Rotating machine fault intelligent diagnosis method and system fusing time-frequency domain characteristics
By integrating the time-frequency domain characteristics into the intelligent diagnosis method for rotating machinery faults, vibration frequency and temperature data are collected and calculated in real time to form a fault prediction factor, which solves the problem that rotating machinery faults cannot be predicted in the existing technology and realizes the early prediction and timely maintenance of rotating machinery faults.
Patent Information
- Application Number
- CN202510902384.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-01
- Publication Date
- 2025-09-19
AI Technical Summary
Existing rotating machinery fault diagnosis is unable to predict faults before they occur, resulting in the gradual process of the operating status evolving from potential defects to functional failure being overlooked, making it impossible to intervene in maintenance in advance, which in turn leads to operation interruption.
By fusing time-frequency domain features, collecting vibration frequency and temperature data in real time, calculating frequency fault severity parameters and temperature anomaly parameters, and forming a fault prediction factor, the system triggers an alarm only when vibration disorder and temperature anomaly trends occur simultaneously, thus realizing the prediction of rotating machinery faults.
It achieves early prediction of rotating machinery failures, avoids misjudgment due to interference with a single parameter, and ensures timely maintenance of rotating machinery before potential defects develop into functional failure, avoiding operation interruptions.
Smart Images
Figure CN120668378A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of fault diagnosis, and in particular to an intelligent fault diagnosis method and system for rotating machinery that integrates time-frequency domain features. Background Art
[0002] In rotating machinery fault diagnosis, it is necessary to integrate time-frequency domain features because the vibration, noise and other signals during the operation of rotating machinery are mostly non-stationary signals, and their fault characteristics are often reflected in the dynamic changes of time and frequency at the same time - time domain characteristics can reflect the time and instantaneous characteristics of the fault, and frequency domain characteristics can reveal the frequency components and energy concentration areas corresponding to the fault. The fusion of the two can fully capture the time-varying characteristics of the signal and effectively deal with problems such as weak early fault characteristics and strong noise interference under complex working conditions.
[0003] However, existing mechanical fault diagnosis cannot predict faults before they occur. The gradual process of rotating machinery's operating status developing from potential defects to functional failure is ignored, and maintenance cannot be intervened in advance, which leads to operation interruption. Summary of the Invention
[0004] Technical problems solved
[0005] In response to the deficiencies of the prior art, the present invention provides an intelligent fault diagnosis method and system for rotating machinery that integrates time-frequency domain features to solve the problem that the gradual process of the rotating machinery's operating status from potential defects to functional failure has been ignored.
[0006] Technical Solution
[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: an intelligent diagnosis method and system for rotating machinery faults that integrates time-frequency domain features, including the following specific steps and modules: Step 1: Start real-time collection of vibration frequency data and temperature data, preprocess the vibration frequency data and temperature data and perform comprehensive calculations to obtain frequency fault severity parameters and temperature anomaly parameters, perform comprehensive calculations on the frequency fault severity parameters and temperature anomaly parameters to obtain a fault prediction factor, wherein the fault prediction factor is used to diagnose whether the rotating machinery is faulty; Step 2: Diagnose whether the machinery is about to fail based on the fault prediction factor; Step 3: If it is diagnosed that the machinery is about to fail, an alarm is issued; if it is diagnosed that the machinery is normal, the diagnosis is ended.
[0008] Furthermore, the specific method of obtaining the fault prediction factor is as follows: comprehensively calculating the vibration frequency data to obtain the frequency fault severity parameter, comprehensively calculating the temperature data to obtain the temperature anomaly parameter, and comprehensively calculating the frequency fault severity parameter and the temperature anomaly parameter to obtain the fault prediction factor.
[0009] Furthermore, the specific method for obtaining the fault prediction factor is: ;in, Represents the fault prediction factor, which is a value used to predict whether the rotating machinery has a fault. Indicates the severity of frequency fault parameters, reflecting whether the vibration frequency of the rotating machinery fluctuates. Indicates the temperature abnormality parameter, reflecting whether the temperature of the rotating machinery is abnormal.
[0010] Furthermore, the specific method for obtaining the frequency fault severity parameter is as follows: performing variance calculation on the vibration frequency data to obtain a frequency fluctuation value, and in a time series, comparing the frequency fluctuation value of the next second with the frequency fluctuation value of the previous second, if the frequency fluctuation value of the next second is greater than the frequency fluctuation value of the previous second, then performing a difference calculation on the frequency fluctuation value of the next second and the frequency fluctuation value of the previous second to obtain a frequency fault severity value, because the frequency fluctuation value of the next second is greater than the frequency fluctuation value of the previous second to reflect that the vibration frequency of the rotating machinery tends to be disordered, that is, the greater the probability of a fault, if the frequency fluctuation value of the next second is less than or equal to the frequency fluctuation value of the previous second, then continuing to compare the frequency fluctuation value of the next second with the frequency fluctuation value of the previous second, performing a sum calculation on the frequency fault severity values, and obtaining a frequency fault severity parameter.
[0011] Furthermore, the specific method for obtaining the frequency fault severity parameter is: ;in, Indicates the frequency fault severity parameter, Indicates time, Indicates the frequency fluctuation value of the next second, Indicates the frequency fluctuation value of the last second.
[0012] Furthermore, the frequency fluctuation value is specifically obtained as follows: the vibration frequency data is summed and then averaged to obtain a frequency balance value, which is used as a standard for the vibration frequency to fluctuate around the frequency balance value; the square of the difference between each vibration frequency in the vibration frequency data and the frequency balance value is calculated to obtain the fluctuation value of each vibration frequency; the fluctuation value of each vibration frequency is summed to obtain the frequency fluctuation value.
[0013] Furthermore, the specific method for obtaining the temperature anomaly parameter is as follows: perform a comprehensive calculation on the temperature data to obtain a temperature anomaly value; in a time series, compare the temperature anomaly value of the next second with the temperature anomaly value of the previous second; if the temperature anomaly value of the next second is greater than the temperature anomaly value of the previous second, then perform a difference calculation between the temperature anomaly value of the next second and the temperature anomaly value of the previous second to obtain a temperature anomaly severity value; because as the temperature anomaly value gradually increases, the probability of mechanical failure increases; if the temperature anomaly value of the next second is less than or equal to the temperature anomaly value of the previous second, then continue the comparison, perform a sum calculation on the temperature anomaly severity value, and obtain the temperature anomaly parameter.
[0014] Furthermore, the specific method for obtaining the temperature anomaly parameter is: ;in, represents the temperature anomaly parameter, Indicates time, Indicates the abnormal temperature value of the next second. Indicates the abnormal temperature value of the previous second.
[0015] Furthermore, the specific method of obtaining the temperature anomaly value is as follows: the normal temperature during mechanical operation is averaged to obtain a standard temperature threshold. Since the temperature during mechanical operation increases due to mechanical failure, the actual temperature value in the temperature data and the standard temperature threshold are differenced to obtain the temperature anomaly value.
[0016] Furthermore, the system includes: a real-time data acquisition module, a historical data storage module, a mechanical fault analysis and prediction module, and an alarm module; the real-time data acquisition module is used to collect vibration frequency data and temperature data in real time and perform preprocessing; the historical data storage module is used to store historical normal operation data of rotating machinery; the mechanical fault analysis and prediction module combines real-time vibration frequency data and temperature data with historical normal operation data of rotating machinery for analysis and prediction, and if the analysis shows that a fault is about to occur, the prediction result is sent to the alarm module; the alarm module receives the prediction result and issues an alarm.
[0017] Beneficial effects
[0018] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects:
[0019] 1. For vibration frequency data, calculate the cumulative time series difference of frequency fluctuation values. The difference is accumulated only when the fluctuation value in the next second is greater than that in the previous second. This quantifies the trend of increasing vibration frequency disorder and predicts rotating machinery failures.
[0020] 2. For temperature data, calculate the cumulative time series difference of temperature anomaly values, track the continuous expansion trend of temperature anomalies, and further predict rotating machinery failures.
[0021] 3. The two form diagnostic conditions through a product model. An alarm is triggered only when vibration disturbances and abnormal temperature trends occur simultaneously, indicating that the rotating machinery used in the kitchen environment is about to fail. This avoids misjudgment caused by interference with a single parameter and prevents the gradual process of the rotating machinery's operating status from potential defects to functional failure from being overlooked, preventing early maintenance intervention and leading to operation interruption.
[0022] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] Figure 1 This is a flow chart of the present invention: a method for intelligent diagnosis of rotating machinery faults integrating time-frequency domain features.
[0024] Figure 2 This is the structural diagram of the present invention: an intelligent diagnosis system for rotating machinery faults that integrates time-frequency domain features. DETAILED DESCRIPTION
[0025] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.
[0026] It should be noted that, in this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "include," "comprise," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.
[0027] like Figure 1 As shown, an embodiment of the present invention provides an intelligent diagnosis method for rotating machinery faults integrating time-frequency domain features, which includes the following specific steps:
[0028] Step 1: When rotating machinery fails due to dynamic balancing caused by kitchen fumes and bearing corrosion caused by improper cleaning, the mechanical vibration frequency produces abnormal characteristic frequencies due to component wear or imbalance, and the mechanical temperature rises locally or overall due to increased friction, increased energy loss, or lubrication failure. Therefore, vibration sensors and temperature sensors are used to collect vibration frequency data and temperature data in real time in a time series. The temperature sensors are arranged around high-friction, high-load, and key energy conversion components of the rotating machinery, such as bearings, motors, gears, and couplings. Data cleaning of the vibration frequency data and temperature data helps to improve the data quality of the vibration frequency data and temperature data. The vibration frequency data and temperature data are standardized to eliminate the dimension and perform comprehensive calculations to obtain fault prediction factors.
[0029] The specific method of obtaining the fault prediction factor is as follows:
[0030] Perform comprehensive calculation on vibration frequency data to obtain frequency fault severity parameters, perform comprehensive calculation on temperature data to obtain temperature anomaly parameters, and perform comprehensive calculation on frequency fault severity parameters and temperature anomaly parameters to obtain fault prediction factors;
[0031] ;
[0032] in, Represents the fault prediction factor, which is a value used to predict whether the rotating machinery has a fault. Indicates the severity of frequency fault parameters, reflecting whether the vibration frequency of the rotating machinery fluctuates. Indicates the temperature abnormality parameter, reflecting whether the temperature of the rotating machinery is abnormal.
[0033] The specific method for obtaining the frequency fault severity parameters is as follows:
[0034] The variance of the vibration frequency data is calculated to obtain the frequency fluctuation value. In the time series, the frequency fluctuation value of the next second is compared with the frequency fluctuation value of the previous second. If the frequency fluctuation value of the next second is greater than the frequency fluctuation value of the previous second, the difference between the frequency fluctuation value of the next second and the frequency fluctuation value of the previous second is calculated to obtain the frequency fault severity value. Because the frequency fluctuation value of the next second is greater than the frequency fluctuation value of the previous second, it can reflect that the vibration frequency of the rotating machinery tends to be disordered, that is, the probability of failure is greater. If the frequency fluctuation value of the next second is less than or equal to the frequency fluctuation value of the previous second, the frequency fluctuation value of the next second is continued to be compared with the frequency fluctuation value of the previous second, and the frequency fault severity value is summed to obtain the frequency fault severity parameter;
[0035] ;
[0036] in, Indicates the frequency fault severity parameter, Indicates time, Indicates the frequency fluctuation value of the next second, Indicates the frequency fluctuation value of the last second.
[0037] The specific method of obtaining the frequency fluctuation value is as follows:
[0038] The vibration frequency data is summed and averaged to obtain a frequency balance value, which is used as a standard for the vibration frequency to fluctuate around the frequency balance value. The square of the difference between each vibration frequency in the vibration frequency data and the frequency balance value is calculated to obtain the fluctuation value of each vibration frequency. The fluctuation value of each vibration frequency is summed to obtain the frequency fluctuation value.
[0039] The specific method for obtaining temperature anomaly parameters is as follows:
[0040] Perform comprehensive calculations on the temperature data to obtain the temperature anomaly value. In the time series, compare the temperature anomaly value of the next second with the temperature anomaly value of the previous second. If the temperature anomaly value of the next second is greater than the temperature anomaly value of the previous second, then calculate the difference between the temperature anomaly value of the next second and the temperature anomaly value of the previous second to obtain the temperature anomaly severity value. As the temperature anomaly value gradually increases, the probability of mechanical failure increases. If the temperature anomaly value of the next second is less than or equal to the temperature anomaly value of the previous second, continue the comparison and calculate the temperature anomaly severity value to obtain the temperature anomaly parameter.
[0041] ;
[0042] in, represents the temperature anomaly parameter, Indicates time, Indicates the abnormal temperature value of the next second. Indicates the abnormal temperature value of the previous second.
[0043] The specific method for obtaining temperature anomaly values is as follows:
[0044] The standard temperature threshold is obtained by averaging the normal temperature during mechanical operation. Since the temperature during mechanical operation increases due to mechanical failure, the difference between the actual temperature value in the temperature data and the standard temperature threshold is calculated to obtain the temperature abnormality value.
[0045] Step 2: Set a mechanical failure trend threshold based on historical experience, and perform a real-time comparison between the fault prediction factor and the mechanical failure trend threshold. If the fault prediction factor is greater than the mechanical failure trend threshold, the machine is diagnosed as about to fail. If the fault prediction factor is less than or equal to the mechanical failure trend threshold, the machine is diagnosed as normal.
[0046] Step 3: If the machine is diagnosed to be about to fail, an alarm is issued; if the machine is diagnosed to be normal, the diagnosis is ended.
[0047] like Figure 2 As shown, the embodiment of the present invention provides an intelligent diagnosis system for rotating machinery faults integrating time-frequency domain features: comprising: a real-time data acquisition module, a historical data storage module, a machinery fault analysis and prediction module, and an alarm module;
[0048] Real-time data acquisition module: used to collect vibration frequency data and temperature data in real time and perform pre-processing;
[0049] Historical data storage module: used to store historical normal operation data of rotating machinery;
[0050] Mechanical failure analysis and prediction module: This module combines real-time vibration frequency data and temperature data with historical normal operation data of rotating machinery to perform analysis and prediction. If an impending failure is detected, the prediction result is sent to the alarm module.
[0051] Alarm module: Receives prediction results and issues an alarm, indicating that the rotating machinery used for kitchen operations has been diagnosed as faulty, thereby issuing an alarm.
[0052] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to the specific embodiments described. Obviously, many modifications and variations are possible based on the contents of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.
Claims
1. An intelligent fault diagnosis method for rotating machinery integrating time-frequency domain features, characterized by: The specific steps include: Step 1: Start collecting vibration frequency data and temperature data in real time, pre-process the vibration frequency data and temperature data and perform comprehensive calculations to obtain frequency fault severity parameters and temperature anomaly parameters, and perform comprehensive calculations on the frequency fault severity parameters and temperature anomaly parameters to obtain a fault prediction factor, wherein the fault prediction factor is used to diagnose whether the rotating machinery has a fault; Step 2: Diagnose whether the machine is about to fail based on the failure prediction factors; Step 3: If the machine is diagnosed to be about to fail, an alarm is issued; if the machine is diagnosed to be normal, the diagnosis is ended.
2. The intelligent fault diagnosis method for rotating machinery integrating time-frequency domain features according to claim 1 is characterized in that: The specific method of obtaining the fault prediction factor is as follows: The vibration frequency data is comprehensively calculated to obtain the frequency fault severity parameter, the temperature data is comprehensively calculated to obtain the temperature anomaly parameter, and the frequency fault severity parameter and the temperature anomaly parameter are comprehensively calculated to obtain the fault prediction factor.
3. The intelligent fault diagnosis method for rotating machinery integrating time-frequency domain features according to claim 2 is characterized by: The specific method for obtaining the fault prediction factor is: ； in, Represents the fault prediction factor, which is a value used to predict whether the rotating machinery has a fault. Indicates the severity of frequency fault parameters, reflecting whether the vibration frequency of the rotating machinery fluctuates. Indicates the temperature abnormality parameter, reflecting whether the temperature of the rotating machinery is abnormal.
4. The intelligent fault diagnosis method for rotating machinery integrating time-frequency domain features according to claim 2 is characterized in that: The specific method of obtaining the frequency fault severity parameter is as follows: The variance of the vibration frequency data is calculated to obtain the frequency fluctuation value. In the time series, the frequency fluctuation value of the next second is compared with the frequency fluctuation value of the previous second. If the frequency fluctuation value of the next second is greater than the frequency fluctuation value of the previous second, the difference between the frequency fluctuation value of the next second and the frequency fluctuation value of the previous second is calculated to obtain the frequency fault severity value. Because the frequency fluctuation value of the next second is greater than the frequency fluctuation value of the previous second, it can reflect that the vibration frequency of the rotating machinery tends to be disordered, that is, the greater the probability of failure. If the frequency fluctuation value of the next second is less than or equal to the frequency fluctuation value of the previous second, the frequency fluctuation value of the next second is continued to be compared with the frequency fluctuation value of the previous second, and the frequency fault severity value is summed and calculated to obtain the frequency fault severity parameter.
5. The intelligent fault diagnosis method for rotating machinery integrating time-frequency domain features according to claim 4 is characterized in that: The specific method for obtaining the frequency fault severity parameter is: ； in, Indicates the frequency fault severity parameter, Indicates time, Indicates the frequency fluctuation value of the next second, Indicates the frequency fluctuation value of the last second.
6. The intelligent fault diagnosis method for rotating machinery integrating time-frequency domain features according to claim 4 is characterized in that: The specific method of obtaining the frequency fluctuation value is as follows: The vibration frequency data is summed and averaged to obtain a frequency balance value, which is used as a standard for the vibration frequency to fluctuate around the frequency balance value. The square of the difference between each vibration frequency in the vibration frequency data and the frequency balance value is calculated to obtain the fluctuation value of each vibration frequency. The fluctuation value of each vibration frequency is summed to obtain the frequency fluctuation value.
7. The intelligent fault diagnosis method for rotating machinery integrating time-frequency domain features according to claim 2 is characterized by: The specific method of obtaining the temperature anomaly parameter is as follows: Perform comprehensive calculations on the temperature data to obtain the temperature anomaly value. In the time series, compare the temperature anomaly value of the next second with the temperature anomaly value of the previous second. If the temperature anomaly value of the next second is greater than the temperature anomaly value of the previous second, the temperature anomaly value of the next second and the temperature anomaly value of the previous second are calculated to obtain the temperature anomaly severity value. As the temperature anomaly value gradually increases, the probability of mechanical failure increases. If the temperature anomaly value of the next second is less than or equal to the temperature anomaly value of the previous second, continue the comparison, sum the temperature anomaly severity values, and obtain the temperature anomaly parameter.
8. The intelligent fault diagnosis method for rotating machinery integrating time-frequency domain features according to claim 7 is characterized in that: The specific method for obtaining the temperature anomaly parameter is: ； in, represents the temperature anomaly parameter, Indicates time, Indicates the abnormal temperature value of the next second. Indicates the abnormal temperature value of the previous second.
9. The intelligent fault diagnosis method for rotating machinery integrating time-frequency domain features according to claim 7, characterized in that: The specific method for obtaining the temperature anomaly value is as follows: The standard temperature threshold is obtained by averaging the normal temperature during mechanical operation. Since the temperature during mechanical operation increases due to mechanical failure, the difference between the actual temperature value in the temperature data and the standard temperature threshold is calculated to obtain the temperature abnormality value.
10. An intelligent diagnosis system for rotating machinery faults integrating time-frequency domain features, used to implement the intelligent diagnosis method for rotating machinery faults integrating time-frequency domain features according to any one of claims 1 to 9, characterized in that: The system includes: a real-time data acquisition module, a historical data storage module, a mechanical failure analysis and prediction module, and an alarm module; The real-time data acquisition module is used to collect vibration frequency data and temperature data in real time and perform pre-processing; The historical data storage module is used to store historical normal operation data of the rotating machinery; The mechanical failure analysis and prediction module combines real-time vibration frequency data and temperature data with historical normal operation data of the rotating machinery to perform analysis and prediction. If an impending failure is detected, the prediction result is sent to the alarm module. The alarm module receives the prediction results and issues an alarm.