Turbofan engine compressor blade crack multi-mode monitoring system

Abstract

The invention discloses a turbofan engine compressor blade crack multi-modal monitoring system, and belongs to the field of multi-modal monitoring, and the system comprises a multi-modal acquisition module which obtains the multi-modal data of each compressor blade, carries out the dynamic filtering denoising and temperature compensation preprocessing, completes the time axis alignment, and carries out the data reconstruction and feature extraction; and constructing a multi-modal feature vector. And the triggering module carries out preliminary abnormality diagnosis according to the judgment result, judges whether the blade is abnormal or not through a key feature threshold, determines whether to trigger depth evaluation or not by combining a periodic threshold, identifies a potential crack blade and generates an excitation decision. And when an excitation decision is executed, the intelligent diagnosis module monitors the strain gradient change to perform excitation monitoring, compares the vibration characteristics of a potential crack blade with the vibration characteristics of a contrast blade, judges whether a crack exists or not, and realizes multi-stage early warning, so as to solve the problem that the crack of the turbofan engine compressor blade is difficult to quickly detect in real time in the prior art. And safe and stable operation of the engine is guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of multimodal monitoring, and more specifically to a multimodal monitoring system for cracks in compressor blades of turbofan engines. Background Art

[0002] The existing crack monitoring technology system for compressor blades of turbofan engines is mainly based on offline detection methods. Its core limitation lies in the time decoupling between the detection behavior and the operating state of the equipment. Traditional non-destructive testing technologies such as ultrasonic phased array and array eddy current need to be implemented in the shutdown state. The detection process is accompanied by complex physical contact and mechanical disassembly, resulting in a significant time delay in the monitoring behavior. This detection mode is difficult to adapt to the operating characteristics of high load and long flight time of the engine, and cannot effectively capture the dynamic evolution process from crack initiation to propagation. In particular, it lacks the ability to provide real-time early warning for sudden crack propagation events that may occur during the maintenance interval.

[0003] The current technology system in the field of online monitoring has multi-dimensional technical bottlenecks. Although acoustic emission detection can sense the stress wave signals of crack propagation, limited by the noise interference and signal attenuation in the complex flow field environment, its spatial positioning accuracy is difficult to meet the engineering requirements. Although fiber Bragg grating sensing technology can achieve real-time monitoring of the strain field, the implantable installation method of its sensor array significantly increases the structural complexity and maintenance cost. And there is also the problem that neither can provide early warning for blade cracks and locate the position of the blade with cracks. Therefore, in order to overcome these limitations, the present invention proposes a multimodal monitoring system for cracks in compressor blades of turbofan engines. Summary of the Invention

[0004] Aiming at the deficiencies of the existing technology, the purpose of the present invention is to provide a multimodal monitoring system for cracks in compressor blades of turbofan engines, which breaks through the technical bottlenecks of traditional crack monitoring methods for compressor blades of turbofan engines in non-stationary signal processing, dynamic condition adaptation and early crack identification, and realizes real-time dynamic monitoring of cracks in compressor blades of turbofan engines and weak feature capture.

[0005] To achieve the above purpose, the present invention provides the following technical solutions: A multimodal monitoring system for cracks in compressor blades of turbofan engines, comprising: Obtain multimodal data of the compressor blade and perform time parameter alignment to construct a multimodal feature vector; Based on the multimodal feature vector, conduct a preliminary abnormal diagnosis on the compressor blade to determine the abnormal blade and mark it, and trigger a depth evaluation for the abnormal blade by constructing a joint trigger rule, construct a trend eigenvalue sequence, and identify potential crack blades by measuring the difference degree of the trend eigenvalues of the compressor blade, and generate an excitation decision for the potential crack blades; During the execution of the excitation decision, excitation monitoring is performed by monitoring the change in the strain gradient characteristics of the potential cracked blade, and a control blade of the potential cracked blade is selected. By comparing and analyzing the vibration data characteristics of the control blade and the potential cracked blade, it is determined whether there is a crack in the potential cracked blade; It also includes performing multi-level early warnings according to the results of the comparative analysis.

[0006] Specifically, the specific steps of the preliminary anomaly diagnosis include: Receiving the multi-modal feature vectors of the compressor blades, selecting key features through recursive feature elimination and combining domain prior knowledge; at the same time, obtaining the operating condition parameters of the compressor of the turbofan engine and constructing the operating condition feature vectors; Based on Gaussian process regression, a non-linear mapping relationship between the operating condition parameters and the key feature thresholds is established, and according to the operating condition feature vectors, the key feature threshold ranges are adjusted in real time and dynamically; Comparing the real-time key feature values with the corresponding key feature threshold ranges, and judging whether all the real-time key feature values of the compressor blades are within the corresponding key feature threshold ranges. If so, it is determined that the compressor blades are in a normal state; otherwise, it is determined that the compressor blades are in an abnormal state and they are marked as abnormal blades.

[0007] Specifically, the specific steps of the preliminary anomaly diagnosis also include: If it is determined that the compressor blades are in an abnormal state, a monitoring threshold is configured to measure the time length for monitoring the compressor blades in an abnormal state. The joint trigger rule is adopted, and according to the number of key features of the abnormal blades exceeding the key feature threshold ranges within the monitoring threshold and the degree of exceeding of the key feature values, it is judged whether to trigger a depth evaluation; A periodic threshold is configured to measure the periodic interval for each compressor blade to trigger a depth evaluation. If it is determined that the compressor blades are in a normal state, it is judged whether the time interval from the last depth evaluation of the compressor blades is equal to the periodic threshold. If so, a depth evaluation is triggered.

[0008] Specifically, the specific steps of the depth evaluation include: Based on the periodic threshold, an evaluation period is set according to the sliding time window management strategy. The sliding time window management strategy is used to dynamically adjust the data acquisition window. Taking the current moment as the reference, the multi-modal feature vectors of each compressor blade within the evaluation period are obtained; For each compressor blade, the values of the key features within the evaluation period are arranged in chronological order to construct the key feature value sequence of the compressor blade; Trend features are extracted from the key feature value sequences of the compressor blades respectively. The trend features include: monotonic features, non-linear trend features, and periodic features. Each trend feature value is calculated by non-parametric statistical methods, and the trend feature value sequences of each compressor blade are constructed.

[0009] Specifically, the specific steps of in-depth evaluation also include: Use a clustering algorithm to cluster the trend feature value sequences of the compressor blades, identify the groups of compressor blades with trend deviations, and mark them as deviated blades; Exclude the trend feature value sequences marked as abnormal blades and deviated blades respectively, and calculate the mean values of the trend feature value sequences of the remaining compressor blades respectively to construct a baseline trend feature value sequence; Use distance metrics to calculate the degree of difference between the trend feature value sequences of each deviated blade and abnormal blade and the baseline trend feature value sequence respectively; Configure a difference threshold to judge whether the degree of difference between the trend feature value sequence of the deviated blade or abnormal blade and the baseline trend feature value sequence is greater than the difference threshold. If it is greater, mark it as a potentially cracked blade.

[0010] Specifically, the specific steps of generating the excitation decision for potentially cracked blades include: Obtain the circumferential position coordinates and their degree of difference of the potentially cracked blades in real time, and dynamically adjust the laser pulse energy in combination with the degree of difference and the basic energy of the laser pulse; Obtain the natural frequency of the potentially cracked blades according to the real-time rotational speed, and set the laser excitation frequency band centered on the natural frequency; According to the position of the potentially cracked blade in the compressor blade group, select a transmitter, and set the azimuth angle and elevation angle of the transmitter according to the relative position between the transmitter coordinates and the circumferential position coordinates of the potentially cracked blade; Emit the laser beam to the surface of the potentially cracked blade through the transmitter. When there are multiple potentially cracked blades in a group of compressor blades, parallel excitation is implemented, and the laser beam is emitted in a time-sharing manner according to the circumferential order of the potentially cracked blades.

[0011] Specifically, the specific steps of determining whether there are cracks in the potentially cracked blades include: Extract the stress data from the multi-modal data of the potentially cracked blades, and perform stress diagnosis on the blade group where the potentially cracked blades are located, including: Extract the strain gradient feature value according to the stress data, configure a strain threshold. If the strain gradient feature value of the potentially cracked blade is greater than the strain threshold, stop the laser excitation and trigger an excitation warning; Based on the principle of spatial symmetry, determine the position coordinates of the potentially cracked blade in the compressor blade group, and select a reference blade according to the spatial layout structure of the compressor blades; Extract the vibration data features in the multimodal data of the potentially cracked blade and the control blade. The vibration data features include vibration order amplitude and time-frequency spectrum diagram features. Configure the order amplitude threshold. For the vibration order amplitude, calculate the relative deviation of the amplitudes of the potentially cracked blade and the control blade at the same order. If the relative deviation of the amplitudes of the potentially cracked blade and the control blade at the same order exceeds the order amplitude threshold, it is determined that the potentially cracked blade has a crack.

[0012] Specifically, the specific steps to determine whether the potentially cracked blade has a crack also include: Configure the principal component threshold , perform empirical mode decomposition on the vibration data of the potentially cracked blade and the control blade, decompose the vibration data into multiple intrinsic mode functions, and select the first intrinsic mode functions with the largest energy proportion for Hilbert transform to obtain the time-frequency spectrum diagram; Perform characteristic frequency identification on the time-frequency spectrum diagram through peak detection. On the time-frequency spectrum diagram, set the frequency bandwidth around the characteristic frequency, and calculate the energy density of the characteristic frequency through integration; Set the energy deviation threshold, calculate the relative deviation of the energy densities of the characteristic frequencies of the potentially cracked blade and the control blade. If it is greater than the energy deviation threshold, it is determined that the potentially cracked blade has a crack; When it is determined that the potentially cracked blade has a crack, stop exciting the potentially cracked blade and issue a first-level blade warning; Configure the excitation duration, monitor the continuous duration of exciting the potentially cracked blade. If it is equal to the excitation duration and it is not determined that the potentially cracked blade has a crack, issue a second-level blade warning.

[0013] Specifically, the specific steps for constructing the multimodal feature vector include: Obtain the real-time speed of the compressor, which is used to resample the vibration data in the multimodal data to generate vibration data with equal angular intervals; and define the target order according to the number of compressor blades in each group; Delimit the resonance frequency band, perform filtering processing on the resampled vibration data, obtain the filtered vibration data and perform Hilbert transform to generate the orthogonal data of the vibration data, calculate the envelope data, and perform Fourier transform on the envelope data to obtain the envelope spectrum; Perform windowing processing on the resampled vibration data and then perform Fourier transform to obtain the spectrum of the resampled vibration data; According to the target order, map the frequency axes of the spectrum of the resampled vibration data and the envelope spectrum to the order axis, and use the interpolation algorithm to generate the corresponding order spectrum; Extract features from the order spectra respectively, and combine with the time-domain features of the multimodal data to splice and form the multimodal feature vector of each compressor blade.

[0014] Specifically, the combined trigger rules include: Within the monitoring threshold time range, count the number of key features that exceed the corresponding key feature threshold range. When the number of key features that exceed the key feature threshold range exceeds the preset quantity threshold, the first rule is satisfied; Calculate the relative deviation degree of each key feature value that exceeds the key feature threshold range with respect to its corresponding key feature threshold range. When there is a relative deviation degree that exceeds the preset deviation degree threshold, the second rule is satisfied; When both the first rule and the second rule are satisfied, that is, when the number of key features that exceed the key feature threshold range reaches the preset quantity threshold and the relative deviation degree of at least one key feature reaches the preset deviation degree threshold, it is determined to trigger a depth evaluation.

[0015] Advantages of the present invention: The present invention realizes the accurate extraction and spatio-temporal alignment of multi-dimensional features such as blade vibration and stress through multi-modal data fusion technology, significantly improving the accuracy of crack feature recognition; the dynamic threshold adaptation mechanism combines Gaussian process regression and combined trigger rules to effectively reduce the false alarm rate under complex working conditions; the trend feature analysis based on a sliding time window and the baseline comparison algorithm successfully capture the weak change trend of early cracks; the dynamic optimization strategy of laser excitation parameters and the parallel time-sharing emission technology greatly improve the crack verification efficiency while ensuring excitation safety; the multi-level early warning mechanism realizes fault classification response and closed-loop management through strain gradient monitoring and vibration data comparison analysis. A full-process monitoring system covering real-time monitoring, early crack warning, accurate positioning and verification is constructed, with real-time dynamic adaptation ability, weak feature recognition ability and multi-target collaborative processing ability, ultimately significantly improving the safety and reliability of the operation of the compressor of the turbofan engine. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a functional schematic diagram of the multi-modal monitoring system for cracks in the compressor blades of the turbofan engine of the present invention; Figure 2 It is a flowchart of the specific steps of the preliminary anomaly diagnosis of the present invention; Figure 3 It is a flowchart of the specific steps of the depth evaluation of the present invention; Figure 4 It is a flowchart of the specific steps of the stress diagnosis of the present invention; Figure 5 It is a flowchart of the specific steps of determining whether there is a crack in the potentially cracked blade of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] Please refer to Figure 1, this embodiment introduces a multi-modal monitoring system for cracks in the compressor blades of a turbofan engine, including: a multi-modal acquisition module, a trigger module, and an intelligent diagnosis module; The multi-modal acquisition module is used to obtain the multi-modal data of each compressor blade through distributed sensors, including vibration data and stress data, and perform data preprocessing on the acquired multi-modal data, including: dynamically adjusting the filter parameters according to the engine speed, denoising the vibration data to filter out non-blade vibration interference, performing temperature compensation on the stress data, and after time-axis alignment of the preprocessed multi-modal data, extracting features through multi-modal data reconstruction to construct the multi-modal feature vector of each compressor blade.

[0018] In this embodiment, based on the stress distribution, aerodynamic load characteristics, and crack propagation mechanism of the compressor blade, a distributed design is adopted. Two orthogonal strain sensors are arranged at the root tenon of each compressor blade to monitor the circumferential and radial strain data of the compressor blade under centrifugal load, and MEMS accelerometers are evenly and symmetrically installed on the inner wall circumference of the compressor casing to monitor the full-circumferential vibration data; according to the operating speed and vibration characteristics of the compressor blade, the acquisition frequency of the vibration data is set. For example, the acquisition frequency is more than twice the highest vibration frequency of the blade to avoid signal aliasing. The acquisition frequency of the stress data is determined according to the rate of strain change. Since the strain change is relatively slow, the acquisition frequency can be reduced, but it should be ensured that the dynamic change of the stress can be captured. A band-pass filter is used to filter out the non-blade vibration interference of low frequency and high frequency. The mapping relationship between the engine speed and the filter parameters is established, and through experiments and theoretical analysis, the variation law of the vibration frequency of the blade at different engine speeds is determined. For example, when the engine speed increases, the vibration frequency of the blade also increases accordingly. At this time, the passband range of the filter is adjusted to enable it to better capture the vibration signal of the blade. During the data acquisition process, the engine speed is monitored in real time, and the filter parameters are dynamically adjusted according to the mapping relationship. A temperature sensor is arranged near the strain sensor to monitor the temperature change at the position where the strain sensor is located in real time. The temperature sensor should be in close contact with the strain sensor to ensure that the temperature at the same position can be accurately measured. The output data of the strain sensor at different temperatures are collected through experiments, and the relationship model between the temperature and the strain measurement value is established. Methods such as polynomial fitting and neural network are used to establish the model. During the data acquisition process, the measurement value of the temperature sensor is obtained in real time, and the acquired stress data is compensated in real time according to the temperature compensation model. After time-axis alignment of the vibration data and the stress data using the interpolation algorithm, the features of the vibration data and the stress data are extracted, including time-domain features, frequency features, joint features, and rate-of-change features, and a high-dimensional multi-modal feature vector is formed by splicing.

[0019] Preferably, the specific steps for constructing the multi-modal feature vector of each compressor blade include: Obtain the real-time rotational speed of the compressor, and resample the vibration data based on the real-time rotational speed to generate vibration data with equal angular intervals, that is: according to the real-time rotational speed, use the interpolation algorithm to resample the vibration signal into a sequence with equal angular intervals as the resampled vibration data to eliminate the spectral ambiguity caused by rotational speed fluctuations; the phase information of the resampled vibration data is completely retained, providing an accurate angular domain signal basis for subsequent order analysis to ensure the correspondence between frequency components and mechanical structure characteristics.

[0020] Define the target order according to the number of compressor blades in each group. For example, if the number of blades in a certain stage of the compressor is 50, then define the target order as 50 to focus on analyzing the 50th order and its harmonics; directly associate the vibration energy distribution with the characteristics of the blade rotating machinery to improve the recognition of crack-related features.

[0021] Combine the historical vibration data, delimit the high-frequency resonance frequency band, use a band-pass filter to filter the resampled vibration data, obtain the filtered vibration data and perform Hilbert transform to generate its orthogonal data to calculate the envelope data, perform Fourier transform on the envelope data to obtain the envelope spectrum; decouple the low-frequency modulation information in the high-frequency carrier signal, where the low-frequency modulation information includes the crack impact period, generate the envelope spectrum, and highlight the modulation characteristics related to the blade rotation frequency.

[0022] Perform windowing processing on the resampled vibration data to reduce spectral leakage, perform discrete Fourier transform to obtain the spectrum of the resampled vibration data; provide an accurate spectral basis for order mapping.

[0023] According to the target order, map the frequency axes of the spectrum of the resampled vibration data and the envelope spectrum to the order axis, and use the interpolation algorithm to generate the corresponding order spectrum to focus on the blade resonance characteristics; Extract features from the order spectra corresponding to the order spectra of the resampled vibration data and the envelope spectrum respectively, including amplitude features and phase features; combine the time-domain features of the vibration data including peak value and kurtosis with the time-domain features of the stress data, that is, the strain gradient feature, and splice them to form a multi-modal feature vector for each compressor blade.

[0024] The trigger module is used to perform preliminary abnormal diagnosis on each compressor blade during the operation of the compressor of the turbofan engine based on the obtained multi-modal feature vector, quickly judge whether there is an abnormal phenomenon, and combine the periodic threshold to judge whether to trigger a depth evaluation to identify potential cracked blades and generate a decision to stimulate potential cracked blades; Please refer to Figure 2 , preferably, the specific steps of the preliminary abnormal diagnosis include: Receive the multimodal data and multimodal feature vectors of each compressor blade in real time. Select the key features of the multimodal feature vectors through recursive feature elimination and combined with domain prior knowledge, including: vibration order amplitude, envelope spectrum sideband energy ratio, stress gradient, strain rate, etc., for rapid diagnosis of each compressor blade; construct a linear model based on the historical multimodal feature vectors of each compressor blade, using the multimodal feature vectors as input and whether the blade is abnormal as output. Then, calculate the weight of each feature, gradually remove the feature with the smallest weight, retrain the model, and repeat this process until the remaining number of features reaches the preset value to determine the features closely related to the crack propagation of the blade, including vibration order amplitude, envelope spectrum sideband energy ratio, stress gradient, etc. Merge and adjust these features with the features selected by recursive feature elimination to finally determine the key features.

[0025] At the same time, obtain the operating condition parameters of the current compressor of the turbofan engine, including real-time speed, load, and temperature, and construct an operating condition feature vector; collect historical operating condition parameters and corresponding key feature values, divide them into a training set and a test set, and establish a nonlinear mapping relationship between the operating condition parameters and the key feature thresholds based on Gaussian process regression to dynamically adjust the range of each key feature threshold; Compare the real-time key feature values with the corresponding key feature threshold ranges to determine whether all real-time key feature values are within the corresponding key feature threshold ranges. If so, it is determined that the compressor blade is in a normal state; otherwise, it is determined that the compressor blade is in an abnormal state, and this blade is marked as an abnormal blade; If it is determined that the compressor blade is in an abnormal state, configure a monitoring threshold to measure the length of time for monitoring the compressor blade in an abnormal state, and adopt a joint trigger rule to determine whether to trigger a deep evaluation based on the number of key features of the abnormal blade exceeding the key feature threshold range and the degree of exceeding of the key feature values within the monitoring threshold; Configure a periodic threshold to measure the periodic interval for each compressor blade to trigger a deep evaluation. If it is determined that all blades in the current blade group are in a normal state, determine whether the time interval since the last deep evaluation of the compressor blade is equal to the periodic threshold. If it is equal, trigger a deep evaluation; otherwise, do nothing.

[0026] The joint trigger rule includes: Within the monitoring threshold time range, count the number of key features exceeding the corresponding key feature threshold range. When the number of key features exceeding the key feature threshold range exceeds the preset number threshold, the first rule is satisfied. This condition is set to ensure that multiple key features are abnormal at the same time because the abnormality of a single feature may be caused by accidental factors, and the abnormality of multiple features at the same time can better reflect the existence of substantial problems in the blade group.

[0027] Meanwhile, calculate the relative deviation degree of each key feature value beyond the critical feature threshold range with respect to its corresponding critical feature threshold range. When there is a relative deviation degree exceeding the preset deviation degree threshold, the second rule is satisfied. This condition emphasizes that at least one key feature has shown a significant abnormal change, which may be an important signal indicating the development of the blade group fault to a certain extent.

[0028] When both the first rule and the second rule are satisfied, that is, the number of key features beyond the critical feature threshold range reaches the preset quantity threshold and the relative deviation degree of at least one key feature reaches the preset deviation degree threshold, it is determined that the depth evaluation is triggered. This combined triggering method comprehensively considers the quantity and degree of feature anomalies, can more accurately capture the abnormal state of the blade group, avoid false triggering caused by accidental fluctuations of individual features, and can also timely detect potential serious fault hazards, providing a reliable basis for further diagnosis and processing. If the combined triggering rule is satisfied, the depth evaluation is triggered; otherwise, the multi-modal feature vectors of each compressor blade are evaluated in real time.

[0029] Please refer to Figure 3 , preferably, the specific steps of the depth evaluation include: Based on the periodic threshold, set the evaluation period according to the sliding time window management strategy. The sliding time window management strategy is used to comprehensively consider the data of multiple historical periods before the current moment by dynamically adjusting the data window, so as to more comprehensively capture the change trend of the compressor blade performance. The periodic threshold represents a basic time unit, and the evaluation period will be set based on this. That is, the time length covered by the evaluation period is a positive integer multiple of the periodic threshold. Taking the current moment as the benchmark, trace back the time range covered by the evaluation period forward to form a dynamic data window; Obtain the multi-modal feature vectors of each compressor blade within the evaluation period. For each compressor blade, arrange the values of the key features within the evaluation period in chronological order to construct the key feature value sequence of the compressor blade; Extract the trend features from the key feature value sequences of the compressor blades respectively. The trend features include: monotonic features, non-linear trend features and periodic features. Calculate the trend feature values through non-parametric statistical methods including the Mann-Kendall test statistic Z value, Sen's slope, Cox-Stuart trend strength, autocorrelation analysis, etc.; combine the monotonic feature values calculated by non-parametric statistical methods for each compressor blade, such as the Z value and Sen's slope, the non-linear trend feature values including the p value of the Cox-Stuart test, the polynomial fitting coefficient, and the periodic feature values including the lag order corresponding to the peak value of the autocorrelation function to construct the trend feature value sequence of each compressor blade.

[0030] Construct a sequence of trend eigenvalue for each compressor blade, and use a clustering algorithm to cluster the sequence of trend eigenvalue of the compressor blades, identify the group of compressor blades with trend deviation, and label them as deviated blades. The density-based spatial clustering algorithm can be used to cluster the sequence of trend eigenvalue of the compressor blades, and the blades are divided into core points, boundary points and noise points. By analyzing the clustering results, the blades in different clusters are identified as groups of blades with different trend characteristics, and the group of blades deviating from the normal cluster is labeled as deviated blades. For example, if there is a cluster that contains most of the blades, while a few other blades form small clusters or are noise points alone, then these few blades can be labeled as deviated blades.

[0031] According to the clustering results, identify and remove the sequences of trend eigenvalue labeled as abnormal blades and deviated blades. Conduct statistical analysis on the trend eigenvalue of the remaining normal blades, calculate the mean value of each trend eigenvalue respectively, and construct a baseline sequence of trend eigenvalue. During the calculation of the mean value, the weighted average method can be used to assign different weights to different blades according to the distance from the blade to the clustering center to improve the accuracy of the baseline.

[0032] Use distance metric as a method to measure the degree of difference between the sequence of trend eigenvalue of each deviated blade and abnormal blade and the baseline sequence of trend eigenvalue, and calculate the degree of difference between the sequence of trend eigenvalue of each deviated blade and abnormal blade and the baseline sequence of trend eigenvalue respectively; Through the analysis of historical data and in combination with actual fault cases, configure a difference threshold, and judge whether the degree of difference between the sequence of trend eigenvalue of the deviated blade or abnormal blade and the baseline sequence of trend eigenvalue is greater than the difference threshold. If it is greater, label it as a potentially cracked blade; otherwise, do not perform any processing.

[0033] Preferably, the specific steps for generating an excitation decision include: Obtain the circumferential position coordinates and its degree of difference of the potentially cracked blade in real time, and dynamically adjust the laser pulse energy according to the degree of difference to enhance the crack response signal, that is: ; Among them, is the laser pulse energy after dynamic adjustment, is the basic laser pulse energy, which is determined by the material damage threshold experiment, is the energy gain coefficient, and the value range is 0.5 ≤ ≤ 1.5 to avoid overload, is the maximum allowable difference threshold, is the degree of difference of potential crack blades; when the degree of difference is large, it indicates that there may be relatively serious cracks in the blades. Appropriately increase the energy gain coefficient and the laser pulse energy to enhance the crack response signal; when the degree of difference is small, reduce the energy gain coefficient and the laser pulse energy to reduce unnecessary influence on the compressor blades while ensuring the detection effect. At the same time, set the maximum allowable difference threshold. When the degree of difference exceeds this threshold, adopt special energy adjustment strategies or send warning signals.

[0034] Monitor the engine speed in real time. According to the structural parameters of the blades, including the number of blades, mass distribution, stiffness, and real-time speed information, calculate the natural frequency of the potential crack blades. With the natural frequency as the center, set the laser excitation frequency band; the range of this frequency band is determined according to the dynamic characteristics of the blades and the detection requirements, and can be set to 1.2 - 2.0 times the natural frequency. Emit laser pulses within the laser excitation frequency band, which can better stimulate the resonance effect of the blades, enhance the crack response signal, and improve the sensitivity of crack detection.

[0035] According to the position of the potential crack blades in the compressor blade group, select the emitters. The emitters are evenly arranged on the inner wall of the compressor casing in a circumferentially uniform distribution. Each group of emitters contains 3 orthogonal direction units. This layout design can cover all areas within the compressor to ensure effective detection of all blades. At the same time, a certain gap is maintained between the emitter and the blade tip. This gap is accurately calculated and experimentally verified to ensure that the laser beam can accurately irradiate the blade surface and avoid collision or interference between the emitter and the blade. Calculate the relative position relationship between the two based on the coordinates of the emitter and the circumferential position coordinates of the potential crack blades, and adjust the azimuth angle and pitch angle of the emitter in real time to ensure that the laser beam accurately aligns with the surface of the potential crack blades and improve the excitation effect.

[0036] Emit the laser beam to the surface of the potential crack blades through the emitter. When there are multiple potential crack blades in a group of compressor blades, perform parallel excitation. According to the circumferential order of the potential crack blades, emit the laser beam at different times, and set a certain time interval between laser pulses to avoid signal interference between different blades and ensure the accuracy of the detection results.

[0037] The intelligent diagnosis module is used to monitor the excitation process through stress diagnosis when making excitation decisions, select reference blades, and judge whether there are cracks in the potential crack blades by comparing the characteristics of the vibration data of the potential crack blades and the reference blades, and perform multi-level early warning; Preferably, the specific steps for judging whether there are cracks in the potential crack blades include: Please refer to Figure 4, obtain the stress data in the multi-modal data of the potentially cracked blade. For the blade group where the potentially cracked blade is located, perform stress diagnosis, including: Extract the strain gradient eigenvalue. The strain gradient eigenvalue reflects the change of strain in the blade under stress. Its calculation method is based on the strain data collected by the orthogonal strain sensors at the blade root tenon. By calculating the ratio of the difference in strain at adjacent positions to the distance, the strain gradient is obtained.

[0038] Configure the strain threshold. The determination of the strain threshold is based on the statistical analysis of the strain gradient data of a large number of normal blades under the same excitation conditions, and comprehensively considers factors such as the material properties of the compressor blade, the designed stress-bearing range, and the actual operating conditions. Conduct stress diagnosis and judgment. If the strain gradient eigenvalue of the potentially cracked blade is greater than the strain threshold, immediately stop the laser excitation. When the strain gradient eigenvalue is greater than the strain threshold, it indicates that the mechanical response of the blade during the laser excitation is abnormal, and there may be situations such as an accelerated crack propagation rate and excessive damage to the blade structure. At this time, stopping the excitation in time can effectively avoid further irreparable damage to the compressor blade. At the same time, trigger an excitation warning and send an alarm signal to the operator to remind them to conduct a more in-depth inspection and evaluation of the compressor blade. The alarm signal can be presented in various ways such as an audible and visual alarm device and an information pop-up window of the engine monitoring system to ensure the safe and stable operation of the engine.

[0039] Please refer to Figure 5 , based on the principle of spatial symmetry, determine the position coordinates of the potentially cracked blade in the compressor blade group. According to the spatial layout structure of the compressor blade, identify the blade corresponding to its spatially symmetric position, and use this blade as the initially selected reference blade.

[0040] If the reference blade at the spatially symmetric position of the potentially cracked blade is still a potentially cracked blade, then re-select the next blade at the spatially symmetric position of the potentially cracked blade as the reference blade in the priority order from top to bottom and from left to right. The directions of "up", "down", "left", and "right" here are determined based on the fixed coordinate system of the compressor blade group. For example, the compressor axial direction is the up-down direction, and the blade circumferential direction is the left-right direction. When selecting, check the adjacent blades in sequence until a non-potentially cracked blade is found as the reference blade.

[0041] Extract the vibration data characteristics in the multi-modal data of the potentially cracked blade and the reference blade. The vibration data characteristics include vibration order amplitude and time-frequency spectrum diagram characteristics; different vibration data characteristics have different sensitivities to blade faults. For example, changes in vibration order amplitude may be related to blade resonance or imbalance, and the time-frequency spectrum diagram characteristics can reveal the change law of the vibration signal in time and frequency, providing rich information for subsequent crack judgment.

[0042] Compare and analyze the vibration data characteristics of potentially cracked blades and control blades, including: Configure the order amplitude threshold. For the vibration order amplitude, calculate the relative deviation of the amplitudes of the potentially cracked blade and the control blade at the same order. If the relative deviation of the amplitudes of the potentially cracked blade and the control blade at the same order exceeds the order amplitude threshold, it is determined that the potentially cracked blade has a crack; when the relative deviation exceeds the order amplitude threshold, it indicates that the vibration amplitude of the potentially cracked blade has an abnormal change, and this change may be caused by structural damage inside the blade, thus providing a strong basis for the determination of cracks and improving the accuracy of crack diagnosis.

[0043] Configure the principal component threshold , perform empirical mode decomposition on the vibration data of the potentially cracked blade and the control blade, decompose the vibration data into multiple intrinsic mode functions, and select the first intrinsic mode functions with the largest energy proportion for Hilbert transform to obtain the time-frequency spectrogram, which reflects the change of the energy of the data with time and frequency, highlights the essential characteristics of the signal, makes the analysis of the blade vibration state more in-depth and detailed, and helps to discover potential fault information; Identify the characteristic frequency of the time-frequency spectrogram through peak detection. On the time-frequency spectrogram, set the frequency bandwidth around the characteristic frequency, and calculate the energy density of the characteristic frequency through integration to further mine the useful information in the vibration signal. The change of the characteristic frequency and its energy density can reflect the change of the dynamic characteristics of the blade. When the blade has a crack, the characteristic frequency and energy distribution of its vibration will change, and it can more accurately judge whether the blade has a crack and the degree of crack development.

[0044] Set the energy deviation threshold, calculate the relative deviation of the energy density of the characteristic frequency of the potentially cracked blade and the control blade. If it is greater than the energy deviation threshold, it indicates that there is a significant difference in the energy density of the characteristic frequency of the potentially cracked blade compared with the control blade, and it is determined that the potentially cracked blade has a crack; When it is determined that the potentially cracked blade has a crack, stop exciting the potentially cracked blade, and issue a first-level blade warning. The first-level blade warning uses a strong alarm method, including a high-volume alarm sound, a flashing red warning light, etc., and at the same time generate a first-level blade warning message, including the location of the cracked blade and relevant characteristic data, to remind the operator to take measures as soon as possible.

[0045] Configure the excitation duration, which should be comprehensively determined according to factors such as the material properties of the blade and the intensity of laser excitation. Monitor the continuous duration of exciting the potentially cracked blade. If it is equal to the excitation duration and no crack is determined in the potentially cracked blade, a secondary blade warning is issued. The secondary blade warning is used to prompt the operator that although no crack is detected in this blade, the excitation duration has reached and there is blade abnormality, and further attention is required.

[0046] Working principle and its effects: Realize the full-process monitoring of the compressor blade cracks of a turbofan engine through multi-modal data fusion and intelligent diagnosis technology: First, collect multi-dimensional data such as vibration, stress, and temperature in real time through a distributed sensor network. Use the resampling technology based on rotational speed synchronization to eliminate the rotational speed fluctuation interference in non-stationary signals. Combine Hilbert transform and envelope analysis to construct a multi-modal feature vector including time-domain features, frequency-domain features, and stress gradient features, solving the problem of incomplete feature extraction in traditional methods for non-stationary signal processing. In the abnormal diagnosis stage, screen key features based on the recursive feature elimination algorithm, and combine Gaussian process regression to establish a non-linear mapping model between operating parameters and feature thresholds to achieve dynamic threshold update. Compared with the traditional fixed threshold method, the false alarm rate under complex operating conditions is significantly reduced. The joint trigger rule effectively balances detection sensitivity and specificity by counting the number of features exceeding the threshold within the monitoring period and calculating the deviation degree of key features, avoiding misjudgment of a single feature. Adopt an adaptive sliding time window strategy to extract the long-term change trend of key features, use non-parametric statistical methods to quantify trend features such as monotonically increasing and periodic changes, and combine density clustering algorithm to identify the blade group with trend deviation. By constructing a baseline trend feature value sequence and calculating the dynamic time warping distance difference degree, the feature drift caused by early cracks can be effectively detected. For potentially cracked blades, based on modal analysis, adjust the laser excitation parameters in real time, and use spatial coordinate positioning technology to achieve coordinated control of the azimuth angle and pitch angle of the emitter. The parallel time-sharing excitation strategy ensures the safety and efficiency of multi-target processing. During the excitation process, synchronously monitor the sudden change of strain gradient and the deviation of vibration order amplitude, combine empirical mode decomposition and Hilbert transform to extract the time-frequency domain feature differences, and the multi-level warning mechanism realizes the full-process management from excitation abnormality to crack confirmation through a hierarchical response strategy.

[0047] The system breaks through the technical bottlenecks of traditional monitoring technologies in non-stationary signal processing, dynamic working condition adaptation, and early fault identification, and constructs a closed-loop monitoring system including data acquisition, feature fusion, intelligent diagnosis, precise excitation, and verification and early warning. The crack recognition accuracy is improved through multi-dimensional feature fusion, the adaptability to complex working conditions is significantly enhanced by the dynamic threshold mechanism, the early warning algorithm realizes the capture of weak features, the precise excitation technology ensures the safety and efficiency of crack verification, and the multi-level early warning mechanism optimizes the maintenance decision response process. These technological innovations jointly improve the intelligent level of the health management of turbofan engine blades, providing a solid guarantee for the safe and reliable operation of the engine.

[0048] The above are only the preferred embodiments of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions falling within the concept of the present invention belong to the protection scope of the present invention. It should be noted that for those of ordinary skill in the art, several improvements and refinements made without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A multi-modal monitoring system for cracks in turbofan engine compressor blades, characterized in that: include: Acquire the multimodal data of the compressor blades and perform time parameter alignment to construct a multimodal feature vector; According to the multi-modal feature vector, the tablet press blades are preliminarily diagnosed to identify abnormal blades and mark them. The abnormal blades are triggered in depth by constructing a joint trigger rule, a trend feature value sequence is constructed, and the potential crack blades are identified by measuring the difference degree of the trend feature values ​​of the tablet press blades, and an incentive decision for the potential crack blades is generated. During the execution of the excitation decision, the excitation monitoring is performed by monitoring the change of the strain gradient characteristics of the potential crack blade, and a control blade of the potential crack blade is selected, and by comparing and analyzing the vibration data characteristics of the control blade and the potential crack blade, it is determined whether the potential crack blade has a crack; It also includes multi-level early warning based on the results of comparative analysis.

2. The turbofan engine compressor blade crack multi-modal monitoring system according to claim 1, characterized in that: The specific steps of the preliminary abnormality diagnosis include: Receive the multi-modal feature vector of the compressor blade, select key features through recursive feature elimination and combine with domain prior knowledge; at the same time, obtain the operating condition parameters of the turbofan engine compressor and construct the operating condition feature vector; Based on Gaussian process regression, a nonlinear mapping relationship between operating condition parameters and key feature thresholds is established, and the range of each key feature threshold is dynamically adjusted in real time according to the operating condition feature vector; The real-time key feature value is compared with the corresponding key feature threshold range to determine whether all real-time key feature values ​​of the compressor blades are within the corresponding key feature threshold range. If so, the compressor blades are judged to be in a normal state; otherwise, the compressor blades are judged to be in an abnormal state and are marked as abnormal blades.

3. The turbofan engine compressor blade crack multi-modal monitoring system according to claim 2, characterized in that: The specific steps of the preliminary abnormality diagnosis also include: If the compressor blade is determined to be in an abnormal state, a monitoring threshold is configured to measure the length of time for monitoring the compressor blade in the abnormal state. A joint trigger rule is used to determine whether to trigger a deep assessment based on the number of key features of the abnormal blade that exceeds the key feature threshold range within the monitoring threshold and the degree of excess of the key feature value. A periodic threshold is configured to measure the periodic interval for triggering a depth assessment of each compressor blade. If the compressor blade is determined to be in a normal state, it is determined whether the time interval from the last depth assessment of the compressor blade is equal to the periodic threshold. If so, a depth assessment is triggered.

4. The turbofan engine compressor blade crack multi-modal monitoring system according to claim 3, characterized in that: The specific steps of the in-depth assessment include: Based on the periodic threshold, an evaluation period is set according to a sliding time window management strategy, wherein the sliding time window management strategy is used to dynamically adjust the data value window, and obtain the multi-modal feature vector of each compressor blade within the evaluation period based on the current moment; For each compressor blade, the values ​​of the key features in the evaluation period are arranged in chronological order to construct the key feature value sequence of the compressor blade; Trend features are extracted from the key eigenvalue sequences of compressor blades respectively. The trend features include monotonic features, nonlinear trend features and periodic features. Each trend eigenvalue is calculated by non-parametric statistical methods to construct the trend eigenvalue sequence of each compressor blade.

5. The turbofan engine compressor blade crack multi-modal monitoring system according to claim 4, characterized in that: The specific steps of the in-depth assessment also include: A clustering algorithm is used to cluster the trend characteristic value sequence of compressor blades, identify the compressor blade groups with trend deviation, and mark them as deviated blades; The trend characteristic value sequences marked as abnormal blades and deviated blades are respectively eliminated, and the mean values ​​of the trend characteristic value sequences of the remaining compressor blades are respectively calculated to construct a baseline trend characteristic value sequence; The distance metric is used to calculate the difference between the trend characteristic value sequence of each deviated leaf and abnormal leaf and the baseline trend characteristic value sequence; A difference threshold is configured to determine whether the difference between the trend characteristic value sequence of the deviated blade or abnormal blade and the baseline trend characteristic value sequence is greater than the difference threshold. If so, it is marked as a potential crack blade.

6. The turbofan engine compressor blade crack multi-modal monitoring system according to claim 1, characterized in that: The specific steps of generating the incentive decision of the potential cracked blade include: The circumferential position coordinates of the potential crack blade and its difference degree are obtained in real time, and the laser pulse energy is dynamically adjusted by combining the difference degree with the basic energy of the laser pulse; The natural frequency of the blade with potential cracks is obtained according to the real-time rotation speed, and the laser excitation frequency band is set with the natural frequency as the center; According to the position of the compressor blade group where the potential crack blade is located, a transmitter is selected, and according to the relative position of the transmitter coordinates and the circumferential position coordinates of the potential crack blade, the azimuth angle and the pitch angle of the transmitter are set; The laser beam is emitted to the surface of the potential crack blade through the transmitter. When a group of compressor blades has multiple potential crack blades, parallel excitation is implemented, and the laser beam is emitted in time-sharing order according to the circumferential order of the potential crack blades.

7. The turbofan engine compressor blade crack multi-modal monitoring system according to claim 1, characterized in that: The specific steps of determining whether a potential crack blade has cracks include: Extract stress data from the multimodal data of the potential crack blade, and perform stress diagnosis on the blade group where the potential crack blade is located, including: Extract the strain gradient characteristic value according to the stress data and configure the strain threshold. If the strain gradient characteristic value of the blade with potential crack is greater than the strain threshold, stop the laser excitation and trigger the excitation warning. Based on the principle of spatial symmetry, the position coordinates of the potential cracked blades in the compressor blade group are determined, and the control blades are selected according to the spatial layout structure of the compressor blades. Extracting vibration data features from multimodal data of potential crack blades and control blades, the vibration data features including vibration order amplitude and time-frequency spectrum features; Configure the order amplitude threshold. For the vibration order amplitude, calculate the relative deviation of the amplitude between the potential crack blade and the control blade at the same order. If the relative deviation of the amplitude between the potential crack blade and the control blade at the same order exceeds the order amplitude threshold, it is determined that the potential crack blade has a crack.

8. The turbofan engine compressor blade crack multi-modal monitoring system according to claim 7, characterized in that: The specific step of determining whether the potential crack blade has cracks also includes: Configuring principal component thresholds , perform empirical mode decomposition on the vibration data of the potential crack blade and the control blade, decompose the vibration data into multiple intrinsic mode functions, and select the energy proportion Perform Hilbert transform on the intrinsic mode functions to obtain the time-frequency spectrum; The characteristic frequency of the time-frequency spectrum is identified by peak detection. On the time-frequency spectrum, the frequency bandwidth is set around the characteristic frequency, and the energy density of the characteristic frequency is calculated by integration. An energy deviation threshold is set to calculate the relative deviation of the energy density of the characteristic frequency of the potential crack blade and the control blade. If it is greater than the energy deviation threshold, it is determined that the potential crack blade has a crack. When it is determined that a potential crack blade has cracks, the stimulation of the potential crack blade is stopped and a first-level blade warning is issued; Configure the excitation duration to monitor the duration of excitation for blades with potential cracks. If it is equal to the excitation duration and the blades with potential cracks are not judged to have cracks, a secondary blade warning is issued.

9. The turbofan engine compressor blade crack multi-modal monitoring system according to claim 1, characterized in that: The specific steps of constructing the multimodal feature vector include: The real-time speed of the compressor is obtained to resample the vibration data in the multi-modal data and generate vibration data with equal angle intervals; and the target order is defined according to the number of compressor blades in each group; Delimiting the resonance frequency band, filtering the resampled vibration data, obtaining the filtered vibration data and performing Hilbert transform, generating orthogonal data of the vibration data, calculating the envelope data, and performing Fourier transform on the envelope data to obtain the envelope spectrum; Performing a windowing process on the resampled vibration data and then performing a Fourier transform to obtain a frequency spectrum of the resampled vibration data; According to the target order, the frequency axis of the spectrum of the resampled vibration data and the envelope spectrum is mapped to the order axis, and the corresponding order spectrum is generated by using an interpolation algorithm; The order spectra are extracted separately and combined with the time domain characteristics of the multimodal data to form a multimodal feature vector for each compressor blade.

10. The turbofan engine compressor blade crack multi-modal monitoring system according to claim 3, characterized in that: The joint triggering rules include: Within the monitoring threshold time range, the number of key features exceeding the corresponding key feature threshold range is counted, and when the number of key features exceeding the key feature threshold range exceeds a preset number threshold, the first rule is satisfied; Calculate the relative deviation of each key feature value exceeding the key feature threshold range relative to its corresponding key feature threshold range, and when there is a relative deviation exceeding a preset deviation threshold, the second rule is satisfied; When the first rule and the second rule are satisfied at the same time, that is, when the number of key features exceeding the key feature threshold range reaches the preset number threshold and the relative deviation degree of at least one key feature reaches the preset deviation degree threshold, it is determined that the depth assessment is triggered.

Images