The noise reduction signal includes a determination submodule for determining whether or not a noise component is present in at least two eigenmode components corresponding to the noise-reduced signal; a noise removal submodule for removing the noise component if a noise component is present and obtaining a dominant component after the noise removal; and a reconstruction submodule for adding the residual component, the dominant component after the noise removal, and the eigenmode components other than the noise component to obtain a reconstructed signal.

[0081] In some alternative embodiments, the determination module: The system includes a first extraction submodule for extracting feature parameters from the voiceprint signal, a second extraction submodule for performing dimension reduction processing on the feature parameters and extracting signal principal components from the feature parameters after dimension reduction, and a recognition submodule for recognizing the signal principal components and obtaining the crack state of the target structural member.

[0082] In some alternative embodiments, each predetermined location of the target structural member in the analysis module is: The method is determined by an acquisition submodule for acquiring operating condition parameters within a first predetermined time period, a calculation submodule for loading the operating condition parameters within the first predetermined time period into a first finite element structural model as boundary conditions and obtaining the stress distribution status of the target structural member within the first predetermined time period, and a determination submodule for determining a position to be studied of the target structural member based on the stress distribution status and determining the position to be studied as a predetermined position.

[0083] In some alternative embodiments, after the analysis module, the device: The system further includes a second prediction module for predicting the remaining life of the target structural member based on the stress data respectively corresponding to each predetermined position within a second predetermined time period to obtain a second prediction result.

[0084] Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments, and therefore, redundant descriptions will be omitted here.

[0085] In this embodiment, the apparatus for predicting fatigue damage in a structural member is represented in the form of a functional unit, where a unit is an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory executing one or more software or fixed programs, and / or other device capable of providing the above functionality.

[0086] An embodiment of the present invention further provides a computer device having the apparatus for predicting fatigue damage of a structural member shown in FIG. 4 above.

[0087] Referring to FIG. 5, FIG. 5 is a schematic diagram of a computer device according to an alternative embodiment of the present invention. As shown in FIG. 5, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the various components, including high-speed and low-speed interfaces. The various components are interconnected by different buses and may be implemented on a common motherboard, or may be implemented in other forms as needed. The processors can process instructions, including instructions stored in or on memory, executed within the computer device to display graphical information of a GUI on an external input / output device (e.g., a display device coupled to the interface). In some alternative embodiments, multiple processors and / or multiple buses, along with multiple memory, can be used as needed. Similarly, multiple computer devices (e.g., a server array, a group of blade servers, or a multiprocessor system) can be connected, each providing a portion of the required operations. In FIG. 5, one processor 10 is used as an example.

[0088] The processor 10 may be a central processing unit, a network processor, or a combination thereof. Here, the processor 10 may further include a hardware chip. The hardware chip may be an application specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable logic gate array, a universal array logic, or any combination thereof.

[0089] Here, the memory 20 stores instructions executable by the at least one processor 10 to cause the at least one processor 10 to perform the methods illustrated in the above embodiments.

[0090] The memory 20 may include a program storage area capable of storing an operating system and / or application programs necessary for at least one function, and a data storage area capable of storing data generated in response to use of the computer device. Furthermore, the memory 20 may include high-speed random access memory, and may further include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory located remotely from the processor 10, and such remote memory may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, a corporate intranet, a local area network, a mobile communication network, and combinations thereof.

[0091] Memory 20 may include volatile memory, such as random access memory, or non-volatile memory, such as flash memory, a hard disk, or a solid state drive, or memory 20 may include a combination of the above types of memory.

[0092] The computing device may include a communications interface 30 for communicating with the computing device and other devices or communications networks.

[0093]

[0013] Embodiments of the present invention further provide a computer-readable storage medium, and the methods according to the above-described embodiments of the present invention can be implemented in hardware, firmware, or as computer code recordable in a storage medium, or downloaded over a network, originally stored in a remote storage medium or a non-transitory machine-readable storage medium, and stored in a local storage medium, so that the methods described herein can be processed by software stored in a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Here, the storage medium may be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, solid-state drive, etc., and may also include a combination of the above types of memory. It should be understood that a computer, processor, microprocessor controller, or programmable hardware includes a storage assembly capable of storing or receiving software or computer code, and when the software or computer code is accessed and executed by the computer, processor, or hardware, the methods shown in the above-described embodiments are realized.

[0094] It should be noted that, in this specification, relational terms such as "first" and "second" are merely used to distinguish one entity or operation from another and do not necessarily require or imply the existence of any such actual relationship or order between those entities or operations. Furthermore, the terms "comprise," "include," or any other variation thereof are intended to cover a non-exclusive inclusion, whereby a process, method, article, or device comprising a set of elements not only includes those elements, but also includes other elements not expressly listed or elements inherent in such process, method, article, or device. Absent further limitations, an element qualified by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, article, or device that includes the element.

[0095] The foregoing are only specific embodiments of the present disclosure, enabling those skilled in the art to understand or realize the present disclosure. Various modifications to these examples will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other examples without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure is not intended to be limited to these examples herein, but is accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. a step of acquiring a voiceprint signal generated by a target structural member within a first predetermined time, a first finite element structural model corresponding to the target structural member, material characteristic parameters of the target structural member, and driving situation parameters within a second predetermined time, wherein the first predetermined time has a current time as an end time and the second predetermined time has a current time as a start time, and the driving situation parameters are determined based on history-synchronized driving situation parameters; determining a crack state of the target structural member based on the voiceprint signal, and placing a crack in the first finite element structural model based on the crack state to obtain a second finite element structural model; Loading the operating status parameters within the second predetermined time period into the first finite element structural model as boundary conditions, and combining them with the material characteristic parameters to perform transient dynamic analysis, thereby obtaining stress data corresponding to each predetermined position of the target structural member within the second predetermined time period; and applying corresponding stress data at each of the predetermined positions of the second finite element structural model within the second predetermined time, and performing a simulation prediction of the fatigue damage status of the target structural member within the second predetermined time to obtain a first prediction result.

2. After the step of acquiring a voiceprint signal generated by the target structural member within a first predetermined time, performing a noise reduction process on the voiceprint signal to obtain a noise-reduced signal; decomposing the noise-reduced signal into a residual component and at least two eigenmode components using an ensemble empirical mode decomposition method; 2. The method of claim 1, further comprising the steps of: reconstructing the noise-reduced signal based on the eigenmode component and the residual component to obtain a reconstructed signal; and determining a crack state of the target structural member based on the reconstructed signal.

3. The step of reconstructing the noise-reduced signal based on the eigenmode components and the residual components to obtain a reconstructed signal includes: determining whether a noise component is present in at least two of the eigenmode components corresponding to the noise-reduced signal; If the noise component exists, performing noise removal on the noise component to obtain a noise-removed dominant component; and adding the residual component, the dominant component after noise removal, and the eigenmode component other than the noise component to obtain the reconstructed signal.

4. The step of determining the crack state of the target structural member based on the voiceprint signal includes: extracting feature parameters from the voiceprint signal; performing a dimension reduction process on the feature parameters and extracting signal principal components from the dimension-reduced feature parameters; and recognizing the signal main component to obtain a crack state of the target structural member.

5. Each predetermined position of the target structural member is Acquire driving situation parameters within a first predetermined time period; Loading the operating condition parameters for the first predetermined time into the first finite element structural model as boundary conditions to obtain a stress distribution condition of the target structural member for the first predetermined time; 2. The method according to claim 1, wherein the position to be studied of the target structural member is determined based on the stress distribution state, and the position to be studied is determined as a predetermined position.

6. After the step of loading the operating situation parameters within the second predetermined time as boundary conditions into the first finite element structural model and combining them with the material characteristic parameters to perform transient dynamic analysis, and obtaining stress data corresponding to each predetermined position of the target structural member within the second predetermined time, 2. The method of claim 1, further comprising: predicting a remaining life of the target structural member based on stress data corresponding to each predetermined position within the second predetermined time period to obtain a second prediction result.

7. an acquisition module for acquiring a voiceprint signal generated by a target structural member within a first predetermined time, a first finite element structural model corresponding to the target structural member, material characteristic parameters of the target structural member, and driving situation parameters within a second predetermined time, wherein the first predetermined time has a current time as an end time, the second predetermined time has a current time as a start time, and the driving situation parameters are determined based on history-synchronized driving situation parameters; a determining module for determining a crack state of the target structural member based on the voiceprint signal, and for locating a crack in the first finite element structural model based on the crack state to obtain a second finite element structural model; an analysis module for loading the operating status parameters within the second predetermined time period as boundary conditions into the first finite element structural model, combining the operating status parameters with the material characteristic parameters, and performing a transient dynamic analysis to obtain stress data corresponding to each predetermined position of the target structural member within the second predetermined time period; and a first prediction module for applying corresponding stress data at each of the predetermined positions of the second finite element structural model within the second predetermined time period, performing a simulation prediction of the fatigue damage status of the target structural member within the second predetermined time period, and obtaining a first prediction result.

8. After the acquisition module, a noise reduction module for performing noise reduction processing on the voiceprint signal to obtain a noise-reduced signal; a decomposition module for decomposing the noise-reduced signal into a residual component and at least two eigenmode components using an ensemble empirical mode decomposition method; The apparatus of claim 7, further comprising a reconstruction module for reconstructing the noise-reduced signal based on the eigenmode component and the residual component to obtain a reconstructed signal, and determining a crack state of the target structural member based on the reconstructed signal.

9. The reconstruction module a determination sub-module for determining whether a noise component exists in the at least two eigenmode components corresponding to the noise-reduced signal; a noise removal module for removing the noise component if the noise component exists and obtaining a dominant component after the noise removal; and a reconstruction sub-module for adding the residual component, the dominant component after noise removal, and the eigenmode component other than the noise component to obtain the reconstructed signal.

10. The determination module: a first extraction sub-module for extracting feature parameters from the voiceprint signal; a second extraction submodule for performing dimension reduction processing on the feature parameters and extracting signal principal components from the dimension-reduced feature parameters; and a recognition sub-module for recognizing the signal main component and obtaining a crack state of the target structural member.

11. Each predetermined position of the target structural member in the analysis module is an acquisition submodule for acquiring driving situation parameters within a first predetermined time; a calculation sub-module for loading the operating status parameters within the first predetermined time into the first finite element structural model as boundary conditions to obtain a stress distribution status of the target structural member within the first predetermined time; 8. The apparatus according to claim 7, further comprising a determination submodule for determining a position to be studied of the target structural member based on the stress distribution situation, and determining the position to be studied as a predetermined position.

12. After the analysis module, 8. The apparatus of claim 7, further comprising a second prediction module for predicting a remaining life of the target structural member based on stress data corresponding to each predetermined position within the second predetermined time period to obtain a second prediction result.

13. A computer device comprising: a memory and a processor, the memory and the processor being connected to each other so as to be able to communicate with each other; computer instructions being stored in the memory; and the processor executing the computer instructions to perform the method for predicting fatigue damage in a structural member according to any one of claims 1 to 6.

14. A computer-readable storage medium storing computer instructions for causing a computer to execute the method for predicting fatigue damage in a structural member according to any one of claims 1 to 6.

Images