Respiration module turbine speed change fault prediction method of life support system
By acquiring, filtering, and extracting turbine signals, nonlinear and structural features are identified, and a turbine abnormal fluctuation model is established. This solves the problem of the accuracy of turbine fault prediction and improves the stability and safety of the system.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-16
- Publication Date
- 2026-03-31
AI Technical Summary
When faced with drastic changes in the concentration of particulate matter in the air, the traditional speed monitoring and fault diagnosis methods for the turbine unit of the breathing module in existing life support systems cannot accurately predict potential speed change faults, thus affecting the stability of the system.
By acquiring turbine speed signals and performing signal filtering, the amplitude and frequency domain characteristics of speed fluctuations are extracted, nonlinear speed fluctuations and structural correlation characteristics are identified, an abnormal fluctuation model is established, the turbine transmission fault trend is predicted, and an early warning signal is output.
It improves the sensitivity of turbine fault diagnosis, enhances the safety and stability of the life support system in complex environments, and ensures the reliable operation of the system.
Smart Images
Figure CN121765248A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of fault prediction technology, and more specifically, to a method for predicting faults in the turbine transmission of a breathing module in a life support system. Background Technology
[0002] The respiratory modules of existing life support systems typically operate in clean and constant environmental conditions, relying on turbine devices to continuously and stably provide auxiliary airflow.
[0003] In real-world clinical applications, such as emergency transport, rescue, or sudden pollution events, the concentration of airborne particles entering the respiratory module can change drastically, affecting the turbine blade surface and causing nonlinear micro-deformation. Traditional speed monitoring and fault diagnosis methods cannot accurately predict potential speed change faults, impacting the stability of the life support system. Summary of the Invention
[0004] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a method for predicting the failure of the turbine transmission of the breathing module in a life support system to solve the problems mentioned in the background art.
[0005] To achieve the above objectives, the present invention provides the following technical solution: A method for predicting turbine transmission failure in the respiratory module of a life support system includes the following steps: S1: Collect the real-time rotational speed signal of the turbine of the breathing module under different particle concentration conditions and perform signal filtering to obtain the turbine speed filtered signal; S2: Based on the turbine speed filtering signal, extract the turbine speed fluctuation amplitude feature and fluctuation frequency domain feature respectively, and generate speed amplitude feature data and frequency domain feature data; S3: Use rotational speed amplitude characteristic data to identify nonlinear velocity fluctuations caused by micro-deformation of turbine blades and generate turbine nonlinear fluctuation identification results; S4: Utilize frequency domain feature data to perform spectrum analysis of turbine speed signals, determine the structural correlation characteristics of turbine speed fluctuations, and output the structural feature analysis results; S5: Based on the turbine nonlinear fluctuation identification results and structural feature analysis results, establish a turbine abnormal fluctuation model and output the turbine abnormal fluctuation assessment results; S6: Predict the turbo gearbox failure trend based on the turbo abnormal fluctuation assessment results, and output a turbo gearbox failure early warning signal.
[0006] In a preferred embodiment, S1 specifically refers to: A speed sensor is used to collect the turbine speed signal of the breathing module turbine under different particle concentration conditions in real time; The real-time turbine speed signal is filtered using a signal filter to remove high-frequency interference noise, resulting in a filtered turbine speed signal.
[0007] In a preferred embodiment, S2 specifically refers to: A time-domain analysis of the turbine speed filter signal is performed based on the amplitude statistical analysis method to extract the amplitude characteristics of turbine speed fluctuations and obtain speed amplitude characteristic data. Frequency domain analysis of the turbine speed filter signal is performed based on the Fast Fourier Transform method to extract the frequency domain features of turbine speed fluctuations and obtain frequency domain feature data.
[0008] In a preferred embodiment, S3 specifically refers to: A preset threshold for identifying nonlinear speed fluctuations is established, and threshold comparisons are performed based on rotational speed amplitude characteristic data. Within a preset sliding time window, trend analysis is performed on the speed amplitude characteristic data to obtain the slope of the data curve; When the slope of the curve exceeds the nonlinear velocity fluctuation identification threshold and the speed amplitude characteristic data shows a sudden change, the nonlinear velocity fluctuation caused by the micro-deformation of the turbine blade is determined, and the turbine nonlinear fluctuation identification result is output.
[0009] In a preferred embodiment, S4 specifically refers to: The main frequency range of turbine speed fluctuation is determined based on frequency domain feature data, and the corresponding spectral amplitude data of the main frequency range is extracted. Perform a peak value extraction operation on the spectral amplitude data to obtain the spectral peak value corresponding to the turbine speed fluctuation; By comparing the spectral peaks using preset structural correlation feature judgment criteria, the structural correlation features of turbine speed fluctuations are determined, and the structural feature analysis results are output.
[0010] In a preferred embodiment, S5 specifically refers to: A model for abnormal turbine fluctuations was constructed based on the results of turbine nonlinear fluctuation identification and structural feature analysis. By fusing the turbine nonlinear fluctuation identification results and structural feature analysis results using a turbine abnormal fluctuation model, the correlation between turbine speed fluctuation and turbine blade micro-deformation is determined. The degree of abnormal fluctuation is assessed based on the correlation between turbine speed fluctuation and turbine blade micro-deformation, and the turbine abnormal fluctuation assessment results are output.
[0011] In a preferred embodiment, S6 specifically refers to: Preset turbo transmission fault warning threshold; Compare the results of the abnormal fluctuation assessment of the turbocharger with the turbocharger transmission fault warning threshold; When the abnormal fluctuation assessment result of the turbine reaches or exceeds the turbine transmission failure warning threshold, it is determined that there is a risk of transmission failure in the turbine, and a turbine transmission failure warning signal is output.
[0012] The technical effects and advantages of the present invention regarding a method for predicting turbine transmission failure in a respiratory module of a life support system are as follows: By acquiring and filtering turbine speed signals in real time, high-frequency interference is effectively removed, ensuring the accuracy of feature extraction. By extracting turbine speed fluctuation amplitude and frequency domain features, a basis for discrimination of different types of fluctuations is established. By identifying nonlinear speed fluctuations and structurally related spectral anomalies respectively, the ability to perceive latent faults caused by turbine micro-deformation is improved. Based on the turbine nonlinear fluctuation identification results and structural feature analysis results, a turbine abnormal fluctuation model is established to achieve a comprehensive judgment of turbine speed change abnormal states. Based on the turbine abnormal fluctuation assessment results, turbine speed change fault trend prediction and turbine speed change fault early warning output are realized, improving the sensitivity of turbine fault diagnosis and enhancing the safety and stability assurance capability of the life support system in complex operating environments. Attached Figure Description
[0013] Figure 1 This is a schematic diagram of a method for predicting turbine transmission failure in the respiratory module of a life support system according to the present invention. Detailed Implementation
[0014] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention. Example
[0015] Figure 1 This invention provides a method for predicting turbine transmission failure in the respiratory module of a life support system, comprising the following steps: S1: Collect the real-time rotational speed signal of the turbine of the breathing module under different particle concentration conditions and perform signal filtering to obtain the turbine speed filtered signal; S2: Based on the turbine speed filtering signal, extract the turbine speed fluctuation amplitude feature and fluctuation frequency domain feature respectively, and generate speed amplitude feature data and frequency domain feature data; S3: Use rotational speed amplitude characteristic data to identify nonlinear velocity fluctuations caused by micro-deformation of turbine blades and generate turbine nonlinear fluctuation identification results; S4: Utilize frequency domain feature data to perform spectrum analysis of turbine speed signals, determine the structural correlation characteristics of turbine speed fluctuations, and output the structural feature analysis results; S5: Based on the turbine nonlinear fluctuation identification results and structural feature analysis results, establish a turbine abnormal fluctuation model and output the turbine abnormal fluctuation assessment results; S6: Predict the turbo gearbox failure trend based on the turbo abnormal fluctuation assessment results, and output a turbo gearbox failure early warning signal.
[0016] S1: Acquire real-time turbine rotation speed signals of the breathing module under different particle concentration conditions and perform signal filtering to obtain a turbine rotation speed filtered signal, including: A speed sensor is used to collect the turbine speed signal of the breathing module turbine under different particle concentration conditions in real time; In the respiratory module of a life support system, the turbine, as a precision rotating mechanical component, is typically used to assist in the delivery of airflow for human respiration. The turbine's rotational speed directly reflects its operational status. To monitor the turbine's actual operation under different particulate concentrations, especially in environments with sudden changes in cleanliness, such as emergency medical care and patient transport, where the concentration of airborne particles can change significantly or even generate sudden high concentrations of dust or particulate pollutants, it is crucial to monitor the turbine's rotational speed in real time. Different particulate concentrations can have subtle physical effects on the turbine blades, causing changes in turbine speed.
[0017] High-precision speed sensors, such as magnetoelectric or photoelectric speed sensors, are installed at appropriate locations on the outside of the turbine bearings or at the blade tips of the respiratory module in the life support system. These sensors are used to collect the speed signals generated during turbine blade rotation in real time, ensuring continuous and accurate measurement of the turbine's operating status. Taking a magnetoelectric speed sensor as an example, its working principle is as follows: a magnet is placed at the tip of the turbine blade. During turbine rotation, whenever the magnet passes the sensor's sensing head, the sensor generates an electrical pulse signal proportional to the rotational speed. By collecting these electrical pulse signals in real time, the current real-time turbine speed can be calculated, thus obtaining accurate real-time speed signals under different particle concentration conditions. Real-time speed signals refer to the continuous recording of turbine speed values at a high sampling frequency, such as collecting more than 1000 data points per second, thereby making data analysis and abnormal fluctuation identification more accurate.
[0018] The real-time turbine speed signal is filtered using a signal filter to remove high-frequency interference noise, resulting in a filtered turbine speed signal. During the acquisition of real-time turbine speed signals, high-frequency interference noise can be generated due to external environmental factors or electromagnetic interference and mechanical vibration of the equipment itself. This high-frequency interference noise, mixed in with the real-time turbine speed signals, can severely interfere with the validity and accuracy of the actual signal. For example, life support systems in hospital environments are often accompanied by the operation of various medical devices, such as defibrillators, high-frequency electrosurgical units, and MRI scanners. These devices generate high-frequency electromagnetic interference during use, affecting the turbine speed signal of the respiratory module in the life support system, manifesting as a large amount of high-frequency noise superimposed on the speed signal. If this high-frequency interference noise is not filtered out in a timely manner, it can lead to misjudgments in the analysis process, making it impossible to accurately determine whether there are critical conditions such as abnormal deformation of the turbine blades, thereby reducing the accuracy and reliability of fault prediction methods.
[0019] Therefore, a signal filter is used to effectively filter the real-time turbine speed signal. The signal filter can be a Butterworth low-pass filter, Chebyshev low-pass filter, or Bessel low-pass filter, which have high filtering accuracy to effectively remove high-frequency noise components from the real-time turbine speed signal. Taking the Butterworth low-pass filter as an example, the filtering method is as follows: First, determine a suitable cutoff frequency based on the true characteristic frequency range of the turbine speed signal. For example, if the frequency of normal turbine speed fluctuations is below 100Hz, the filter cutoff frequency can be set to around 120Hz. The real-time turbine speed signal is then processed by the Butterworth low-pass filter. The Butterworth low-pass filter effectively suppresses high-frequency noise components with frequencies exceeding the preset cutoff frequency, retaining the lower-frequency true turbine speed fluctuation signal. After filtering, the obtained turbine speed filtered signal is an effective and clean signal, which can be used for the analysis of nonlinear speed fluctuation characteristics and frequency domain structural characteristics, thus providing basic data support for the accurate prediction of turbine transmission faults.
[0020] S2: Based on the turbine speed filtering signal, extract the turbine speed fluctuation amplitude features and fluctuation frequency domain features respectively, and generate speed amplitude feature data and frequency domain feature data, including: A time-domain analysis of the turbine speed filter signal is performed based on the amplitude statistical analysis method to extract the amplitude characteristics of turbine speed fluctuations and obtain speed amplitude characteristic data. The turbine speed filter signal is a real-time signal that has undergone filtering to remove high-frequency noise, containing accurate information about the actual turbine speed changes. Amplitude statistical analysis is used to perform time-domain analysis on the turbine speed filter signal. This method analyzes the variation of the turbine speed filter signal over time, including but not limited to calculating multiple statistical quantities such as the average, peak, trough, root mean square, variance, and range of the turbine speed filter signal to characterize the amplitude fluctuations. For example, a time-domain analysis window of a certain length is first set; for instance, with a data acquisition frequency of 1000 data points per second, a 1-second window can be selected. Statistical analysis is then performed on all data points of the turbine speed filter signal within the 1-second window. By calculating the average amplitude, maximum amplitude, and minimum amplitude of the turbine speed filter signal within the time-domain analysis window, the range and pattern of amplitude variation are obtained.
[0021] For example, when the turbine of the breathing module operates under different particulate concentrations, assuming the normal turbine speed amplitude varies between 1500 rpm and 1550 rpm, under sudden changes in particulate concentration, due to slight blade deformation or uneven stress, the maximum amplitude of the turbine speed filter signal may exceed 1560 rpm or the minimum amplitude may be lower than 1490 rpm. Through amplitude statistical analysis, subtle but diagnostically significant abnormal changes in the speed signal in the time domain can be effectively captured and quantified, thus forming speed amplitude characteristic data.
[0022] The obtained speed amplitude characteristic data is a set of statistical index data, such as average speed value, maximum amplitude of speed fluctuation, minimum amplitude, range, standard deviation, etc.
[0023] Frequency domain analysis of the turbine speed filter signal is performed based on the fast Fourier transform method to extract the frequency domain features of turbine speed fluctuations and obtain frequency domain feature data. Frequency domain analysis is an analytical method that transforms a signal from the time axis to the frequency axis. It can reveal the presence and amplitude of different frequency components in the signal, thereby reflecting the periodic or non-periodic characteristics of turbine speed changes.
[0024] Specifically, the Fast Fourier Transform (FFT) method is used to perform frequency domain analysis on the turbine speed filtered signal. This involves first performing a Fourier transform on all data points of the turbine speed filtered signal within a preset analysis time window to obtain a series of frequency-corresponding spectral distributions. These spectral distributions represent the intensity and distribution of the turbine speed filtered signal at various frequencies.
[0025] For example, suppose the turbine speed filter signal contains a main fluctuating frequency component of 25Hz within a preset analysis window. When the turbine blades are impacted by particles, causing micro-deformation, the speed signal may exhibit additional abnormal frequency components, such as 12.5Hz or 50Hz. In frequency domain analysis, this can be clearly revealed using the Fast Fourier Transform (FFT) method. Assuming a sampling frequency of 1000Hz, 1000 data points are obtained within a one-second sampling window. Through FFT, the original turbine speed filter signal can be transformed into a spectrum containing frequency information, thereby identifying the frequency components of the turbine speed fluctuations.
[0026] Extracting frequency domain feature data means extracting frequency components with significant characteristics from the spectrum, including the main frequency, secondary main frequency, harmonic frequency components, and the amplitude at the corresponding frequency.
[0027] S3: Utilize rotational speed amplitude characteristic data to identify nonlinear velocity fluctuations caused by micro-deformation of turbine blades, and generate turbine nonlinear fluctuation identification results, including: A preset threshold for identifying nonlinear speed fluctuations is established, and threshold comparisons are performed based on rotational speed amplitude characteristic data. The nonlinear speed fluctuation identification threshold is a baseline threshold set for abnormal states that may occur during the operation of the turbine in the breathing module of the life support system. This threshold serves as a reference standard for determining whether turbine speed fluctuations are abnormal, and is used to identify nonlinear fluctuations caused by minute deformations of the turbine blades. These minute deformations typically originate from the impact of particles on the turbine blades when air flows into the turbine from environments with varying particle concentrations. This impact causes uneven local stress and minute deformations, resulting in nonlinear speed fluctuations in the speed signal.
[0028] The method for setting the nonlinear speed fluctuation identification threshold is as follows: A normal fluctuation range for the speed amplitude characteristic data is established using a large amount of historical experimental data and normal operating condition data. Based on the normal operating state of the turbine in the breathing module of the life support system, turbine speed signals are measured under multiple sets of different particle concentration conditions, and the normal distribution range of a large amount of speed amplitude characteristic data is obtained. For example, under normal turbine operating conditions, the original measured real-time turbine speed filter signal ranges from 1500 to 1550 rpm. At this time, the statistical results of the speed amplitude characteristic data typically show: an average speed of approximately 1525 rpm, a maximum amplitude of approximately 1550 rpm, a minimum amplitude of approximately 1500 rpm, a speed range of approximately 50 rpm, and a small standard deviation (e.g., less than 5 rpm).
[0029] Under abnormal conditions where turbine blades undergo minute deformation due to particle impact, the original values of the turbine real-time speed filtering signal will deviate significantly from the above range. For example, the maximum value may exceed 1560 rpm or the minimum value may fall below 1490 rpm. The corresponding statistics in the speed amplitude characteristic data will also change significantly. For instance, the range may increase significantly (e.g., exceeding 70 rpm), and the standard deviation may also increase significantly (e.g., exceeding 10 rpm). Based on these significantly changed statistical characteristics, it is possible to distinguish between the normal operating state and the abnormal operating state of the turbine.
[0030] Within a preset sliding time window, trend analysis is performed on the speed amplitude characteristic data to obtain the slope of the data curve; A preset sliding time window refers to a dynamic time domain range selected during the operation of the life support system to continuously observe the changing trends of turbine speed amplitude characteristic data. The sliding time window can be set to a certain duration according to actual needs, such as 1 second. The data point set within the window is updated every certain time interval (e.g., 0.1 seconds) to achieve continuous analysis. Each time the sliding window contains the latest speed amplitude characteristic data, and the oldest data points outside the window are removed, forming data continuity and real-time performance, facilitating the observation of the changing trends of turbine speed amplitude characteristic data.
[0031] Trend analysis within a sliding time window involves using mathematical analysis or numerical calculation methods to linearly fit the turbine speed amplitude characteristic data and obtain the slope of the data curve. For example, with 1000 data points within a sliding window, using the x-axis as the time of data collection and the y-axis as the turbine speed amplitude characteristic data, a least-squares linear regression analysis is performed on the data points to obtain the slope of the fitted curve. The magnitude of the curve slope characterizes the speed and trend of change in the turbine speed amplitude characteristic data: a positive slope indicates an upward trend in turbine speed within the window, a negative slope indicates a downward trend, and a larger slope value indicates a more pronounced trend change.
[0032] For example, under normal turbine operation, the slope of the turbine speed amplitude characteristic data curve is generally stable within a small range, such as between -0.1 and 0.1, which is close to a stable state. When the turbine blades undergo micro-deformation due to sudden changes in particle concentration, the amplitude characteristic data may change rapidly in a very short time. For example, the slope within the window may reach 0.5 or even higher. At this time, it can be judged that the turbine speed data shows an obvious abnormal trend, indicating possible nonlinear fluctuation phenomena.
[0033] When the slope of the curve exceeds the nonlinear velocity fluctuation identification threshold and the speed amplitude characteristic data changes abruptly, the nonlinear velocity fluctuation caused by the micro-deformation of the turbine blade is determined, and the turbine nonlinear fluctuation identification result is output. The judgment process is as follows: when the slope of the data curve obtained within the preset sliding time window exceeds the preset nonlinear speed fluctuation identification threshold. For example, if the curve slope exceeds the nonlinear speed fluctuation identification threshold of 0.4, and at the same time, the speed amplitude characteristic data itself also shows a sudden change, such as a rapid jump from 1540 rpm to 1570 rpm or even higher within a window period, then it meets the abnormal characteristic conditions of turbine nonlinear speed fluctuation.
[0034] If the above situation occurs, it can be determined that the current turbine speed fluctuation is a nonlinear velocity fluctuation phenomenon caused by the micro-deformation of the turbine blades. Nonlinear velocity fluctuations are directly related to the micro-deformation of the turbine blades. The micro-deformation causes slight changes in the aerodynamic structure of the turbine blades, thereby affecting the turbine rotational stability under the action of airflow, which manifests as a significant abrupt change in the speed amplitude.
[0035] S4: Utilize frequency domain feature data to perform spectral analysis of the turbine speed signal, determine the structural correlation characteristics of turbine speed fluctuations, and output the structural feature analysis results, including: The main frequency range of turbine speed fluctuation is determined based on frequency domain feature data, and the corresponding spectral amplitude data of the main frequency range is extracted. Frequency domain characteristic data represents the frequency distribution of turbine speed signals. Different frequency components and their corresponding amplitudes reflect the dynamic characteristics of turbine operation and potential anomalies. To determine the main characteristics of turbine speed fluctuations, the dominant frequency range of turbine speed fluctuations is identified using frequency domain characteristic data. The dominant frequency range refers to the frequency range in the frequency domain characteristic data where the spectral amplitude is significantly higher and more concentrated, reflecting the most prominent periodic or harmonic characteristics of turbine speed fluctuations.
[0036] For example, during the operation of a turbine in a life support system, when the turbine is in normal operation, the frequency domain characteristic data typically exhibits a specific and stable dominant frequency range, such as the spectral amplitude being most prominent in the 20Hz to 30Hz frequency range, which represents the dominant frequency range of turbine speed fluctuations. When the turbine blades are subjected to sudden changes in particle concentration, resulting in minor deformation or vibration, additional secondary frequency ranges may appear in the spectral data. For instance, the dominant frequency range may extend to between 15Hz and 35Hz, or multiple distinct peak ranges may appear. To accurately determine the dominant frequency range, automatic or semi-automatic methods are employed. For example, firstly, spectral peak identification rules are defined, using a minimum threshold for spectral amplitude or a specific proportion of spectral amplitude exceeding the average spectral amplitude (e.g., greater than 20% of the average spectral amplitude) as a standard. Point-by-point analysis is then used to determine the most prominent frequency distribution range in the frequency domain characteristic data.
[0037] After determining the dominant frequency range, the corresponding spectral amplitude data is extracted from the frequency domain feature data. Spectral amplitude data refers to the amplitude at each frequency point within the dominant frequency range, reflecting the fluctuation intensity of the turbine speed signal at that frequency point. For example, if the dominant frequency range is determined to be 20Hz to 30Hz, the spectral amplitude data corresponding to each frequency point within this range will be extracted, such as an amplitude of 0.5 for 20Hz, 0.8 for 25Hz, and 0.4 for 30Hz, forming complete and continuous spectral amplitude data.
[0038] Perform a peak value extraction operation on the spectral amplitude data to obtain the spectral peak value corresponding to the turbine speed fluctuation; Spectrum peak extraction involves data analysis of spectral amplitude data obtained within the main frequency range. Spectrum peaks are the most prominent feature in spectral amplitude data, typically appearing as local maxima and representing the core frequency characteristics of turbine speed fluctuations. Specifically, for the spectral amplitude data corresponding to the main frequency range, a spectrum peak identification algorithm or analysis rule is used, such as the local maximum identification method. This involves analyzing the spectral amplitude data point by point, and determining the frequency containing the spectrum peak when the spectral amplitude at a certain frequency point is higher than that of the adjacent frequency points.
[0039] For example, within the main frequency range (20Hz to 30Hz), analysis using spectral peak extraction revealed that the spectral amplitude at 25Hz was significantly higher than that at the adjacent 24Hz and 26Hz frequencies. Therefore, 25Hz was identified as the frequency of the spectral peak, and the corresponding spectral amplitude (e.g., 0.8) represents the spectral peak corresponding to the turbine speed fluctuation. A peak was identified when the spectral amplitude exceeded that of adjacent frequencies by more than 10%, thus accurately capturing the significant frequency characteristics in the turbine speed fluctuation spectrum.
[0040] By comparing the spectral peaks using preset structural correlation feature judgment criteria, the structural correlation features of turbine speed fluctuations are determined, and the structural feature analysis results are output. Structural correlation features refer to the frequency domain characteristics that are directly related to turbine speed fluctuations and turbine blade micro-deformation, and can effectively identify the correlation between turbine operating conditions and structural anomalies. Setting criteria for structural correlation feature judgment, i.e., pre-defining the basis for feature judgment, can be the location, number, and relative amplitude of spectral peaks.
[0041] For example, the criteria for determining structural correlation characteristics can be set as one or a combination of the following criteria: when the position of the spectral peak deviates significantly from the normal spectral peak position, for example, the main peak is at 25Hz in normal conditions, but moves to 28Hz or has an additional peak position such as 12.5Hz in abnormal conditions; or when the number of spectral peaks increases, with a single obvious peak in normal conditions and two or more obvious peaks in abnormal conditions; or when the peak amplitude is significantly higher (e.g., the amplitude increases by more than 30%) or significantly lower than in normal conditions, it is determined that there is a structural correlation between turbine speed fluctuations and turbine structural changes.
[0042] When the spectral peak value is compared with the structurally relevant feature judgment criteria, and one or more criteria are met, it is determined that the turbine speed fluctuation has structurally relevant characteristics, indicating that the turbine has experienced structurally relevant abnormal fluctuations, and the structural feature analysis result is output. The structural feature analysis result is the output result of the feature comparison, usually output as a judgment result or structural feature type, such as outputting conclusions like "structural feature abnormal" or "structural feature normal," as the basis for turbine transmission fault prediction.
[0043] S5: Based on the turbine nonlinear fluctuation identification results and structural feature analysis results, establish a turbine abnormal fluctuation model and output turbine abnormal fluctuation assessment results, including: A model for abnormal turbine fluctuations was constructed based on the results of turbine nonlinear fluctuation identification and structural feature analysis. The turbine abnormal fluctuation model is a model used to comprehensively analyze and evaluate abnormal fluctuations in turbine speed. It uses a fusion analysis method to jointly analyze the nonlinear fluctuation characteristics and structural frequency domain characteristics of the turbine. Through data correlation analysis, it can accurately identify abnormal fluctuations that may exist during turbine operation and are related to minor structural changes.
[0044] Specifically, when constructing a turbine anomaly fluctuation model, a fusion analysis method is used to establish identification rules for anomaly fluctuation characteristics. For example, the nonlinear fluctuation identification result is set as one of the input variables of the turbine anomaly fluctuation model, while the structural feature analysis result is set as another input variable. A comprehensive evaluation system or calculation rule that can integrate the turbine nonlinear fluctuation identification result and the structural feature analysis result is constructed. For example, when the turbine nonlinear fluctuation identification result is obviously abnormal and the structural feature analysis result also clearly shows abnormal spectral fluctuations, a corresponding rule can be defined in the turbine anomaly fluctuation model. For example, the rule can be defined as: when both the nonlinear fluctuation identification anomaly and the structural feature analysis anomaly are satisfied, the turbine is determined to be in a highly correlated abnormal state. The model is established through rule-based processing to achieve a comprehensive analysis of the turbine's abnormal operating state.
[0045] By fusing the turbine nonlinear fluctuation identification results and structural feature analysis results using a turbine abnormal fluctuation model, the correlation between turbine speed fluctuation and turbine blade micro-deformation is determined. For example, mathematical fusion methods can be used to implement the fusion calculation process, such as weighted scoring, fusion index, or multi-factor comprehensive analysis. If the weighted scoring method is used, first define the weight coefficients. For example, the weight of the turbine nonlinear fluctuation identification result is 0.6, and the weight of the structural feature analysis result is 0.4. When the turbine nonlinear fluctuation identification result shows a clear abnormality (score of 0.9), and the structural feature analysis result shows a low degree of abnormality (score of 0.5), the total score is calculated through weighted fusion calculation, which is (0.6×0.9+0.4×0.5)=0.74, representing the degree of correlation between turbine speed fluctuation and turbine blade micro-deformation.
[0046] The determined correlation reflects the close relationship between abnormal fluctuations in turbine speed signals and micro-deformation of blades. The higher the value, the higher the correlation, indicating that the micro-deformation of turbine blades has a more significant impact on turbine speed fluctuations, thus reflecting the degree of influence of micro-deformation of blades on turbine operation.
[0047] The degree of abnormal fluctuation is assessed based on the correlation between turbine speed fluctuation and turbine blade micro-deformation, and the turbine abnormal fluctuation assessment results are output. An abnormal fluctuation severity grading method or threshold assessment method is employed. For example, the correlation severity can be divided into multiple levels: a correlation severity value in the range of 0–0.3 is defined as normal turbine operation; a correlation severity value in the range of 0.31–0.6 is defined as slightly abnormal turbine operation, indicating that the turbine blades may have minor deformation but the abnormal fluctuation is not severe; a correlation severity value in the range of 0.61–0.8 is defined as significantly abnormal turbine operation, indicating significant micro-deformation of the turbine blades and marked abnormal operation; a correlation severity value in the range of 0.81–1.0 is defined as severely abnormal turbine operation, representing severe turbine blade deformation, and the turbine operation may have entered a dangerous state. After completing the abnormal fluctuation severity assessment, the turbine abnormal fluctuation assessment results are output.
[0048] S6: Predicts turbo transmission failure trends based on turbo abnormal fluctuation assessment results and outputs turbo transmission failure early warning signals, including: Preset turbo transmission fault warning threshold; The turbocharger transmission fault warning threshold refers to the critical standard or reference value used to determine whether the assessment result of abnormal turbocharger fluctuations reaches the threshold required to issue a fault warning signal. The purpose of the warning threshold is to define and represent the alert standard for turbocharger transmission fault risk. When the assessment result of abnormal turbocharger fluctuations reaches or exceeds the turbocharger transmission fault warning threshold, it is determined that there is a risk in the turbocharger's operating condition, and a warning message should be issued promptly to remind staff to take appropriate emergency measures to prevent the turbocharger's operating condition from further deteriorating.
[0049] The method for setting the turbine transmission fault warning threshold is as follows: First, collect and analyze historical operating data and abnormal fluctuation data under fault conditions of the turbine in the respiratory module of the life support system. Historical data includes, but is not limited to, the speed signals, amplitude characteristics, frequency domain characteristics, and abnormal fluctuation assessment results generated when the turbine is operating normally and abnormally under different particulate concentration conditions. Based on the historical data, statistical analysis and numerical optimization are performed to determine the critical threshold, which is the turbine transmission fault warning threshold.
[0050] For example, through analysis of historical data, assuming that the turbine abnormal fluctuation assessment result is stable below 0.3 under normal operating conditions, slightly abnormal operating conditions are usually between 0.31 and 0.6, abnormal operating conditions are between 0.61 and 0.8, and severely abnormal operating conditions exceed 0.81, the turbine transmission fault warning threshold is usually selected at around 0.6. That is, when the turbine abnormal fluctuation assessment result reaches 0.6 or above, it is determined that there is a risk in turbine operation.
[0051] Compare the results of the abnormal fluctuation assessment of the turbocharger with the turbocharger transmission fault warning threshold; The abnormal fluctuation assessment result of the turbine is compared with the turbine transmission failure warning threshold. For example, if the abnormal fluctuation assessment result of the turbine is 0.74 and the preset turbine transmission failure warning threshold is 0.6, the abnormal fluctuation assessment result of the turbine is higher than the preset turbine transmission failure warning threshold. Thus, it is concluded that the turbine operating state has reached the critical condition of abnormality and there is a risk of transmission failure.
[0052] When the abnormal fluctuation assessment result of the turbine reaches or exceeds the turbine transmission failure warning threshold, it is determined that there is a risk of transmission failure in the turbine and a turbine transmission failure warning signal is output. When the abnormal fluctuation assessment result of the turbine is compared with the turbine transmission failure warning threshold, if the abnormal fluctuation assessment result reaches or exceeds the preset turbine transmission failure warning threshold, it is determined that the current operating state of the turbine is abnormal and there is a risk of transmission failure. The warning operation must be executed immediately.
[0053] The judgment logic is as follows: if the turbine abnormal fluctuation assessment result reaches or exceeds the turbine transmission failure warning threshold, it indicates that the turbine's operating state has reached a significantly abnormal or dangerous state. The turbine may experience nonlinear or structural abnormal fluctuations due to blade micro-deformation, mechanical structural damage, or changes in airflow channels, which may lead to turbine instability or even damage, causing the life support system to malfunction. In this case, a turbine transmission failure warning signal needs to be issued promptly. The turbine transmission failure warning signal can be a clear alarm message, such as a voice alarm, audible and visual alarm, interface text warning, or a warning message sent to the remote monitoring system, to prompt the life support system operators to pay attention and take necessary emergency measures.
[0054] For example, when the abnormal fluctuation assessment result of the turbine is 0.74, which exceeds the turbine transmission failure warning threshold of 0.6, the control unit in the life support system automatically triggers the alarm mechanism and outputs a turbine transmission failure warning signal. For example, a clear alarm prompt is displayed on the monitoring display interface of the life support system, such as "Warning: There is a risk of transmission failure in the turbine. Please check and handle it immediately!" At the same time, an alarm sound is emitted or a warning light flashes to ensure that the operator can detect it in time and take appropriate maintenance measures.
[0055] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.
[0056] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.
[0057] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0058] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.
[0059] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.
[0060] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.
[0061] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.
[0062] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0063] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
[0064] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A method for predicting turbine transmission failure in the respiratory module of a life support system, characterized in that, Includes the following steps: S1: Collect the real-time rotational speed signal of the turbine of the breathing module under different particle concentration conditions and perform signal filtering to obtain the turbine speed filtered signal; S2: Based on the turbine speed filtering signal, extract the turbine speed fluctuation amplitude feature and fluctuation frequency domain feature respectively, and generate speed amplitude feature data and frequency domain feature data; S3: Use rotational speed amplitude characteristic data to identify nonlinear velocity fluctuations caused by micro-deformation of turbine blades and generate turbine nonlinear fluctuation identification results; S4: Utilize frequency domain feature data to perform spectrum analysis of turbine speed signals, determine the structural correlation characteristics of turbine speed fluctuations, and output the structural feature analysis results; S5: Based on the turbine nonlinear fluctuation identification results and structural feature analysis results, establish a turbine abnormal fluctuation model and output the turbine abnormal fluctuation assessment results; S6: Predict the turbo gearbox failure trend based on the turbo abnormal fluctuation assessment results, and output a turbo gearbox failure early warning signal.
2. The method for predicting turbine transmission failure in a respiratory module of a life support system according to claim 1, characterized in that, S1 includes: A speed sensor is used to collect the turbine speed signal of the breathing module turbine under different particle concentration conditions in real time; The real-time turbine speed signal is filtered using a signal filter to remove high-frequency interference noise, resulting in a filtered turbine speed signal.
3. The method for predicting turbine transmission failure in a respiratory module of a life support system according to claim 2, characterized in that, S2 includes: A time-domain analysis of the turbine speed filter signal is performed based on the amplitude statistical analysis method to extract the amplitude characteristics of turbine speed fluctuations and obtain speed amplitude characteristic data. Frequency domain analysis of the turbine speed filter signal is performed based on the Fast Fourier Transform method to extract the frequency domain features of turbine speed fluctuations and obtain frequency domain feature data.
4. The method for predicting turbine transmission failure in a respiratory module of a life support system according to claim 3, characterized in that, S3 includes: A preset threshold for identifying nonlinear speed fluctuations is established, and threshold comparisons are performed based on rotational speed amplitude characteristic data. Within a preset sliding time window, trend analysis is performed on the speed amplitude characteristic data to obtain the slope of the data curve; When the slope of the curve exceeds the nonlinear velocity fluctuation identification threshold and the speed amplitude characteristic data shows a sudden change, the nonlinear velocity fluctuation caused by the micro-deformation of the turbine blade is determined, and the turbine nonlinear fluctuation identification result is output.
5. The method for predicting turbine transmission failure in a respiratory module of a life support system according to claim 4, characterized in that, S4 includes: The main frequency range of turbine speed fluctuation is determined based on frequency domain feature data, and the corresponding spectral amplitude data of the main frequency range is extracted. Perform a peak value extraction operation on the spectral amplitude data to obtain the spectral peak value corresponding to the turbine speed fluctuation; By comparing the spectral peaks using preset structural correlation feature judgment criteria, the structural correlation features of turbine speed fluctuations are determined, and the structural feature analysis results are output.
6. The method for predicting turbine transmission failure in a respiratory module of a life support system according to claim 5, characterized in that, S5 includes: A model for abnormal turbine fluctuations was constructed based on the results of turbine nonlinear fluctuation identification and structural feature analysis. By fusing the turbine nonlinear fluctuation identification results and structural feature analysis results using a turbine abnormal fluctuation model, the correlation between turbine speed fluctuation and turbine blade micro-deformation is determined. The degree of abnormal fluctuation is assessed based on the correlation between turbine speed fluctuation and turbine blade micro-deformation, and the turbine abnormal fluctuation assessment results are output.
7. The method for predicting turbine transmission failure in a respiratory module of a life support system according to claim 6, characterized in that, S6 includes: Preset turbo transmission fault warning threshold; Compare the results of the abnormal fluctuation assessment of the turbocharger with the turbocharger transmission fault warning threshold; When the abnormal fluctuation assessment result of the turbine reaches or exceeds the turbine transmission failure warning threshold, it is determined that there is a risk of transmission failure in the turbine, and a turbine transmission failure warning signal is output.