An Automatic Abnormality Detection Method for a High-Speed Rotating Transmission Device

Abstract

The present invention relates to the technical field of fault diagnosis, and specifically to an automatic anomaly detection method for a high-speed rotating transmission device, which includes the following steps: obtaining the angular acceleration, moment of inertia, and strain data of the high-speed rotating transmission device, calculating the transient strain gradient and matching the rotational speed change rate, adjusting the strain deviation range according to the change of the moment of inertia, screening the strain mutation points, and obtaining the rotational strain matching result. In the present invention, by setting the dynamic strain response interval and adjusting the strain deviation range, screening the strain mutation points, and monitoring the frequency response of the device in real time and comparing it with the normal frequency, the continuous monitoring of the frequency deviation rate is made more accurate, and they jointly act on identifying and clustering abnormal features, improving the accuracy and real-time performance of fault diagnosis. By calculating the deviation degree between the abnormal features and the normal trajectory, the anomaly detection threshold can be adjusted to more effectively identify and process potential faults, reduce the maintenance cost of the high-speed rotating transmission device, and improve the operation efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of fault diagnosis, and particularly to an automatic anomaly detection method for a high-speed rotating transmission device. Background Art

[0002] The technical field of fault diagnosis includes methods and technologies for identifying, analyzing, and processing anomalies or faults that occur during the operation of mechanical equipment and systems. This field focuses on capturing device status data through various sensors and data acquisition technologies, and using analysis models to predict and diagnose potential mechanical faults. The core contents include data collection, feature recognition, condition monitoring, fault prediction, and health management. By systematically integrating technologies, fault diagnosis can help improve the reliability and maintenance efficiency of equipment, reduce unplanned downtime, and optimize maintenance plans.

[0003] Among them, the automatic anomaly detection method for a high-speed rotating transmission device refers to using automation technology to detect abnormal behaviors that occur during the operation of a high-speed rotating transmission device. The technical matters targeted by this technical theme cover the establishment of a data acquisition system, the analysis and processing of multi-dimensional parameters, and the automatic adjustment of process parameters. Specifically, operation data is collected through sensors installed on the device, graph neural networks are used to process the data to identify abnormal patterns, and reinforcement learning is used to optimize and adjust the process parameters during the production process. The combined action of these means realizes the real-time monitoring and automatic adjustment of the state of the transmission device.

[0004] Existing fault diagnosis technologies rely on traditional data collection and processing methods, which show their limitations when dealing with high-speed rotating equipment. Existing technologies cannot provide sufficient sensitivity and response speed when capturing complex dynamic changes and instantaneous abnormal behaviors. This results in the fact that in actual operation, when the equipment encounters atypical faults or sudden anomalies, the fault diagnosis system cannot respond and adjust in a timely and accurate manner, leading to premature wear or faults of mechanical equipment, increasing the operating costs of enterprises and the risk of production interruption. Existing technologies lack adaptability and flexibility to fault patterns in a dynamic environment and are difficult to achieve health management throughout the life cycle of the equipment, which is particularly obvious in high-demand industrial applications. Summary of the Invention

[0005] To address the limitations exhibited by existing technologies when dealing with high-speed rotating equipment, namely, the inability of existing technologies to provide sufficient sensitivity and response speed in capturing complex dynamic changes and instantaneous abnormal behaviors, resulting in the failure of the fault diagnosis system to respond and adjust in a timely and accurate manner during actual operation when the equipment encounters atypical faults or sudden abnormalities, leading to premature wear or failure of mechanical equipment, increasing the enterprise's operating costs and the risk of production interruption; and the lack of adaptability and flexibility of existing technologies to fault patterns in a dynamic environment, making it difficult to achieve the technical problem of health management throughout the life cycle of the equipment. The present invention provides an automatic abnormal detection method for a high-speed rotating transmission device. The technical solution is as follows:

[0006] On the one hand, an automatic abnormal detection method for a high-speed rotating transmission device is provided, and the method includes:

[0007] S1: Obtain the angular acceleration, moment of inertia, and strain data of the high-speed rotating transmission device, calculate the transient strain gradient and match the rotational speed change rate, adjust the strain deviation range according to the change of the moment of inertia, screen the strain mutation points, and obtain the rotational strain matching result;

[0008] S2: Use the rotational strain matching result to record the real-time frequency response of the high-speed rotating transmission device, continuously monitor to obtain the frequency offset rate sequence, and refer to the vibration acceleration, impact signal, and transient stress to identify the deviation of the abnormal feature points, and obtain the rotational frequency offset identification result;

[0009] S3: Utilize the rotational frequency offset identification result, combine the vibration acceleration, transient stress, and impact signal, analyze the deviation degree between the abnormal feature points and the normal frequency trajectory, identify the local density distribution, screen the continuous offset abnormality and sudden abnormality, and obtain the abnormal feature clustering result;

[0010] S4: Invoke the abnormal feature clustering result, calculate the fitting residual change rate of the continuous offset abnormality, set the abnormal level according to the change rate, screen the abnormal feature points that meet the set range, obtain the abnormal mode matching result, automatically monitor the abnormal signal of the high-speed rotating transmission device, and give an early warning.

[0011] As a further solution of the present invention, the rotational strain matching result includes the transient strain gradient threshold, strain response interval, strain deviation range, and strain mutation situation, the rotational frequency offset identification result includes the frequency response ratio, frequency offset rate threshold, and deviation amplitude of the feature points, and the abnormal feature clustering result includes the trajectory deviation degree, local density distribution feature, and continuous offset duration threshold.

[0012] As a further solution of the present invention, the specific steps for obtaining the rotational strain matching result are as follows:

[0013] S101: Obtain the angular acceleration, moment of inertia, and strain data of the high-speed rotating transmission device, calculate the change in the moment of inertia, calculate the instantaneous increment of the angular acceleration based on the change in the moment of inertia, match the instantaneous increment of the angular acceleration with the strain data, analyze the degree of change in the transient strain, and obtain the transient strain matching data;

[0014] S102: Based on the transient strain matching data, identify the time change trend of the moment of inertia, analyze the distribution of the moment of inertia at different time points, set the reference range for the change in the moment of inertia, screen the transient strain data that meet the conditions, and obtain the dynamic strain range;

[0015] S103: Call the dynamic strain range, analyze the deviation degree of the strain data, combine with the change trend of the moment of inertia, screen the mutation points, and compare with the instantaneous increment of the angular acceleration to obtain the rotational strain matching result.

[0016] As a further solution of the present invention, the steps for obtaining the rotational frequency deviation identification result are specifically as follows:

[0017] S201: Call the rotational strain matching result, record the real-time frequency response of the high-speed rotating transmission device, compare the real-time operating frequency with the normal frequency, and obtain the frequency deviation identification result;

[0018] S202: According to the frequency deviation identification result, continuously monitor the ratio data, calculate the frequency change amplitude according to the time series, obtain the frequency deviation rate sequence, call the frequency deviation rate sequence, identify the dynamic range according to the frequency change trend, match and analyze the dynamic range data with the frequency deviation rate, evaluate the corresponding relationship between the frequency change range and time, screen the data points that meet the change range, and obtain the frequency deviation dynamic range;

[0019] S203: Use the frequency deviation dynamic range, refer to the vibration acceleration, impact signal, and transient stress, analyze the deviation of the abnormal characteristic points, screen the frequency deviation data points, calculate the frequency deviation characteristic value, and compare with the corresponding transient stress fluctuation to obtain the rotational frequency deviation identification result.

[0020] As a further solution of the present invention, the formula for calculating the frequency deviation characteristic value is as follows:

[0021] ;

[0022] Wherein, represents the frequency deviation characteristic value, represents the frequency value of the th data point, represents the average frequency value of the data points, represents the The vibration acceleration amplitude corresponding to the data point is Represents the total number of data points.

[0023] As a further solution of the present invention, the step of obtaining the abnormal feature clustering result is specifically as follows:

[0024] S301: using the rotation frequency offset identification result, combined with vibration acceleration, transient stress and impact signal, identifying the distribution of abnormal feature points, comparing the abnormal feature points with the normal frequency trajectory, identifying the degree of deviation between the two, and obtaining frequency trajectory deviation data;

[0025] S302: calling the frequency trajectory deviation data, identifying the local density distribution according to the degree of deviation, calculating the number of abnormal feature points, identifying the density change in the local area, comparing the density fluctuation amplitude under the time series, and obtaining the local density distribution characteristics;

[0026] S303: Analyze the temporal variation trend of local density by using the local density distribution characteristics, screen out continuous deviation anomalies and sudden anomalies, classify abnormal feature points according to the amplitude and stability of density variation, and classify anomaly categories to obtain abnormal feature clustering results.

[0027] As a further solution of the present invention, the formula for calculating the number of abnormal feature points is as follows:

[0028] ;

[0029] in, Represents the number of abnormal feature points, Representative The deviation of feature points, Represents the average value of the deviation of feature points, Represents the standard deviation of the feature point deviation, Represents the total number of feature points in the differentiated area, Representative The feature point and The associated weight of the reference point, Representative The feature point and The influence factor of the reference point, Represents the total number of reference points.

[0030] As a further solution of the present invention, the step of obtaining the abnormal pattern matching result is specifically:

[0031] S401: extracting the data sequence of the continuous shift anomaly based on the abnormal feature clustering result, calculating the fitting residuals of the differentiated time points, comparing the residual difference values of adjacent time points, and obtaining the continuous shift residual change rate;

[0032] S402: Invoke the continuous offset residual change rate, analyze the anomaly level based on the change rate distribution, classify the change rate according to numerical intervals, set the level criteria, mark the corresponding levels of the anomaly points, and obtain the anomaly level classification list;

[0033] S403: Use the anomaly level classification list, for sudden anomalies, calculate the instantaneous signal change amplitude of the mutation factor, compare the change amplitude with the set range, screen the anomaly feature points that meet the range, and obtain the anomaly pattern matching result.

[0034] As a further solution of the present invention, the method further includes step S5:

[0035] S5: Based on the anomaly pattern matching result, calculate the difference between the real-time frequency offset rate and the normal trajectory, invoke the difference to adjust the anomaly detection threshold, calculate the proportion of risk feature points according to the distribution of anomaly feature points, obtain the abnormal rotation state evaluation result, and adjust the signal acquisition weight and data sampling frequency;

[0036] The abnormal rotation state evaluation result includes the anomaly detection threshold, the proportion of risk feature points, and the frequency offset risk index.

[0037] As a further solution of the present invention, the steps for obtaining the abnormal rotation state evaluation result are specifically as follows:

[0038] S501: Invoke the anomaly pattern matching result, calculate the real-time frequency offset rate, extract the frequency data of the normal trajectory, compare the values of the two, calculate the difference between the real-time frequency offset rate and the normal trajectory, and obtain the frequency offset difference data;

[0039] S502: Use the frequency offset difference data to adjust the anomaly detection threshold according to the numerical range, compare the frequency offset situations at different differential time points, calculate the proportion of anomaly feature points, identify the distribution of anomaly feature points, obtain the abnormal rotation state evaluation result, and adjust the signal acquisition weight.

[0040] The beneficial effects brought by the technical solution provided by the embodiments of the present invention at least include:

[0041] By carefully acquiring and analyzing the angular acceleration, rotational inertia, and strain data of a high-speed rotating transmission device, precise calculation of the transient strain gradient and matching of the rotational speed change rate are achieved. The data-driven method optimizes the setting of the dynamic strain response interval and the adjustment of the strain deviation range, effectively screening for strain mutation points. The frequency response of the device is monitored in real time and compared with the normal frequency, making the continuous monitoring of the frequency deviation rate more accurate. These jointly act on identifying and clustering abnormal features, further improving the accuracy and real-time performance of fault diagnosis. By calculating the deviation degree between the abnormal features and the normal trajectory, the abnormal detection threshold can be adjusted to more effectively identify and handle potential faults, reducing the maintenance cost of the high-speed rotating transmission device and improving the operation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a schematic diagram of the working process of the present invention; DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] The technical solutions in the present invention will be described below with reference to the accompanying drawings.

[0044] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to represent examples, illustrations, or explanations. Any embodiment or design solution described as an "example" in the present invention should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Exactly speaking, the use of the word "example" is intended to present concepts in a specific manner. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either one of the two.

[0045] To make the technical problems, technical solutions, and advantages to be solved by the present invention clearer, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

[0046] Please refer to Figure 1 , the embodiments of the present invention provide an automatic abnormal detection method for a high-speed rotating transmission device. The processing flow of this method can include the following steps:

[0047] S1: Acquire the angular acceleration, rotational inertia, and strain data of the high-speed rotating transmission device, calculate the transient strain gradient and match the rotational speed change rate, set the dynamic strain response interval, adjust the strain deviation range according to the change of the rotational inertia, screen for strain mutation points, and obtain the rotational strain matching result;

[0048] S2: Adopt the rotational strain matching result, record the real-time frequency response of the high-speed rotating transmission device, calculate the ratio of the real-time operating frequency to the normal frequency, call the ratio for continuous monitoring to obtain the frequency deviation rate sequence, call the dynamic range of the frequency deviation rate, and refer to the vibration acceleration, impact signal, and transient stress to identify the deviation of abnormal feature points, and obtain the rotational frequency deviation identification result;

[0049] S3: Using the recognition result of rotational frequency offset, combined with vibration acceleration, transient stress, and impact signals, analyze the abnormal feature points and the deviation degree of the normal frequency trajectory, identify the local density distribution, call the change characteristics of the local density to screen continuous offset anomalies and sudden anomalies, and obtain the abnormal feature clustering result;

[0050] S4: Call the abnormal feature clustering result, calculate the fitting residual change rate of the continuous offset anomaly, set the anomaly level according to the change rate, call the mutation factor of the sudden anomaly to calculate the instantaneous signal change amplitude, screen the abnormal feature points that meet the set range, and obtain the abnormal pattern matching result;

[0051] S5: Based on the abnormal pattern matching result, calculate the difference between the real-time frequency offset rate and the normal trajectory, call the difference to adjust the anomaly detection threshold, calculate the proportion of risk feature points according to the distribution of abnormal feature points, obtain the abnormal evaluation result of the rotation state, and adjust the signal acquisition weight and data sampling frequency;

[0052] The rotation strain matching result includes the transient strain gradient threshold, strain response interval, strain deviation range, strain mutation situation. The rotation frequency offset recognition result includes the frequency response ratio, frequency offset rate threshold, and deviation amplitude of feature points. The abnormal feature clustering result includes the trajectory deviation degree, local density distribution feature, and continuous offset duration threshold. The abnormal pattern matching result includes the continuous offset anomaly level, sudden anomaly amplitude, and credibility of abnormal feature points.

[0053] The specific steps for obtaining the rotation strain matching result are as follows:

[0054] S101: Obtain the angular acceleration, moment of inertia, and strain data of the high-speed rotating transmission device, calculate the change amount of the moment of inertia, calculate the instantaneous increment of the angular acceleration according to the change amount of the moment of inertia, match the instantaneous increment of the angular acceleration with the strain data, analyze the change degree of the transient strain, and obtain the transient strain matching data;

[0055] Obtain the angular acceleration, moment of inertia, and strain data of a high-speed rotating transmission device. The angular acceleration is obtained using a gyroscope sensor, which is installed on the rotating component and set to an appropriate sampling frequency (e.g., 10 kHz) to ensure that high-frequency signals are not lost. The data is transmitted to the computing unit in real-time through a data acquisition card for storage and processing. The moment of inertia can be obtained using a dynamic testing method, where the torque on the drive shaft is measured by a high-precision torque sensor and combined with the angular acceleration to calculate the moment of inertia. The strain data is generally obtained using strain gauges, which are pasted on key parts such as bearing seats, gear meshing areas, or vulnerable parts under stress. The resistance change is measured through a Wheatstone bridge and converted into a strain value. The change in the moment of inertia is calculated from the difference in inertia between adjacent time points, and a sliding window smoothing process is used to reduce noise interference. The instantaneous increment of the angular acceleration is calculated based on the time difference method. When matching the angular acceleration increment with the strain data, the data needs to be processed, including detrending, filtering, and normalization, to ensure comparability between different physical quantities. The matching algorithm can use the dynamic time warping (DTW) method, which calculates the optimal matching path between the instantaneous increment sequence of the angular acceleration and the strain sequence and calculates the similarity score based on the matching path. The matching similarity threshold is set to 0.8, and this value is set based on the statistics of batch experimental data. The specific calculation method is to take the historical matching score distribution of a certain device under different working conditions, calculate its mean and standard deviation, and set as the lower limit of the matching reference for this device, that is: ;

[0056] Set the mean and standard deviation in the matching score dataset, then the matching similarity threshold is taken as 0.8, and only the pairs of angular acceleration increment and strain data with a matching degree exceeding 0.8 are retained to obtain the transient strain matching data.

[0057] S102: Based on the transient strain matching data, identify the time variation trend of the moment of inertia, analyze the distribution of the moment of inertia at different time points, set the reference range for the change in the moment of inertia, and screen the transient strain data that meets the conditions to obtain the dynamic strain interval;

[0058] To calculate the time variation trend of the moment of inertia, it is necessary to analyze the distribution of the moment of inertia over time, which can be achieved by statistically analyzing the inertia values at different time points. The processing of inertia data uses regression analysis, and polynomial fitting or moving average is set to eliminate high-frequency fluctuations in the data. During the analysis process, the time window technique can be adopted, that is, calculate the mean and standard deviation of the inertia change within a fixed time interval (such as 100 ms), and classify the inertia data. For example, 0 - 1000 rpm is the low-speed range, 1000 - 5000 rpm is the medium-speed range, and above 5000 rpm is the high-speed range. Calculate the rate of change of inertia within each range to identify outliers. The reference range of inertia change is set as:

[0059] ;

[0060] where, is the mean inertia of the device in the corresponding rotational speed range, is the standard deviation of the inertia within this range;

[0061] Set a certain device in the medium-speed range of 1000 - 5000 rpm, with its mean inertia , and standard deviation , then the reference range of inertia change is calculated as follows:

[0062] ;

[0063] Data exceeding this range will be regarded as outliers. When screening transient strain data that meets the conditions, the sliding window method is used for local statistics. The window size can be set according to the rotational speed characteristics of the device. For a system with a rotational speed exceeding 5000 rpm, the sliding window size is set to 50 ms, and for a system below 1000 rpm, it is set to 200 ms. This setting is based on ensuring that each window contains at least 10 data points to ensure statistical stability and screen out the dynamic strain range.

[0064] S103: Invoke the dynamic strain range, analyze the deviation degree of the strain data, combine with the change trend of the moment of inertia, screen out the mutation points, and compare with the instantaneous increment of the angular acceleration to obtain the rotational strain matching result;

[0065] It is necessary to analyze the deviation degree of the strain data. The deviation of the strain data is measured by the relative change rate, and the calculation formula is as follows:

[0066] ;

[0067] where, is the strain value at the current time point, is the reference strain value, taking the mean of the previous time window;

[0068] Set the average value of the strain measured within a certain time window to 200, and the strain value at the current time point is 230. Then the deviation degree is calculated as follows:

[0069] ;

[0070] The deviation threshold is set to 10%. The basis for setting this value is the statistical range of deviation fluctuations of the device under stable operating conditions. The calculation method is:

[0071] ;

[0072] Among them, is the number of data points under normal operating conditions;

[0073] Set among 10,000 data points under normal operating conditions, and calculate , then the threshold is:

[0074] ;

[0075] Combined with the change trend of the moment of inertia to screen for mutation points. The judgment index of mutation points can be set as the moment of inertia change rate:

[0076] ;

[0077] If the moment of inertia change rate at a certain time point exceeds the set threshold , then it can be judged that this point is a mutation point. The value is calculated based on the 95% confidence interval of the change trend of the transmission device's moment of inertia and compared with the instantaneous increment of angular acceleration. Set the angular acceleration increments before and after the mutation point to 200 and 240 respectively. Then the increment change rate is calculated as follows:

[0078] ;

[0079] Set the instantaneous increment change threshold to 15%. The calculation method of this value is based on the distribution of angular acceleration change rate under normal operating conditions, and take as the threshold. Set , , then is:

[0080] ;

[0081] If the calculation result exceeds 15%, it is determined that significant deformation occurs at this time point, and the rotational strain matching result is obtained.

[0082] The specific steps for obtaining the rotational frequency deviation identification result are as follows:

[0083] S201: Call the rotational strain matching result, record the real-time frequency response of the high-speed rotating transmission device, compare the real-time operating frequency with the normal frequency, and obtain the frequency deviation identification result;

[0084] Record the real-time frequency response of the high-speed rotating transmission device. The real-time operating frequency is obtained by using a high-precision rotational speed sensor, such as an optical encoder or a laser velocimeter. The installation position of the sensor is generally set at the shaft end or the gear meshing position to ensure the measurement accuracy. The data acquisition system records the frequency change at a sampling rate of 1 kHz or higher. When comparing the real-time operating frequency with the normal frequency, it is necessary to set the normal operating frequency reference value. The setting of this value is based on the designed rated rotational speed of the equipment and empirical statistical data. The specific method is to statistically analyze the frequency data of the equipment under normal operating conditions and take its average value as the reference value and set as the normal operating range. Set the rated rotational speed of a certain equipment to 3000 rpm, and the historical operating data statistics show rpm, rpm. Then the normal frequency range is set to rpm. If the real-time operating frequency exceeds this range, it is considered that there is a frequency deviation. The frequency deviation is calculated using the relative deviation formula:

[0085] ;

[0086] where is the real-time operating frequency, is the normal operating frequency;

[0087] Set the real-time measured rotational speed to 3050 rpm, then the deviation calculation is as follows:

[0088] ;

[0089] If the set frequency deviation threshold is 2%, then this data point is marked as abnormal, and the frequency deviation identification result is obtained.

[0090] S202: According to the frequency deviation identification result, continuously monitor the ratio data, calculate the amplitude of frequency change according to the time series, obtain the frequency offset rate sequence, call the frequency offset rate sequence, identify the dynamic range based on the frequency change trend, perform matching analysis on the dynamic range data and the frequency offset rate, evaluate the corresponding relationship between the frequency change range and time, screen the data points that meet the change range, and obtain the dynamic range of frequency offset;

[0091] Continuously monitor the comparison value data, calculate the amplitude of frequency change using time series analysis method, and adopt the sliding window method for the calculation method. The size of each window is set to 500 ms, and the difference between the maximum and minimum frequencies within the window is calculated as the amplitude of change to obtain the frequency offset rate sequence. The calculation of the frequency offset rate adopts the normalization method, using the upper and lower limits of the normal frequency range as the benchmark, calculate the frequency change rate at the current moment, call the frequency offset rate sequence, and identify the dynamic range based on the frequency change trend. The dynamic range is set based on the 95% confidence interval of the offset rate, and the calculation formula is:

[0092] ;

[0093] Among them, and are the mean and standard deviation of the frequency offset rate respectively;

[0094] Set the frequency offset rate data of a certain device to be statistically obtained as , , then the dynamic range is calculated as follows:

[0095] ;

[0096] Match and analyze the dynamic range data with the frequency offset rate, calculate the change trend of the frequency offset rate within the time series using the sliding mean method, evaluate the corresponding relationship between the frequency change range and time, set the change range threshold to 2.5%, and this value is set based on the statistical maximum volatility of the device under different load conditions. Screen the data points that meet the change range to obtain the frequency offset dynamic range.

[0097] S203: Utilize the frequency offset dynamic range, refer to the vibration acceleration, impact signal and transient stress, analyze the deviation of the abnormal characteristic points, screen the frequency offset data points, calculate the frequency offset characteristic value, and compare the corresponding transient stress fluctuation situation to obtain the rotation frequency offset identification result;

[0098] The formula for calculating the frequency offset characteristic value is as follows:

[0099] ;

[0100] Among them, represents the frequency offset characteristic value, represents the th data point's frequency value, represents the average frequency value of the data points, represents the vibration acceleration amplitude corresponding to the th data point, represents the total number of data points;

[0101] Parameter meaning and formula calculation derivation process:

[0102] Frequency value Obtained by measurement with a high-precision spectrum analyzer, which performs Fourier transform on the vibration signals of rotating machinery, extracts the main frequency components and detects the offset components therein;

[0103] Mean frequency Calculated from all the collected frequency data points, and the calculation method is:

[0104] ;

[0105] Vibration acceleration amplitude Measured by an acceleration sensor, installed on the rotating component, and monitors the vibration signal in real time, with the unit of ; Total number of data points Depends on the acquisition duration and sampling rate. A high sampling rate above 10 kHz is adopted to ensure the measurement accuracy;

[0106] The calculation steps are as follows:

[0107] The measured frequency offset data points and the corresponding vibration acceleration amplitudes. Suppose 5 groups of data are collected ( ), and the data are as follows:

[0108] Hz, ;

[0109] Hz, ;

[0110] Hz, ;

[0111] Hz, ;

[0112] Hz, ;

[0113] Calculate the mean frequency:

[0114] ;

[0115] Calculate the absolute value of the offset of each data point and multiply it by the vibration acceleration amplitude:

[0116] ;

[0117] ;

[0118] ;

[0119] ;

[0120] ;

[0121] Summation operation:

[0122] ;

[0123] ;

[0124] ;

[0125] The result shows that in the current measurement data, the frequency offset eigenvalue weighted by the vibration acceleration amplitude is 1.08 Hz. This value reflects the overall characteristics of the rotational frequency offset and can be used to analyze the frequency fluctuations under abnormal conditions and further analyze in combination with transient stress data.

[0126] The specific steps for obtaining the abnormal feature clustering result are as follows:

[0127] S301: Using the rotational frequency offset identification result, combined with vibration acceleration, transient stress, and impact signals, identify the distribution of abnormal feature points, compare the abnormal feature points with the normal frequency trajectory, identify the deviation degree between the two, and obtain the frequency trajectory deviation data;

[0128] Combined with vibration acceleration, transient stress, and impact signals, analyze the abnormal feature points during the operation of the equipment. Collect the rotational frequency, vibration acceleration, transient stress, and impact signal data under the normal operation state of the equipment to establish a normal operation reference data set. During the operation of the equipment, monitor the rotational frequency in real time and compare it with the reference data. If the offset exceeds the set threshold, it indicates that the equipment is abnormal. Record the vibration signal during the operation of the equipment, perform data filtering and feature extraction on the vibration acceleration, calculate the vibration amplitude, average value, and peak value. When the average value or peak value of the vibration acceleration is significantly higher than the normal level, there is an abnormality. In the transient stress analysis, measure the stress condition of the equipment through strain gauges or pressure sensors, record the change curve of the transient stress, and judge whether the stress value exceeds the allowable range of the material. If there is a sudden change or excessive stress, the equipment faces a structural problem. Obtain the impact signal through an acceleration sensor or impact detector, calculate the impact amplitude and impact occurrence frequency, and compare with the normal impact mode. If the impact amplitude increases abnormally or the impact interval is significantly shortened, it indicates that the equipment is subjected to abnormal external forces. Mark all the detected abnormal feature points, compare the feature points with the frequency trajectory under the normal operation state, and calculate the deviation degree. If the deviation degree exceeds the set safety threshold, obtain the frequency trajectory deviation data.

[0129] S302: Invoke the call frequency trajectory deviation data, identify the local density distribution based on the deviation degree, calculate the number of abnormal feature points, identify the density change in the local area, and compare the density fluctuation amplitude in the time series to obtain the local density distribution characteristics;

[0130] The formula for calculating the number of abnormal feature points is as follows:

[0131] ;

[0132] Wherein, represents the number of abnormal feature points, represents the deviation degree of the th feature point, represents the average value of the deviation degrees of the feature points, represents the standard deviation of the deviation degrees of the feature points, represents the total number of feature points in the differential area, represents the th feature point and the th reference point's correlation weight, represents the th feature point and the th reference point's influence factor, represents the total number of reference points;

[0133] Parameter acquisition and calculation method:

[0134] represents the deviation degree of the th feature point, which is obtained by calculating the frequency trajectory deviation amount of the feature point within the current time window. The calculation method is:

[0135] ;

[0136] Wherein, represents the current frequency value of the th feature point, which is collected by a high-precision frequency monitoring device, represents the reference frequency value of this feature point, which is obtained as the mean value through long-term data statistics;

[0137] Set the feature point data collected in a certain area:

[0138] ;

[0139] Reference frequency value:

[0140] ;

[0141] Then:

[0142] ;

[0143] ;

[0144] ;

[0145] represents the average deviation degree of feature points:

[0146] ;

[0147] wherein, (number of feature points);

[0148] ;

[0149] represents the standard deviation of the deviation degree of feature points:

[0150] ;

[0151] Calculation:

[0152] ;

[0153] represents the th feature point and the th reference point's correlation weight, obtained by calculating the correlation coefficient between the change in feature point frequency and the reference point data, and set by data analysis calculation:

[0154] ;

[0155] represents the th feature point and the th reference point's influence factor, calculated by the distance attenuation function from the reference point to the feature point:

[0156] ;

[0157] wherein, represents the spatial distance from the feature point to the reference point, measured by the GIS system, and set:

[0158] ;

[0159] Then:

[0160] ;

[0161] ;

[0162] ;

[0163] Calculate the local density adjustment factor:

[0164] ;

[0165] Calculate each feature point:

[0166] ;

[0167] ;

[0168] ;

[0169] calculate :

[0170] ;

[0171] ;

[0172] The results show that the number of abnormal feature points is 2.138, which reflects the degree of density deviation of feature points within the statistical differentiation area and can be used for further density change analysis.

[0173] S303: Analyze the temporal variation trend of local density by using local density distribution characteristics, screen out continuous deviation anomalies and sudden anomalies, classify abnormal feature points according to the amplitude and stability of density variation, and classify anomaly categories to obtain abnormal feature clustering results;

[0174] Abnormal feature points are classified, and the judgment criteria for continuous offset anomalies and sudden anomalies are set. Under normal circumstances, the density of abnormal points in each grid unit of a local area of a certain device is stable at around 5%. When the density exceeds 10% and does not decrease for multiple time windows, it is defined as a continuous offset anomaly. If the density surges in a short period of time, such as a sudden increase from 5% to 20%, it is defined as a sudden anomaly. In industrial equipment fault monitoring, if the density of abnormal points remains at 12% for five consecutive time windows during the operation of a certain device, it means that the abnormality in this area is a stable offset type. If the density of abnormal points suddenly increases from 6% to 25% in a certain time window, it indicates that a sudden failure occurs in this area. The stability of density fluctuation is calculated, and the fluctuation threshold is set, such as calculating the density mean and standard deviation. If the standard deviation is small, it indicates that the density is stable, and if the standard deviation is large, it indicates that the density changes violently. Based on the abnormal category collection criteria, the abnormal feature points are classified. If the density offset of a certain area exceeds 5%, it is classified as category 1, and if the density growth rate exceeds 10%, it is classified as category 2, forming a clustering result of abnormal features.

[0175] The specific steps for obtaining the abnormal pattern matching results are as follows:

[0176] S401: based on the abnormal feature clustering results, extract the data sequence of the continuous shift anomaly, calculate the fitting residuals of the differentiated time points, compare the residual differences of adjacent time points, and obtain the continuous shift residual change rate;

[0177] Extract the data series of continuous deviation anomalies, filter the classified anomaly data, arrange them in chronological order, form time series data, ensure the consistency of time intervals, and avoid calculation errors caused by missing data. In this process, if missing data points appear, use linear interpolation or historical mean filling to ensure data integrity. Select a suitable fitting method to perform trend fitting on the time series, such as using the least squares method for linear fitting or polynomial fitting to improve the trend matching accuracy. After obtaining the fitting function, calculate the fitting residual at each time point. The fitting residual is defined as the difference between the actual measured value and the fitting value. Record the residual data at each time point, and calculate the residual difference of adjacent time points to determine the trend of residual change. When the residual difference is small, it indicates that the abnormal change is relatively stable, and when the residual difference is large, It indicates that the abnormal change is more drastic. The threshold of the change rate of the residual difference is set. The threshold is set according to the residual change range during the normal operation period. The specific method is to select the residual change rate during the stable operation period of the equipment, calculate the mean and standard deviation, and set the threshold to the mean plus twice the standard deviation to ensure the sensitivity of anomaly detection and avoid false alarms. It is set that the residual change rate of a certain equipment is 0.05 and the standard deviation is 0.02 in the stable operation stage, then the threshold is set to 0.05+2×0.02=0.09. When the residual change rate exceeds 0.09, it is determined that the equipment has a continuous offset abnormality. If the equipment type or operating conditions are different, the standard deviation multiple can be adjusted according to the actual measurement data. By comparing the change rates in multiple time windows and analyzing the continuous changes of the abnormal points, the continuous offset residual change rate is obtained.

[0178] S402: calling the continuous offset residual change rate, analyzing the abnormality level according to the change rate distribution, classifying the change rate according to the numerical range, setting the level standard, marking the corresponding level of the abnormal point, and obtaining the abnormality level classification list;

[0179] Analyze the abnormal level based on the distribution of the change rate, conduct statistical analysis on the calculated change rate data, obtain the maximum value, minimum value, mean value, and standard deviation of the change rate, set the numerical interval of the change rate, divide it into multiple levels, such as setting three levels: low, medium, and high. The classification standard can be based on the operating requirements and safety range of the equipment. The specific method is to calculate the distribution of the change rate in the abnormal dataset. Set the low-level range as less than the mean value of the change rate minus one standard deviation, the medium-level range as from the mean value minus one standard deviation to the mean value plus one standard deviation, and the high-level range as the part greater than the mean value plus one standard deviation. The change rate data distribution of a certain mechanical equipment is a mean value of 0.06 and a standard deviation of 0.015. Then the low-level range is the change rate < 0.045, the medium-level range is 0.045 ≤ change rate ≤ 0.075, and the high-level range is the change rate > 0.075. After completing the classification, mark the abnormal points and count the number of abnormal points in each level to form a list of abnormal level classifications. If the operating conditions of the equipment are different, such as the change rate of the vibration signal of fan and pump equipment is low, adjust the standard deviation multiple to between 1.5 - 2 to ensure the accuracy of the abnormal level classification, which can intuitively display the distribution of abnormal points in different levels and provide basic data support for further abnormal analysis to obtain the list of abnormal level classifications.

[0180] S403: Adopt the list of abnormal level classifications. For sudden abnormalities, calculate the instantaneous signal change amplitude of the mutation factor, compare the change amplitude with the set range, and screen the abnormal feature points that meet the range to obtain the abnormal mode matching result;

[0181] For sudden abnormalities, calculate the instantaneous signal change amplitude of the mutation factor. Select the data of the sudden abnormal points to obtain the instantaneous signal, such as vibration signal, current signal, or temperature signal, and calculate the change amplitude of the signal. Define the change amplitude as the difference between the signal value at the current moment and the signal value at the previous moment. For example, for a certain equipment at the moment, the vibration amplitude is mm / s, and at the moment, the vibration amplitude is mm / s. Then the change amplitude is calculated as:

[0182] ;

[0183] Calculate the instantaneous signal change amplitude for all time points, compare the change amplitude with the set range, and the setting of this range is based on the normal fluctuation range of the device. The specific calculation method is to obtain the signal change amplitude data during the normal operation period of the device, calculate the mean value and standard deviation, and set the mutation judgment threshold as the mean value plus 2 - 3 times the standard deviation. When a certain mechanical device is operating normally, the mean value of the vibration signal change amplitude is 1.2 mm / s, and the standard deviation is 0.5 mm / s, then the set threshold is 1.2 + 2×0.5 = 2.2 mm / s. When the change amplitude exceeds 2.2 mm / s, it is determined as a mutation anomaly. If the working condition of a certain device fluctuates greatly, the threshold range can be adjusted to 2.5 - 3 times the standard deviation to reduce the false alarm rate. For example, if the above calculation result of 3.3 mm / s is greater than the threshold of 2.2 mm / s, then this point is determined as a sudden anomaly. Screen all abnormal feature points that meet the range to obtain the abnormal pattern matching result.

[0184] The steps for obtaining the evaluation result of the abnormal rotation state are specifically as follows:

[0185] S501: Call the abnormal pattern matching result, calculate the real-time frequency offset rate, extract the frequency data of the normal trajectory, compare the values of the two, calculate the difference between the real-time frequency offset rate and the normal trajectory, and obtain the frequency offset difference data;

[0186] Calculate the real-time frequency offset rate, obtain the current operating data of the device, extract the real-time frequency data, and calculate the degree of offset relative to the reference frequency. During this process, ensure that the time interval of data acquisition is consistent to avoid the influence of data imbalance on the calculation result. At the same time, extract the frequency data of the normal operating trajectory of the device, establish a reference frequency curve, compare the real-time frequency offset data with the reference trajectory, and analyze the offset trend at different time points. If it is found that the frequency offset is persistent or sudden, further mark the abnormal moment, and calculate the change of the real-time offset rate. During the comparison process, use a sliding time window for data smoothing processing to reduce the influence of sudden interference on the calculation result and ensure the continuity of the data trend. After obtaining the offset rate, calculate its numerical difference from the normal trajectory to ensure the rationality of the offset rate, and form the frequency offset difference data.

[0187] S502: Use the frequency offset difference data to adjust the abnormal detection threshold according to the numerical range, compare the frequency offset situations at different differential time points, calculate the proportion of abnormal feature points, identify the distribution of abnormal feature points, obtain the evaluation result of the abnormal rotation state, and adjust the signal acquisition weight;

[0188] Adjust the anomaly detection threshold according to the numerical range, conduct statistical analysis on the frequency offset difference data, extract the maximum value, minimum value and distribution interval, set the strategy for adjusting the threshold according to different working conditions. If the offset difference is within a certain fixed range, the anomaly detection threshold can be appropriately increased or decreased to avoid false positives or missed detections. Compare the frequency offset situations at different time points, analyze the distribution law of anomaly points in the time series, calculate the proportion of anomaly feature points among all data points to judge the severity of the anomaly phenomenon, identify the distribution characteristics of anomaly feature points in the equipment operation cycle. If the anomaly points are concentrated in a specific time period, they are related to certain operating states. Through further data screening and adjustment, form the evaluation result of the rotation state anomaly, and adjust the acquisition weight for the signal acquisition process to improve the data acquisition accuracy of the key frequency bands and optimize the accuracy of subsequent anomaly detection.

[0189] The above are only the specific implementation manners of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. An automatic abnormal detection method for a high-speed rotating transmission device, characterized in that, It includes the following steps: S1: Obtain the angular acceleration, moment of inertia, and strain data of the high-speed rotating transmission device, calculate the transient strain gradient and match the rotational speed change rate, set the dynamic strain response interval, adjust the strain deviation range according to the change of the moment of inertia, screen the strain mutation points, and obtain the rotational strain matching result; S2: Adopt the rotational strain matching result, record the real-time frequency response of the high-speed rotating transmission device, calculate the ratio of the real-time operating frequency to the normal frequency, call the ratio to continuously monitor and obtain the frequency deviation rate sequence, and refer to the vibration acceleration, impact signal, and transient stress to identify the deviation of the abnormal feature points, and obtain the rotational frequency deviation identification result; S3: Utilize the rotational frequency deviation identification result, combine the vibration acceleration, transient stress, and impact signal, analyze the deviation degree between the abnormal feature points and the normal frequency trajectory, identify the local density distribution, screen the continuous deviation abnormality and sudden abnormality, and obtain the abnormal feature clustering result; S4: Call the abnormal feature clustering result, calculate the fitting residual change rate of the continuous deviation abnormality, set the abnormal level according to the fitting residual change rate, screen the abnormal feature points that meet the set range, obtain the abnormal mode matching result, automatically monitor the abnormal signal of the high-speed rotating transmission device, and issue a warning.

2. The automatic abnormal detection method for the high-speed rotating transmission device according to claim 1, characterized in that, The rotational strain matching result includes the transient strain gradient threshold, strain response interval, strain deviation range, and strain mutation situation. The rotational frequency deviation identification result includes the frequency response ratio, frequency deviation rate threshold, and deviation amplitude of the feature points. The abnormal feature clustering result includes the trajectory deviation degree, local density distribution feature, and continuous deviation duration threshold.

3. The automatic abnormality detection method for the high-speed rotating transmission device according to claim 1, characterized in that The specific steps for obtaining the rotational strain matching result are as follows: S101: Obtain the angular acceleration, moment of inertia, and strain data of the high-speed rotating transmission device, calculate the change amount of the moment of inertia, calculate the instantaneous increment of the angular acceleration according to the change amount of the moment of inertia, match the instantaneous increment of the angular acceleration with the strain data, analyze the change degree of the transient strain, and obtain the transient strain matching data; S102: Based on the transient strain matching data, identify the time change trend of the moment of inertia, analyze the distribution of the moment of inertia at different time points, set the reference range of the moment of inertia change, screen the transient strain data that meet the conditions, and obtain the dynamic strain interval; S103: Call the dynamic strain interval, analyze the deviation degree of the strain data, combine the change trend of the moment of inertia, screen the mutation points, and compare them with the instantaneous increment of the angular acceleration to obtain the rotational strain matching result.

4. The automatic abnormal detection method for the high-speed rotating transmission device according to claim 1, characterized in that The specific steps for obtaining the rotational frequency deviation identification result are as follows: S201: Call the rotational strain matching result, record the real-time frequency response of the high-speed rotating transmission device, compare the real-time operating frequency with the normal frequency, calculate the ratio of the real-time operating frequency to the normal frequency, and obtain the frequency deviation identification result; S202: According to the frequency deviation identification result, continuously monitor the comparison value data, calculate the frequency change amplitude according to the time series, obtain the frequency offset rate sequence, call the frequency offset rate sequence, identify the dynamic range according to the frequency change trend, match and analyze the dynamic range data with the frequency offset rate, evaluate the corresponding relationship between the frequency change range and time, select the data points that meet the change range, and obtain the frequency offset dynamic range; S203: Utilizing the frequency offset dynamic range, referring to vibration acceleration, impact signal and transient stress, analyzing the deviation of abnormal feature points, screening frequency offset data points, calculating frequency offset feature values, and comparing the corresponding transient stress fluctuations to obtain a rotation frequency offset identification result.

5. The automatic abnormal detection method of the high-speed rotating transmission device according to claim 4, characterized in that, The formula for calculating the frequency offset characteristic value is as follows: ； in, represents the frequency shift characteristic value, Representative The frequency value of the data point, represents the average frequency value of the data points, Representative The vibration acceleration amplitude corresponding to the data point is Represents the total number of data points.

6. The automatic abnormal detection method for the high-speed rotating transmission device according to claim 1, wherein The steps for obtaining the abnormal feature clustering result are specifically as follows: S301: using the rotation frequency offset identification result, combined with vibration acceleration, transient stress and impact signal, identifying the distribution of abnormal feature points, comparing the abnormal feature points with the normal frequency trajectory, identifying the degree of deviation between the two, and obtaining frequency trajectory deviation data; S302: calling the frequency trajectory deviation data, identifying the local density distribution according to the degree of deviation, calculating the number of abnormal feature points, identifying the density change in the local area, comparing the density fluctuation amplitude under the time series, and obtaining the local density distribution characteristics; S303: Analyze the temporal variation trend of local density by using the local density distribution characteristics, screen out continuous deviation anomalies and sudden anomalies, classify abnormal feature points according to the amplitude and stability of density variation, and classify anomaly categories to obtain abnormal feature clustering results.

7. The automatic abnormality detection method for the high-speed rotating transmission device according to claim 6, characterized in that, The formula for calculating the number of abnormal feature points is as follows: ； Among them, represents the number of abnormal feature points, represents the degree of deviation of the th feature point, represents the average value of the degrees of deviation of the feature points, represents the standard deviation of the degrees of deviation of the feature points, represents the total number of feature points within the differential region, represents the th feature point and the th reference point's correlation weight, represents the th feature point and the th reference point's influence factor, represents the total number of reference points.

8. The automatic abnormality detection method of the high-speed rotating transmission device according to claim 1, wherein The steps for obtaining the abnormal pattern matching result are specifically as follows: S401: extracting the data sequence of the continuous shift anomaly based on the abnormal feature clustering result, calculating the fitting residuals of the differentiated time points, comparing the residual differences of adjacent time points, and obtaining the fitting residual change rate of the continuous shift anomaly; S402: calling the fitting residual change rate of the continuous offset anomaly, analyzing the anomaly level according to the distribution of the fitting residual change rate of the continuous offset anomaly, classifying the fitting residual change rate of the continuous offset anomaly according to the numerical range, setting the level standard, marking the corresponding level of the abnormal point, and obtaining the abnormal level classification list; S403: using the abnormal level classification list, for sudden abnormalities, calculating the instantaneous signal change amplitude of the mutation factor, comparing the change amplitude with the set range, screening abnormal feature points that meet the range, and obtaining abnormal pattern matching results.

9. The automatic abnormal detection method for the high-speed rotating transmission device according to claim 1, characterized in that The method further comprises step S5: S5: Based on the abnormal pattern matching result, the difference between the real-time frequency deviation rate and the normal trajectory is calculated, the difference is called to adjust the abnormal detection threshold, the risk feature point ratio is calculated according to the abnormal feature point distribution, the rotation state abnormality assessment result is obtained, and the signal acquisition weight and data sampling frequency are adjusted; The rotation state abnormality assessment result includes an abnormality detection threshold, a risk feature point ratio, and a frequency offset risk index.

10. The automatic abnormal detection method for the high-speed rotating transmission device according to claim 9, characterized in that, The steps for obtaining the rotation state abnormality evaluation result are specifically as follows: S501: Call the abnormal pattern matching result, calculate the real-time frequency offset rate, extract the frequency data of the normal trajectory, compare the values of the two, calculate the difference between the real-time frequency offset rate and the normal trajectory, and obtain the frequency offset difference data; S502: Use the frequency offset difference data to adjust the abnormal detection threshold according to the numerical range, compare the frequency offset conditions at different time points, calculate the proportion of abnormal feature points, identify the distribution of abnormal feature points, obtain the abnormal evaluation result of the rotation state, and adjust the signal acquisition weight.

Images