Bearing health condition monitoring method, device, equipment and medium

Abstract

The application relates to a bearing health state monitoring method, device, equipment and medium, which calculates the ratio of the actual number of fault characteristic frequencies to the maximum number of fault characteristic frequencies in theory as the bearing health degree index of the measured bearing on the basis of modeling the fault characteristic frequency as a random event. Compared with the traditional technology, the value of the bearing health degree index constructed is between 0 and 1, which is the same as the value range of the bearing residual life, and the bearing health degree index and the bearing residual life have good correlation, the monotonicity of the health degree index itself is also high, and the health degree index is very suitable for representing the health state of the bearing. More importantly, the bearing health degree index is directly constructed based on the fault characteristic frequency of the measured bearing, is not a data-driven fitting index, has good physical interpretability, and thus effectively improves the monitoring performance of the bearing health state.

Description

Technical Field

[0001] This invention belongs to the field of bearing health status monitoring technology, and relates to a bearing health status monitoring method, device, equipment and medium. Background Technology

[0002] For bearing condition monitoring and fault early warning, constructing a reasonable health index is a prerequisite for improving assessment efficiency and diagnostic accuracy. In addition to using directly measurable state quantities (such as temperature, vibration, wear, and acoustic emission) to characterize the health status of bearings, another common approach is to construct new health indices based on sensor measurement signals, combined with signal processing and feature engineering. For example, features can be extracted from the time and frequency domains of the collected acceleration, and then a comprehensive evaluation index can be constructed through dimensionality reduction methods such as principal component analysis.

[0003] For example, by performing VMD (Variational Mode Decomposition) processing on the bearing vibration signal, and then extracting multi-dimensional features from the reconstructed signal and reducing the dimensionality to obtain health features; or by extracting features based on collected multi-source monitoring data, and then inputting them into a health assessment model for the wind turbine main bearing to obtain a comprehensive health index for the bearing; or by using a decomposition algorithm to decompose and reconstruct the original vibration signal to obtain the reconstructed signal of the gas turbine rolling bearing, and based on a trained SVDD (Support Vector Data Description) model, calculating the distance between each sampling point of the reconstructed signal and the health signal, and using this distance as an indicator for assessing the bearing's health status. However, in the process of realizing this invention, the inventors discovered that traditional bearing health status monitoring technologies have technical problems with insufficient monitoring performance. Summary of the Invention

[0004] To address the problems existing in the above-mentioned traditional methods, this invention proposes a bearing health condition monitoring method, a bearing health condition monitoring device, a computer device, and a computer-readable storage medium, which can effectively improve the monitoring performance of bearing health condition.

[0005] To achieve the above objectives, the embodiments of the present invention adopt the following technical solutions:

[0006] On the one hand, a method for monitoring the health status of bearings is provided, including the following steps:

[0007] Obtain the geometric and operating parameters of the bearing under test;

[0008] The theoretical fault characteristic frequency of the bearing under test is calculated based on geometric parameters and operating condition parameters.

[0009] Acquire the acceleration measurement signal of the bearing under test, and extract the actual fault characteristic frequency from the acceleration measurement signal based on the theoretical fault characteristic frequency;

[0010] The actual fault characteristic frequency is modeled as a random event and the number of actual fault characteristic frequencies at any given time is counted.

[0011] The ratio between the actual number of fault characteristic frequencies occurring at any given time and the theoretical maximum number of fault characteristic frequencies is calculated to obtain the bearing health index of the tested bearing at any given time.

[0012] On the other hand, a bearing health condition monitoring device is also provided, including:

[0013] The parameter acquisition module is used to acquire the geometric parameters and operating parameters of the bearing under test.

[0014] The theoretical characteristic calculation module is used to calculate the theoretical fault characteristic frequency of the tested bearing based on geometric parameters and operating condition parameters.

[0015] The actual feature extraction module is used to acquire the acceleration measurement signal of the bearing under test, and extract the actual fault feature frequency from the acceleration measurement signal based on the theoretical fault feature frequency.

[0016] The modeling and statistics module is used to model the actual fault feature frequency as a random event and count the number of actual fault feature frequencies at any given time.

[0017] The index calculation module is used to calculate the ratio between the actual number of fault characteristic frequencies at any given time and the theoretical maximum value of the number of fault characteristic frequencies, thereby obtaining the bearing health index of the tested bearing at any given time.

[0018] In another aspect, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the bearing health condition monitoring method described above.

[0019] Furthermore, a computer-readable storage medium is also provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the bearing health condition monitoring method described above.

[0020] One of the above technical solutions has the following advantages and beneficial effects:

[0021] The aforementioned bearing health monitoring methods, devices, equipment, and media, by modeling fault characteristic frequencies as random events, calculate the theoretical maximum value of the number of fault characteristic frequencies given a fault type and frequency order. Then, the ratio of the actual number of fault characteristic frequencies to the theoretical maximum value is calculated as the bearing health index of the tested bearing. Compared to traditional techniques, the constructed bearing health index value lies between 0 and 1, the same range as the bearing's remaining useful life (RUL). Furthermore, the bearing health index shows a strong correlation with the bearing's RUL and exhibits high monotonicity, making it highly suitable for characterizing bearing health. More importantly, the bearing health index is directly constructed based on the fault characteristic frequencies of the tested bearing, rather than a data-driven fitting index, giving it excellent physical interpretability and effectively improving the monitoring performance of bearing health. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of this application or the conventional technology, the drawings used in the description of the embodiments or the conventional technology will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a flowchart illustrating a bearing health status monitoring method in one embodiment;

[0024] Figure 2 This is a schematic diagram of a frequency observation window in one embodiment;

[0025] Figure 3 This is a schematic diagram illustrating the frequency variance determination threshold in one embodiment;

[0026] Figure 4 This is a schematic diagram of experimental results in one embodiment; where (a) is the bearing's full life cycle acceleration, and (b) is the bearing's health index HI-FCF;

[0027] Figure 5 This is a schematic diagram of the module structure of a bearing health status monitoring device in one embodiment. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0029] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.

[0030] It should be noted that, in this document, the reference to "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The presentation of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will understand that the embodiments described herein can be combined with other embodiments. The term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items, and all possible combinations thereof.

[0031] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0032] In the process of developing this invention, the inventors discovered that although traditional bearing health monitoring technologies can effectively characterize the health status of bearings to a certain extent, they all have certain limitations. For example, constructing health indicators based on purely data-driven methods lacks mechanistic interpretability and has limited generalization ability; a health indicator suitable for one bearing is often not suitable for other bearings. In general, there is currently no universal and effective technology for designing and applying bearing health indicators in academia and industry.

[0033] In practical research and applications, fault characteristic frequencies are an inherent feature of faulty bearings: when one or more components of a bearing fail, after spectral transformation of the acceleration signal, fault spikes will appear at the corresponding fault characteristic frequencies and their higher harmonics. If a bearing health index can be directly constructed based on the fault characteristic frequencies, the modeling efficiency of the health index can be significantly improved, as well as its interpretability in terms of mechanism and its generalization ability among different bearings, thereby effectively improving the monitoring performance of bearing health status.

[0034] Considering that the actual fault characteristic frequency is easily affected by noise and sideband modulation, this invention models it as a random variable (event). A frequency observation window is designed to identify the actual fault characteristic frequency, and the variance of the spectral amplitude within the frequency observation window is analyzed and compared with a predefined threshold. When the former exceeds the latter, it is defined as a true event, indicating that there is a valid fault characteristic frequency in the current frequency window; otherwise, it is recorded as a false event, indicating that there is no valid fault characteristic frequency in the current frequency observation window.

[0035] By modeling fault characteristic frequencies as random events, the number of times this random event is true at any given moment can be counted. Simultaneously, the theoretical maximum value of the number of fault characteristic frequencies can be calculated for a given fault type and the order of the characteristic frequencies, and then the ratio between the two can be calculated. The theoretical maximum value of the number of fault characteristic frequencies indicates that the bearing exhibits fault characteristic frequencies at all orders across all fault types, representing the most severe fault state of the bearing. The ratio between the two represents the similarity between the bearing's current state and its most severe fault state, and therefore can serve as an indicator of the bearing's health.

[0036] Compared with traditional technologies, the technical solution designed in this invention constructs a bearing health index with values ​​between 0 and 1, which is the same range as the bearing's remaining useful life (RUL). Furthermore, the health index shows a strong correlation with RUL and exhibits high monotonicity, making it highly suitable for characterizing the health status of bearings. Moreover, the health index is directly constructed based on the bearing's fault characteristic frequencies, rather than a data-driven fitting index, giving it excellent physical interpretability. In addition, this invention is the first to model the occurrence of fault characteristic frequencies as random events, which changes the traditional modeling methods for fault characteristic frequencies and provides a new technological development approach for research and application in this field.

[0037] In one embodiment, such as Figure 1 As shown, a bearing health status monitoring method is provided, which may include the following processing steps S10 to S18:

[0038] S10, obtain the geometric parameters and operating parameters of the bearing under test;

[0039] S12, calculate the theoretical fault characteristic frequency of the tested bearing based on geometric parameters and operating condition parameters;

[0040] S14: Acquire the acceleration measurement signal of the bearing under test, and extract the actual fault characteristic frequency from the acceleration measurement signal according to the theoretical fault characteristic frequency;

[0041] S16, model the actual fault characteristic frequency as a random event and count the number of actual fault characteristic frequencies at any given moment;

[0042] S18, calculate the ratio between the actual number of fault characteristic frequencies at any given time and the theoretical maximum value of the number of fault characteristic frequencies, to obtain the bearing health index of the tested bearing at any given time.

[0043] It is understood that the geometric parameters of the bearing under test may include, but are not limited to, the bearing ball diameter, the bearing pitch diameter, and the number of bearing balls; the operating parameters of the bearing under test may include, but are not limited to, the shaft frequency and the bearing initial contact angle. The geometric and operating parameters of the bearing under test can be obtained by manual uploading or manual input, or can be read by the monitoring computer from a storage device, server, or third-party database containing the aforementioned parameters.

[0044] like Figure 1 As shown, the theoretical fault characteristic frequency of the bearing under test is first calculated: Specifically, when different components of the bearing under test fail, the corresponding fault characteristic frequency and its higher harmonics will appear in the spectrum of the measured acceleration signal (hereinafter referred to as the acceleration measurement signal).

[0045] In some implementations, the theoretical fault characteristic frequencies of the tested bearing include the fault characteristic frequency corresponding to a fault in the bearing outer race, the fault characteristic frequency corresponding to a fault in the bearing inner race, the fault characteristic frequency corresponding to a fault in the bearing cage, and the fault characteristic frequency corresponding to a fault in the bearing balls. For example, the fault characteristic frequency corresponding to a fault in the bearing outer race is BPFO (Ball Passing Frequency of Outer race); the fault characteristic frequency corresponding to a fault in the bearing inner race is BPFI (Ball Passing Frequency of Inner race); the fault characteristic frequency corresponding to a fault in the bearing cage is FTF (Fundamental Train Frequency); and the fault characteristic frequency corresponding to a fault in the bearing balls is BSF (Ball Spin Frequency). The calculation formulas for the above theoretical fault characteristic frequencies can be as follows:

[0046]

[0047]

[0048]

[0049]

[0050] Where f is the rotational frequency of the shaft, d is the diameter of the bearing balls, D is the bearing's pitch diameter, α is the bearing's initial contact angle, and n is the number of bearing balls. Therefore, given the geometric and operating parameters of the bearing under test, the above four theoretical fault characteristic frequency values ​​can be calculated.

[0051] Then, the acceleration measurement signal of the bearing under test is acquired, and the actual fault characteristic frequency is extracted from it. In practical applications, due to factors such as bearing speed fluctuations, transient contact angle changes, measurement noise, and sideband modulation, the actual fault characteristic frequency often deviates from its theoretical fault characteristic frequency value. In some cases, the actual fault characteristic frequency is even completely submerged in noise and cannot be identified. Therefore, in this embodiment, the peak of the actual fault characteristic frequency observed within the frequency observation window is modeled as a random event. Finally, based on this, the probability of the actual fault characteristic frequency occurring at any given time is calculated, which is the ratio between the number of actual fault characteristic frequencies occurring at any given time and the theoretical maximum value of the number of fault characteristic frequencies. The theoretical maximum value of the number of fault characteristic frequencies indicates that the bearing exhibits fault characteristic frequencies at all orders of all fault types, representing the most severe fault state of the bearing. The ratio of the two values ​​indicates the similarity between the current state of the bearing and the most severe fault state. Therefore, it can be used as a health index of the bearing to present the health status of the bearing under test at any (monitoring) time.

[0052] The aforementioned bearing health status monitoring method, by modeling fault characteristic frequencies as random events, calculates the theoretical maximum number of fault characteristic frequencies given a fault type and frequency order. Then, it calculates the ratio of the actual number of fault characteristic frequencies to the theoretical maximum number as the bearing health index. Compared to traditional techniques, the constructed bearing health index ranges between 0 and 1, the same range as the bearing's remaining life, and shows a strong correlation with the remaining life. Furthermore, the index exhibits high monotonicity, making it highly suitable for characterizing bearing health. More importantly, the bearing health index is directly constructed based on the fault characteristic frequencies of the tested bearing, rather than a data-driven fitting index, giving it excellent physical interpretability and effectively improving the monitoring performance of bearing health status.

[0053] It should be noted that step S14 above may specifically include six sub-steps: performing spectral transformation on the acceleration measurement signal, defining the frequency observation window, calculating the variance of the spectral amplitude of each observation window, calculating the variance of the spectral amplitude of the initial sample, and defining the judgment criteria for the fault characteristic frequency. These will be described in detail below.

[0054] Spectral transformation of acceleration measurement signals: The actual acceleration measurement signals measured on the bearing test bench are generally time-domain signals. To extract the characteristic frequencies of actual faults, it is first necessary to transform the acceleration measurement signals from the time domain to the frequency domain. This can be achieved using the existing Fast Fourier Transform (FFT). The specific spectral transformation methods and procedures can be understood by referring to the existing FFT processing procedures, and will not be elaborated here.

[0055] Define the frequency observation window: The length of the frequency observation window can be manually set according to actual observation needs, so as to observe frequency spikes on the acceleration measurement signal in the frequency domain by sliding the window, thereby searching for high actual fault characteristic frequencies. For example, the length of the fixed frequency observation window can be selected based on the experience of experts in the field or historical observation patterns, as long as it can be used to effectively observe the fault characteristic frequencies required.

[0056] In one embodiment, during the extraction of the actual fault characteristic frequency from the acceleration measurement signal based on the theoretical fault characteristic frequency, a search is performed within a search interval of ±p% near the theoretical fault characteristic frequency. In practical applications, due to factors such as lubrication conditions and external force loading, the bearing balls may slip instead of rolling completely, causing the actual fault characteristic frequency of the bearing to deviate from its theoretical value (i.e., the theoretical fault characteristic frequency). Therefore, after completing the spectrum transformation, the actual fault characteristic frequency should be searched and located within a certain frequency range near the theoretical fault characteristic frequency. To this end, this embodiment proposes an interval search method, such as... Figure 2 The diagram illustrates this interval search method, which involves searching within a range of ±p% around the theoretical fault characteristic frequency. The search interval is defined as [(1-p%)×f k , (1+p%)×f k ], where f k The theoretical value of the characteristic frequency of the k-th order fault is f. k This refers to one of the four fault characteristic frequencies mentioned above (BPFO, BPFI, BSF, and FTF); p% is a configuration parameter that can be configured and optimized according to the specific monitoring needs of the application scenario. A generally recommended optimal value is p% = 5%. Similarly, this interval search method can be used to search for fault characteristic frequencies of higher harmonics. Ultimately, this can be used to achieve more efficient and accurate observation of actual fault characteristic frequencies.

[0057] Calculate the variance of the spectral amplitude for each observation window: The amplitude of the fault characteristic frequency will be significantly greater than the amplitude of the noise. If a fault frequency spike exists within a certain frequency observation window, the variance of the spectral amplitude for the entire frequency observation window will increase significantly. Therefore, the existence of an actual fault characteristic frequency can be determined by calculating the variance of the spectral amplitude for each frequency observation window. The formula for calculating the variance S is shown in equation (5), where... Let be the spectral amplitude at the i-th point within the frequency observation window. This represents the mean amplitude of all frequency points within the current frequency observation window, where n is the number of sample points within the frequency observation window.

[0058]

[0059] Calculate the spectral amplitude variance of the initial sample: After calculating the spectral amplitude variance of each frequency observation window, a comparison benchmark needs to be defined. Since the acceleration amplitude of different bearings or the same bearing under different operating conditions varies greatly, it is difficult to define a definite value. Therefore, in this embodiment, the mean variance of the spectrum of all fault types and all orders of the bearing in the initial state (i.e., the healthy state when it is brand new) is used as the comparison benchmark, as shown in Equation (6), where S0 is the mean amplitude variance of the new bearing in each frequency observation window, which is used as the benchmark for subsequent comparison; S ij (0) represents the variance of the spectral amplitude under the frequency observation window of the i-th fault at the initial time (i.e., t = 0). max N represents the highest order of the fault characteristic frequencies. fault This represents the total number of fault types. Typically, O max The possible values ​​are 5 and N. fault It can take a value of 4, which can be used to represent faults in the outer ring, inner ring, cage, and ball.

[0060]

[0061] In one embodiment, during the process of extracting the actual fault characteristic frequency from the acceleration measurement signal based on the theoretical fault characteristic frequency, the threshold for determining whether there is an actual fault characteristic peak in each frequency observation window is:

[0062] m×S0

[0063] Where m is an adjustment factor greater than 1, and S0 is the mean amplitude variance of the new bearing in each frequency observation window.

[0064] It is understandable that the criteria for determining fault characteristic frequencies can be defined as follows: After calculating the variance of the spectral amplitude under each frequency observation window and defining the comparison benchmark for the spectral variance, further criteria can be defined to determine whether there are actual fault characteristic peaks in each frequency observation window. For example... Figure 3 As shown, in this embodiment, the judgment threshold is defined as m×S0, where m (m>1) is an adjustment factor. The specific value can be set according to the requirements of the threshold robustness. The larger the value of m, the stronger the threshold robustness to noise.

[0065] Next, the actual fault characteristic frequency is modeled as a random event: In practical applications, due to factors such as bearing speed fluctuations, transient contact angle changes, measurement noise, and sideband modulation, the actual fault characteristic frequency often deviates from its theoretical value. In some cases, the actual fault characteristic frequency is even completely submerged in noise and cannot be identified. Therefore, in this embodiment, the peak of the actual fault characteristic frequency observed within the frequency observation window is modeled as a random event, which can be denoted as C. ij (k), as given in equation (7), provides a specific modeling approach.

[0066] Specifically, for any time k, when the amplitude variance S of the frequency observation window ij If (k) is greater than the comparison threshold m×S0, then event C ij (k) is marked as true, i.e., C ij (k) = 1, indicating that there is a valid actual fault characteristic frequency in the current frequency observation window; otherwise, event C ij (k) is denoted as false, i.e., C ij (k) = 0 indicates that there is no valid actual fault characteristic frequency in the current frequency observation window.

[0067]

[0068] Finally, the probability of the occurrence of actual fault characteristic frequencies is calculated. This part may include three steps: counting the number of all actual fault characteristic frequencies (i.e., actual fault characteristic frequencies), calculating the theoretical maximum value of the number of fault characteristic frequencies, and calculating the ratio between the two to generate a bearing health index. The specific implementation methods of each step will be introduced below.

[0069] Statistically count the frequency of all actual fault characteristics: The previous steps have already analyzed whether there are fault characteristic spikes at any order i for any fault type j, when C... ij (k) = 1 indicates the existence of the j-th order actual fault characteristic frequency of fault type i; conversely, C ij (k) = 0, indicating that the actual fault characteristic frequency does not exist. Based on this, the total number of fault characteristic frequencies that actually occur for all possible fault types at all possible orders can be calculated by equation (8).

[0070]

[0071] The theoretical maximum value of the number of fault characteristic frequencies is calculated as follows: There are four types of bearing faults: outer ring fault, inner ring fault, ball (bearing ball) fault, and cage fault. When any of these faults occurs, the spectrum of the bearing's acceleration measurement signal will show spikes at the corresponding fault characteristic frequencies and their higher harmonics. Therefore, the theoretical maximum value of the number of fault characteristic frequencies can be calculated using equation (9), where O max N represents the highest order of the fault characteristic frequency. fault This represents the total number of fault types. In practical engineering applications, O max It can generally be set to 5, N fault This can be determined based on the number of actual fault types.

[0072] O max ×N fault (9)

[0073] Calculate the ratio of the two and generate a bearing health index: At any time k, the actual number of fault characteristic frequencies includes the fault types that have already shown fault characteristic peaks and the order of the fault characteristic frequencies under each type, representing the current fault state of the bearing. The theoretical maximum value of the number of fault characteristic frequencies represents that the bearing has fault characteristic frequencies under all fault types and all orders, representing the most severe failure state of the bearing. Therefore, the ratio of the two can be used to represent the similarity between the current health state of the bearing and complete failure, and can thus be used as a bearing health index. Equation (10) gives the formula for calculating the ratio of the two, which can be expressed as HI-FCF and can be used as a bearing health index.

[0074]

[0075] like Figure 4 As shown, (a) presents the acceleration curve of a bearing throughout its entire life cycle from brand new to complete failure, and (b) presents the HI-FCF health index for the bearing's entire life cycle constructed based on this invention. A comparison reveals that the health index provided by this invention can accurately characterize the changing trend of the bearing's acceleration profile throughout its entire life cycle. The filtered health index is stable, with a variation range between 0 and 1, making it perfectly suitable for bearing condition monitoring and remaining life estimation.

[0076] It should be understood that, although the above process Figure 1 The steps in the diagram are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order in which these steps are executed; they can be performed in other orders. Furthermore, the above process... Figure 1At least some of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0077] In one embodiment, such as Figure 5 As shown, a bearing health status monitoring device 100 is also provided, including a parameter acquisition module 11, a theoretical feature calculation module 13, an actual feature extraction module 15, a modeling and statistics module 17, and an index calculation module 19. The parameter acquisition module 11 is used to acquire the geometric parameters and operating condition parameters of the bearing under test. The theoretical feature calculation module 13 is used to calculate the theoretical fault characteristic frequency of the bearing under test based on the geometric parameters and operating condition parameters. The actual feature extraction module 15 is used to acquire the acceleration measurement signal of the bearing under test and extract the actual fault characteristic frequency from the acceleration measurement signal based on the theoretical fault characteristic frequency. The modeling and statistics module 17 is used to model the actual fault characteristic frequency as a random event and count the number of actual fault characteristic frequencies at any given time. The index calculation module 19 is used to calculate the ratio between the actual number of fault characteristic frequencies occurring at any given time and the theoretical maximum value of the number of fault characteristic frequencies, obtaining the bearing health index of the bearing under test at any given time.

[0078] The aforementioned bearing health monitoring device 100, by modeling fault characteristic frequencies as random events, calculates the theoretical maximum value of the number of fault characteristic frequencies given a fault type and frequency order. Then, it calculates the ratio of the actual number of fault characteristic frequencies to the theoretical maximum value as the bearing health index of the tested bearing. Compared to traditional techniques, the constructed bearing health index value lies between 0 and 1, the same range as the bearing's remaining life, and shows a strong correlation with the bearing's remaining life. Furthermore, the index itself exhibits high monotonicity, making it highly suitable for characterizing bearing health. More importantly, the bearing health index is directly constructed based on the fault characteristic frequencies of the tested bearing, rather than a data-driven fitting index, giving it excellent physical interpretability and effectively improving the monitoring performance of bearing health.

[0079] In one embodiment, the theoretical fault characteristic frequencies of the tested bearing include the fault characteristic frequencies corresponding to faults in the outer ring, the inner ring, the cage, and the balls.

[0080] In one embodiment, during the process of extracting the actual fault characteristic frequency from the acceleration measurement signal based on the theoretical fault characteristic frequency, the actual feature extraction module 15 searches within a search range of ±p% near the theoretical fault characteristic frequency.

[0081] The search range is:

[0082] [(1-p%)×f k , (1+p%)×f k ]

[0083] Among them, f k is the theoretical value of the characteristic frequency of the k-th order fault, and p% is the configuration parameter.

[0084] In one embodiment, during the process of extracting the actual fault feature frequency from the acceleration measurement signal based on the theoretical fault feature frequency, the threshold for the actual feature extraction module 15 to determine whether there is an actual fault feature peak in each frequency observation window is:

[0085] m×S0

[0086] Where m is an adjustment factor greater than 1, and S0 is the mean amplitude variance of the new bearing in each frequency observation window.

[0087] It is understood that for a detailed explanation of the bearing health condition monitoring device 100, please refer to the corresponding explanations of the various embodiments of the bearing health condition monitoring method above, and will not be repeated here. Each module in the bearing health condition monitoring device 100 can be implemented entirely or partially through software, hardware, or a combination thereof. Each module can be embedded in hardware or independently of a device with data processing capabilities, or stored in software in the memory of the aforementioned device, so that the processor can call and execute the operations corresponding to each module. The aforementioned device can be, but is not limited to, various types of fault diagnosis and monitoring devices already existing in the art.

[0088] In one embodiment, a computer device is also provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to perform the following processing steps: acquiring the geometric parameters and operating parameters of the bearing under test; calculating the theoretical fault characteristic frequency of the bearing under test based on the geometric parameters and operating parameters; acquiring the acceleration measurement signal of the bearing under test, and extracting the actual fault characteristic frequency from the acceleration measurement signal based on the theoretical fault characteristic frequency; modeling the actual fault characteristic frequency as a random event and counting the number of actual fault characteristic frequencies at any given time; calculating the ratio between the actual number of fault characteristic frequencies occurring at any given time and the theoretical maximum value of the number of fault characteristic frequencies, to obtain the bearing health index of the bearing under test at any given time.

[0089] It is understood that, in addition to the memory and processor mentioned above, the computer equipment described above also includes other hardware and software components not listed in this specification. The specific components can be determined based on the model of the specific fault assessment computer in different application scenarios, and will not be listed and described in detail in this specification.

[0090] In one embodiment, when the processor executes the computer program, it can also implement the steps or sub-steps added in the various embodiments of the bearing health status monitoring method described above.

[0091] In one embodiment, a computer-readable storage medium is also provided, on which a computer program is stored. When the computer program is executed by a processor, it performs the following processing steps: acquiring the geometric parameters and operating parameters of the bearing under test; calculating the theoretical fault characteristic frequency of the bearing under test based on the geometric parameters and operating parameters; acquiring the acceleration measurement signal of the bearing under test, and extracting the actual fault characteristic frequency from the acceleration measurement signal based on the theoretical fault characteristic frequency; modeling the actual fault characteristic frequency as a random event and counting the number of actual fault characteristic frequencies at any given time; calculating the ratio between the number of actual fault characteristic frequencies at any given time and the theoretical maximum value of the number of fault characteristic frequencies, to obtain the bearing health index of the bearing under test at any given time.

[0092] In one embodiment, when the computer program is executed by the processor, it can also implement the steps or sub-steps added in the various embodiments of the bearing health status monitoring method described above.

[0093] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), memory bus DRAM (RDRAM), and interface DRAM (DRDRAM), etc.

[0094] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0095] The above embodiments merely illustrate several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, all of which fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for monitoring the health status of bearings, characterized in that, Including the following steps: Obtain the geometric and operating parameters of the bearing under test; The theoretical fault characteristic frequency of the tested bearing is calculated based on the geometric parameters and the operating condition parameters. Acquire the acceleration measurement signal of the bearing under test, and extract the actual fault characteristic frequency from the acceleration measurement signal according to the theoretical fault characteristic frequency; The actual fault characteristic frequencies are modeled as random events, and the number of actual fault characteristic frequencies at any given time is counted. A frequency observation window is designed to identify the actual fault characteristic frequencies, and the variance of the spectral amplitude within the frequency observation window is analyzed and compared with a predefined threshold. When the variance exceeds the threshold, it is defined as a true event, indicating that a valid fault characteristic frequency exists in the current frequency window; otherwise, it is recorded as a false event, indicating that a valid fault characteristic frequency does not exist in the current frequency observation window. The threshold for determining whether an actual fault characteristic peak exists in each frequency observation window is: in, m For adjustment factors greater than 1, This represents the mean variance of the amplitude of the new bearing across various frequency observation windows. The ratio between the actual number of fault characteristic frequencies at any given time and the theoretical maximum value of the number of fault characteristic frequencies is calculated to obtain the bearing health index of the tested bearing at that given time; wherein, the theoretical maximum value of the number of fault characteristic frequencies is: in, The highest order of the fault characteristic frequency. This represents the total number of fault types.

2. The bearing health status monitoring method according to claim 1, characterized in that, The theoretical fault characteristic frequencies of the tested bearing include the fault characteristic frequencies corresponding to faults in the outer ring, the inner ring, the cage, and the balls.

3. The bearing health status monitoring method according to claim 1 or 2, characterized in that, During the process of extracting the actual fault characteristic frequency from the acceleration measurement signal based on the theoretical fault characteristic frequency, in the vicinity of the theoretical fault characteristic frequency... Search within the search range; The search range is: in, For the first k The theoretical value of the characteristic frequency of the first-order fault. For configuration parameters.

4. A bearing health status monitoring device, characterized in that, The bearing health status monitoring method according to any one of claims 1 to 3 includes: The parameter acquisition module is used to acquire the geometric parameters and operating parameters of the bearing under test. The theoretical characteristic calculation module is used to calculate the theoretical fault characteristic frequency of the tested bearing based on the geometric parameters and the operating condition parameters; The actual feature extraction module is used to acquire the acceleration measurement signal of the bearing under test, and extract the actual fault feature frequency from the acceleration measurement signal according to the theoretical fault feature frequency. The modeling and statistics module is used to model the actual fault feature frequency as a random event and count the number of actual fault feature frequencies at any given time. The index calculation module is used to calculate the ratio between the actual number of fault characteristic frequencies at any given time and the theoretical maximum value of the number of fault characteristic frequencies, so as to obtain the bearing health index of the tested bearing at any given time.

5. The bearing health status monitoring device according to claim 4, characterized in that, The theoretical fault characteristic frequencies of the tested bearing include the fault characteristic frequencies corresponding to faults in the outer ring, the inner ring, the cage, and the balls.

6. The bearing health status monitoring device according to claim 4 or 5, characterized in that, In the process of extracting the actual fault characteristic frequency from the acceleration measurement signal based on the theoretical fault characteristic frequency, the actual feature extraction module extracts the actual fault characteristic frequency in the vicinity of the theoretical fault characteristic frequency. Search within the search range; The search range is: in, For the first The theoretical value of the characteristic frequency of the first-order fault. For configuration parameters.

7. A computer device, comprising a memory and a processor, characterized in that, The memory stores a computer program, and when the processor executes the computer program, it implements the steps of the bearing health condition monitoring method according to any one of claims 1 to 3.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the bearing health condition monitoring method according to any one of claims 1 to 3.

Images