Real-time monitoring method for bearing lubrication state based on fluid dynamics model correction

Abstract

The present application belongs to the technical field of bearing lubrication state monitoring, and particularly relates to a bearing lubrication state real-time monitoring method based on fluid dynamics model correction, which comprises the following steps: obtaining original vibration acceleration signals and running state data, performing variational modal decomposition on the original vibration acceleration signals to extract cavitation characteristic components; calculating a cavitation kinetic energy impact index according to the extracted cavitation characteristic components and the running state data; dynamically correcting the equivalent fluid density of lubricating oil according to the cavitation kinetic energy impact index and the contact area pressure of the bearing estimated from the running state data; solving the minimum oil film thickness of the lubricating oil film by using the corrected equivalent fluid density, and determining a lubrication safety margin index in combination with the temperature rise rate of the bearing. The present application overcomes the defect that the monitoring result is distorted due to the single-phase fluid assumption, and improves the accuracy of bearing lubrication state monitoring and early risk warning.

Description

Technical Field

[0001] This invention relates to the field of bearing lubrication condition monitoring technology. More specifically, this invention relates to a real-time bearing lubrication condition monitoring method based on a modified fluid dynamics model. Background Technology

[0002] In modern industrial production systems, various rotating machines are core equipment ensuring continuous operation. For critical components such as heavy-duty rolling bearings, their lubrication status under high-speed operation or heavy-load conditions directly determines the stability and service life of the equipment. If the lubrication status inside such bearings cannot be accurately monitored, premature wear or even catastrophic failures can easily occur, causing incalculable economic losses to industrial production.

[0003] Currently, for monitoring the lubrication status of such bearings, the industry generally adopts a real-time monitoring method based on traditional fluid dynamics models. This method mainly uses external sensors to acquire operating status data, including parameters such as real-time bearing speed, load, and temperature. Then, the various operating status data are substituted into the Reynolds equation to calculate the theoretical minimum oil film thickness of the lubricating oil film inside the bearing, thereby assessing the current lubrication status of the bearing and determining its safety.

[0004] However, in actual operating conditions, when bearings are operating at high speeds or when the pressure in the bearing contact area changes drastically, cavitation can easily occur within the lubricating oil film. Traditional fluid dynamics models are based on the assumption of an incompressible single-phase fluid, which assumes a fixed density for the lubricating oil. When cavitation occurs within the lubricating oil film, the oil film instantly transforms into a two-phase flow state, and its actual fluid density and bulk modulus change drastically. If the single-phase fluid assumption is still used for simulation at this time, the calculated oil film load-bearing capacity will be much greater than the actual value, causing the monitoring system to misjudge the current lubrication state of the bearing as being within a safe range, when in reality the oil film inside the bearing may have completely ruptured. In this case, because the single-phase fluid assumption deviates significantly from the actual physical state, bearing failure occurs, leading to distorted monitoring results. Summary of the Invention

[0005] To address the technical problem of cavitation within the lubricating oil film of bearings leading to the failure of the single-phase fluid assumption and resulting in distorted monitoring results, this invention proposes a real-time monitoring method for bearing lubrication status based on a fluid dynamics model. This method can dynamically correct the equivalent fluid density of the lubricating oil to align with the actual physical state, thereby accurately determining the minimum oil film thickness inside the bearing and combining it with the bearing's temperature rise rate to determine the lubrication safety margin index, achieving precise assessment of the bearing lubrication status and early risk warning.

[0006] This invention provides a real-time monitoring method for bearing lubrication status based on a fluid dynamics model, comprising: acquiring raw vibration acceleration signals and operating status data; performing variational mode decomposition on the raw vibration acceleration signals to extract cavitation characteristic components; calculating the cavitation kinetic energy impact index based on the extracted cavitation characteristic components and the operating status data; dynamically correcting the equivalent fluid density of the lubricating oil based on the cavitation kinetic energy impact index and the bearing contact area pressure estimated from the operating status data; solving for the minimum oil film thickness of the lubricating oil film using the corrected equivalent fluid density, and determining the lubrication safety margin index in conjunction with the bearing temperature rise rate.

[0007] This invention effectively removes and purifies cavitation feature components and filters out redundant noise by performing deep signal decoupling processing on the original vibration acceleration signal. Furthermore, the extracted features are mapped to fluid physical properties and a balancing mechanism is introduced to dynamically correct the equivalent fluid density of the lubricating oil. This effectively avoids the calculation divergence and distortion problems that easily occur in the Reynolds equation under extreme conditions under the traditional single-phase fluid assumption, achieving effective alignment between the physical model and the actual physical state. This lays a reliable foundation for accurate assessment of bearing lubrication status and early risk warning.

[0008] Preferably, the cavitation kinetic energy impact index is calculated as follows:

[0009]

[0010] in, The cavitation kinetic energy impact index; This is the energy scaling factor; This is the length of the sampling time window; The signal represents the cavitation feature components after normalization. The real-time rotational speed of the bearing is obtained by the encoder; This refers to the rated reference speed of the bearing; is the logarithmic reference constant.

[0011] This invention constructs a cavitation kinetic energy impact index by extracting cavitation feature components and operational status data. This index fully considers the nonlinear amplification effect of the bearing's real-time rotational speed on the cavitation generation and collapse frequencies, and dynamically adjusts the sampling time window length according to the bearing's real-time rotational speed to ensure a constant angular domain span. This method successfully transforms chaotic instantaneous high-frequency energy into a smooth, stable, and intuitively comparable dimensionless feature index, thereby objectively and accurately reflecting the cavitation generation frequency and intensity of the bearing within a unit rotation cycle.

[0012] Preferably, the equivalent fluid density of the lubricating oil is calculated as follows:

[0013]

[0014] in, This is the equivalent fluid density of the lubricating oil; This is the nominal density of the lubricating oil under standard atmospheric pressure. This represents the maximum theoretical void fraction. The cavitation kinetic energy impact index; This refers to the pressure in the contact area of ​​the bearing. This is the pressure suppression coefficient.

[0015] In the process of dynamically correcting the equivalent fluid density of lubricating oil, this invention fully integrates the inhibitory effect of the bearing contact area pressure estimated from operating data on cavitation expansion, and specifically introduces a pressure suppression coefficient. This improvement constructs a pressure bubble balancing mechanism that conforms to the laws of real fluid physics, forcibly compressing the bubble volume under high pressure to hinder gas expansion. This mechanism effectively avoids the illusion of excessive correction of the equivalent fluid density of lubricating oil caused by strong vibration in high-pressure, heavy-load areas, and effectively ensures the accuracy of physical parameter derivation under gas-liquid mixing conditions.

[0016] Preferably, the method for calculating the lubrication safety margin index is as follows:

[0017]

[0018] in, For lubrication safety margin index; This is the minimum oil film thickness calculated based on the equivalent fluid density; This is the root mean square value of the surface roughness of the inner ring raceway of the bearing. This is the root mean square value of the surface roughness of the bearing rolling elements; This represents the current rate of temperature rise of the bearing. This is the threshold value for the maximum allowable rate of temperature rise in the bearing.

[0019] This invention utilizes a modified equivalent fluid density to determine the minimum oil film thickness of the lubricating oil film, and integrates this with the bearing's temperature rise rate and the root mean square value of the surface roughness of the bearing's inner ring raceway and rolling elements to jointly determine the lubrication safety margin index. This index not only reflects the geometric safety isolation through the film thickness ratio, but also introduces a thermodynamic penalty factor determined by the bearing's temperature rise rate. This allows the system to not only react instantly to oil film rupture, but also to detect early signs of heat accumulation caused by localized dry friction, giving the monitoring system a keen early risk warning capability.

[0020] Preferably, the step of acquiring the original vibration acceleration signal and operating status data, performing variational mode decomposition on the original vibration acceleration signal, and extracting cavitation feature components includes: acquiring the high-frequency vibration acceleration signal of the non-load-bearing area as the original vibration acceleration signal; setting the number of decomposition layers and the penalty factor, and decomposing the original vibration acceleration signal into multiple intrinsic mode components through a variational mode decomposition algorithm; calculating the kurtosis value of each intrinsic mode component, and selecting the intrinsic mode component with the largest kurtosis value as the cavitation feature component.

[0021] Preferably, the contact area pressure of the bearing is estimated by dividing the radial load of the bearing by the effective bearing area.

[0022] Preferably, before performing variational mode decomposition on the original vibration acceleration signal, a wavelet denoising algorithm is used to preprocess the original vibration acceleration signal to remove low-frequency baseline drift and high-frequency environmental white noise.

[0023] Preferably, the sampling time window length is dynamically adjusted according to the real-time rotational speed of the bearing.

[0024] Preferably, the temperature rise rate of the bearing is obtained by extracting the slope through least squares linear fitting of continuously collected bearing temperature data within a sliding time window.

[0025] This invention extracts the slope by performing least-squares linear fitting on continuously acquired bearing temperature data through a sliding time window. This effectively eliminates the interference of high-frequency temperature fluctuation noise on the calculation of the temperature rise rate, thereby ensuring the accuracy of the obtained bearing temperature rise rate.

[0026] Preferably, when the lubrication safety margin index is lower than a set threshold, a system alarm is triggered, indicating that the current bearing lubrication status is in a high-risk range.

[0027] The beneficial effects of this invention are as follows:

[0028] This invention effectively extracts cavitation characteristic components by performing variational mode decomposition and kurtosis optimization on the original vibration acceleration signal, thus effectively eliminating interference from environmental noise and structural resonance. Simultaneously, this invention constructs a cavitation kinetic energy impact index and combines this index with the bearing contact area pressure to dynamically correct the equivalent fluid density of the lubricating oil. This achieves effective alignment between the simulation model and the actual physical reality of the gas-liquid two-phase flow, effectively avoiding the cavitation phenomenon inside the bearing's lubricating oil film that could lead to the failure of the single-phase fluid assumption and distortion of monitoring results. This significantly improves the physical authenticity and calculation accuracy of bearing lubrication condition monitoring.

[0029] Furthermore, this invention introduces a pressure suppression mechanism and a thermodynamic penalty factor for the temperature rise rate into the modified model, constructing a lubrication safety margin index that incorporates both geometric and thermodynamic dimensions. The pressure suppression mechanism fully considers the physical limitations of high-pressure, heavy-load environments on bubble volume expansion, successfully avoiding excessive fluid density correction caused by misleading strong vibration signals. The evaluation system integrating the temperature rise rate allows the monitoring system to not only reflect the safety isolation at the oil film thickness level but also to keenly capture the heat accumulation trend caused by bearing dry friction. This multi-physics-integrated evaluation method enables the monitoring system to detect early signs of pressure field distortion and thermal runaway before the lubricating oil film ruptures, significantly improving the system's sensitivity and early risk warning capabilities, effectively ensuring the long-term stable operation of the bearing. Attached Figure Description

[0030] Figure 1 This is a flowchart illustrating the real-time monitoring method for bearing lubrication status based on fluid dynamics model correction in this invention.

[0031] Figure 2 This is a schematic diagram illustrating the spatial distribution of bearing lubrication states based on multidimensional features in this invention;

[0032] Figure 3 This is a schematic illustration of the modified bearing oil film pressure field distribution thermogram in this invention. Detailed Implementation

[0033] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.

[0034] This invention discloses a method for real-time monitoring of bearing lubrication status based on a modified fluid dynamics model, referring to... Figure 1 This includes steps S1-S4:

[0035] S1. Acquire the original vibration acceleration signal and operating status data, perform variational mode decomposition on the original vibration acceleration signal, and extract cavitation feature components.

[0036] In an optional embodiment, in an actual industrial setting, a high-frequency vibration acceleration sensor is placed in the non-load-bearing area of ​​the bearing to acquire the high-frequency vibration acceleration signal of the non-load-bearing area in real time as the raw vibration acceleration signal. Then, a wavelet denoising algorithm is used to preprocess the raw vibration acceleration signal to remove low-frequency baseline drift and high-frequency environmental white noise. Simultaneously, in conjunction with a photoelectric encoder and a three-dimensional force sensor, the operating status data, mainly including the bearing's real-time rotational speed and radial load, is acquired synchronously. Because the impact signal generated by cavitation collapse often has extremely short bandwidth characteristics, it is easily submerged in the huge bearing structural resonance and environmental background noise. Directly reading the time-domain waveform of the raw vibration acceleration signal makes it difficult to obtain the effective characteristics of cavitation; therefore, deep signal decoupling processing of the raw vibration acceleration signal is necessary.

[0037] Furthermore, to decouple the original vibration acceleration signal, the number of decomposition layers and penalty factor are pre-set, for example, five decomposition layers. The original vibration acceleration signal is adaptively decomposed into five intrinsic modal components using a variational mode decomposition algorithm. Subsequently, the kurtosis value of each intrinsic modal component is calculated. Given that the impact signal generated by cavitation collapse not only has extremely short broadband characteristics but also exhibits significant non-Gaussian impact characteristics, the frequency band with higher kurtosis values ​​contains more concentrated cavitation collapse features. Therefore, the kurtosis values ​​of each intrinsic modal component are directly compared, and the intrinsic modal component with the largest kurtosis value is selected and dimensionlessized as the cavitation feature component.

[0038] Thus, by precisely stripping and purifying the original vibration acceleration signal, the interference of redundant noise is effectively filtered out, and accurate cavitation characteristic components are obtained.

[0039] S2. Calculate the cavitation kinetic energy impact index based on the extracted cavitation feature components and operating status data.

[0040] In an optional embodiment, to accurately measure the kinetic energy impact intensity generated by cavitation collapse, simply observing the vibration amplitude of the cavitation characteristic components is insufficient; the nonlinear amplification effect of the real-time rotational speed in the bearing operating data on the cavitation generation and collapse frequencies must be taken into account. In physics, higher rotational speeds result in stronger shear forces within the fluid and more severe local pressure drops, leading to an exponential increase in the probability of cavitation formation. Based on this physical phenomenon, this invention combines the real-time rotational speed in the bearing operating data to construct a cavitation kinetic energy impact index, which smoothly and objectively reflects the generation intensity of cavitation within the lubricating oil film of the bearing under current operating conditions. Therefore, the calculation method for the cavitation kinetic energy impact index is as follows:

[0041]

[0042] in, The cavitation kinetic energy impact index is dimensionless. This is the energy scaling factor, which is dimensionless and is usually set to 100 to adjust the numerical magnitude to a range that is easy to calculate. This is the length of the sampling time window, in seconds, and its value is determined by the system sampling frequency. The signal is a normalized cavitation feature component signal, which is dimensionless; The bearing's real-time rotational speed, obtained from the encoder, is expressed in revolutions per minute. This refers to the rated reference speed of the bearing, expressed in revolutions per minute. The logarithmic reference constant is dimensionless and its value must be greater than or equal to 1. This ensures that the logarithmic function will not become meaningless when no vibration energy integral enters the system, and that the final calculated cavitation kinetic energy impact index will be zero.

[0043] It should be noted that this invention couples fluid cavitation dynamics with the unsteady elastohydrodynamic (EHL) theory, and the cavitation kinetic energy impact index... The construction follows the nonlinear law of cavitation nucleus explosion caused by fluid shear, i.e., the exponential term. This maps the nonlinear acceleration characteristics of the local flow field pressure drop exceeding the saturated vapor pressure when the bearing speed surges.

[0044] To ensure the strict comparability of the cavitation kinetic energy impact index calculated for the bearing at different speeds, this invention specifies the sampling time window length. Implement dynamic adjustments, that is, adjust according to the real-time speed of the bearing. The sampling time window length is updated in real time according to an inverse relationship. ,Right now ,in This is a constant related to the angle to ensure that the rotation angle of the shaft covered by each integration, i.e., the angular domain span, remains constant. This processing method can eliminate the influence of changes in bearing speed on the time-domain integration length, ensuring a constant angular domain sampling resolution, so that the calculated cavitation kinetic energy impact index can truly reflect the cavitation generation frequency of the bearing within a unit rotation cycle.

[0045] For example, suppose the sampling time window length is set. The rated reference speed of the bearing is 0.1 seconds. The speed is 1000 revolutions per minute. This is when the real-time speed of the bearing is detected. When the speed is increased to 1200 rpm, the exponential term This indicates that the vibrational energy is nonlinearly amplified by approximately 3.32 times at high speed; if at this time... Seconds, energy scaling factor The value is 100, the logarithmic reference constant. If the value is 1, then the cavitation kinetic energy impact index is... All parameters have a clear source and consistent dimensions.

[0046] It should be noted that the above calculations only illustrate the processing flow at a single moment. In actual continuous monitoring, the actual process will vary depending on the real-time rotational speed of the bearing. The monitoring system will adjust the monitoring based on the length of the sampling time window. With real-time speed The inverse relationship is updated in real time. For example, when the real-time rotational speed of the bearing doubles, the sampling time window length will be halved to ensure that the integration interval always covers the same bearing rotation angle.

[0047] Thus, by introducing nonlinear weighting of the real-time rotational speed of the bearing and performing logarithmic smoothing, the chaotic instantaneous high-frequency energy can be transformed into a stable and intuitively comparable cavitation kinetic energy impact index.

[0048] S3. Based on the cavitation kinetic energy impact index and the bearing contact area pressure estimated from the operating status data, dynamically correct the equivalent fluid density of the lubricating oil.

[0049] In one optional embodiment, conventional fluid dynamics models assume that the nominal density of lubricating oil is constant under standard atmospheric pressure. However, under actual heavy-load conditions, the equivalent fluid density of lubricating oil in a gas-liquid mixture must be re-derived by considering the inhibitory effect of bearing contact zone pressure on cavitation expansion. When the bearing contact zone pressure is extremely high, the high-pressure environment forcibly compresses the bubble volume, hindering gas expansion and making the macroscopic behavior of the mixed fluid closer to that of a pure liquid. Conversely, if the bearing contact zone pressure is low and the cavitation kinetic energy impact index spikes, the equivalent fluid density of the lubricating oil will drop significantly. Based on this, the equivalent fluid density of the lubricating oil is constructed, and its calculation method is as follows:

[0050]

[0051] in, This is the equivalent fluid density of the lubricating oil, expressed in kilograms per cubic meter. This refers to the nominal density of lubricating oil under standard atmospheric pressure, expressed in kilograms per cubic meter. The maximum theoretical cavitation fraction is dimensionless and represents the limit of the gas volume ratio of the lubricating oil film under extreme rupture conditions. The cavitation kinetic energy impact index; This is the contact area pressure of the bearing, measured in Pascals, and is estimated by dividing the radial load of the bearing by the effective bearing area. The pressure suppression coefficient, whose corresponding physical unit is the reciprocal of Pascal, enables the product term in the denominator to become a dimensionless value, ensuring that the dimensions on both sides of the relation remain consistent.

[0052] It should be noted that the equivalent fluid density correction model essentially constructs a dynamic equilibrium game-theoretic physical model of bubble formation and collapse. The molecular end of the above formula... This characterizes the driving force of cavitation evolution, while the high-pressure suppression term introduced at the denominator end... This conforms to the objective law in physics that environmental static pressure hinders bubble expansion. This bounded extremum function simulates the density degradation process of oil film gas-liquid two-phase flow.

[0053] For example, the nominal density of Mobil SHC 624 lubricating oil at standard atmospheric pressure. The maximum theoretical cavitation fraction is calibrated to 850 kg / m³. The pressure suppression coefficient is 0.7. Set as Pascal's reciprocal; the cavitation kinetic energy impact index at a certain moment. The value is 5, which is dimensionless; the system estimates the current contact pressure of the bearing. achieve Pascal; at this point, the pressure product term in the denominator... The result is dimensionless. Therefore, the corrected equivalent fluid density... Kilograms per cubic meter. This objective data reflects that the lubricating oil film has undergone significant vaporization and expansion.

[0054] Thus, by introducing the pressure in the bearing contact area to suppress cavitation expansion, the illusion of excessive correction of the equivalent fluid density caused by strong vibration in the high-pressure and heavy-load area of ​​the lubricating oil can be avoided, effectively ensuring the accuracy of the physical parameter derivation.

[0055] S4. Use the corrected equivalent fluid density to solve for the minimum oil film thickness of the lubricating oil film, and combine it with the bearing temperature rise rate to determine the lubrication safety margin index.

[0056] In an optional embodiment, once an equivalent fluid density that more closely reflects physical reality is obtained, the system substitutes it into the discretized Reynolds equation and uses the finite difference method to iteratively solve for the minimum oil film thickness under the current condition. However, relying solely on the minimum oil film thickness is insufficient to comprehensively assess failure risk, because even if the minimum oil film thickness is within a reasonable range, if the bearing experiences localized dry friction leading to a rapid temperature rise, the equipment still faces a very high risk of thermal damage. Therefore, the thermodynamic penalty factor of the bearing's temperature rise rate must be incorporated into the final evaluation system. Based on this, a lubrication safety margin index is constructed, and its calculation method is as follows:

[0057]

[0058] in, It is a lubrication safety margin index, which is dimensionless; The minimum oil film thickness calculated based on the equivalent fluid density, expressed in micrometers; , where is the root mean square value of the surface roughness of the inner ring raceway of the bearing, and the unit is micrometer; , where is the root mean square value of the surface roughness of the rolling elements of the bearing, and the unit is micrometer; The temperature rise rate of the bearing at the current moment is expressed in degrees Celsius per second. This temperature rise rate is obtained by extracting the slope through least squares linear fitting of continuously collected bearing temperature data within a sliding time window, in order to eliminate the interference of high-frequency temperature fluctuation noise on the temperature rise rate calculation. The maximum permissible temperature rise rate threshold for the bearing is defined in degrees Celsius per second. The fractional structure in the first half of the formula is dimensionless after division because both the numerator and denominator are in micrometers, representing the film thickness ratio, which reflects the safety isolation at the geometric level. The second half is dimensionless because the units in the exponent cancel each other out, constituting the thermodynamic penalty factor determined by the bearing's temperature rise rate.

[0059] It should be noted that the lubrication safety margin index The classic film thickness ratio at the microscopic geometric level and the nonlinear thermal decay penalty term at the macroscopic thermodynamic level are orthogonally integrated, filling the algorithmic blind spot of traditional static fluid lubrication models that cannot cope with heavy-load sudden operating conditions from the mathematical foundation.

[0060] For example, assume that the minimum oil film thickness obtained is... The root mean square value of the surface roughness of the bearing inner ring raceway is 1.2 micrometers. The root mean square value of the surface roughness of the bearing rolling elements is 0.3 micrometers. If the thickness is 0.4 micrometers, then the square root of the sum of the squares of the denominators is exactly 0.5 micrometers. At this point, the geometric film thickness ratio of the first half is 2.4, which is within the safe range based on engineering experience. However, if the temperature rise rate of the bearing is monitored at this point... The temperature rise rate reaches 2 degrees Celsius per second, which is the maximum allowable temperature rise rate threshold for the bearing set by the system. If the temperature is 4 degrees Celsius per second, then the lubrication safety margin index is... If the bearing temperature continues to rise uncontrollably, and the lubrication safety margin index falls below the set threshold, a system alarm will be triggered to indicate that the current bearing lubrication status is in a high-risk range.

[0061] In this way, by integrating bearing surface roughness, minimum oil film thickness, and bearing temperature rise rate, the monitoring system can be endowed with a keen early risk warning capability, ensuring stable equipment operation.

[0062] Reference Figure 2 In a three-dimensional Cartesian coordinate system with bearing real-time speed, radial load, and cavitation kinetic energy impact index as axes, a cluster of sample points representing a high risk of cavitation in the lubricating oil can be clearly observed. These clusters are concentrated in the operating region where both bearing real-time speed and radial load are high, and exhibit significant numerical jumps and aggregations along the cavitation kinetic energy impact index axis. Conversely, samples representing a stable lubrication state are steadily distributed in the low-value region. This directly demonstrates that the present invention can effectively isolate bearing fault states in a multi-dimensional space.

[0063] Reference Figure 3 The figure shows the pressure fluctuations of the bearing on the circumferential surface. As can be seen from the figure, the oil film pressure peak is normal in the maximum load area of ​​the bearing, while the oil film pressure field is severely distorted in the cavitation collapse area, which objectively reflects the oil film pressure distortion law caused by cavitation.

[0064] It should be noted that those skilled in the art can make various modifications and improvements without departing from the inventive concept, and these all fall within the scope of protection of this invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A method for real-time monitoring of bearing lubrication status based on a modified fluid dynamics model, characterized in that, include: The original vibration acceleration signal and operating status data are acquired, and variational mode decomposition is performed on the original vibration acceleration signal to extract cavitation feature components. Based on the extracted cavitation feature components and the operational status data, the cavitation kinetic energy impact index is calculated. ,satisfy: This is the energy scaling factor; This is the length of the sampling time window; The signal represents the cavitation feature components after normalization. The real-time rotational speed of the bearing is obtained by the encoder; This refers to the rated reference speed of the bearing; The logarithmic reference constant; Based on the cavitation kinetic energy impact index and the bearing contact area pressure estimated from the operating state data, the equivalent fluid density of the lubricating oil is dynamically corrected. ， satisfy: This is the nominal density of the lubricating oil under standard atmospheric pressure. This represents the maximum theoretical void fraction. The cavitation kinetic energy impact index; This refers to the pressure in the contact area of ​​the bearing. This is the pressure suppression coefficient; The minimum oil film thickness is determined using the corrected equivalent fluid density, and the lubrication safety margin index is determined in conjunction with the bearing's temperature rise rate. ,for: This is the minimum oil film thickness calculated based on the equivalent fluid density; This is the root mean square value of the surface roughness of the inner ring raceway of the bearing. This is the root mean square value of the surface roughness of the bearing rolling elements; This represents the current rate of temperature rise of the bearing. This is the threshold value for the maximum allowable rate of temperature rise in the bearing.

2. The method for real-time monitoring of bearing lubrication status based on fluid dynamics model correction according to claim 1, characterized in that, The steps of acquiring the original vibration acceleration signal and operating status data, performing variational mode decomposition on the original vibration acceleration signal, and extracting cavitation feature components include: acquiring the high-frequency vibration acceleration signal of the non-load-bearing area as the original vibration acceleration signal; setting the number of decomposition layers and penalty factor, and decomposing the original vibration acceleration signal into multiple intrinsic mode components through a variational mode decomposition algorithm; calculating the kurtosis value of each intrinsic mode component, and selecting the intrinsic mode component with the largest kurtosis value as the cavitation feature component.

3. The method for real-time monitoring of bearing lubrication status based on fluid dynamics model correction according to claim 1, characterized in that, The contact area pressure of the bearing is estimated by dividing the radial load of the bearing by the effective bearing area.

4. The real-time monitoring method for bearing lubrication status based on fluid dynamics model correction according to claim 1, characterized in that, Before performing variational mode decomposition on the original vibration acceleration signal, a wavelet denoising algorithm is used to preprocess the original vibration acceleration signal to remove low-frequency baseline drift and high-frequency environmental white noise.

5. The method for real-time monitoring of bearing lubrication status based on fluid dynamics model correction according to claim 1, characterized in that, The sampling time window length is dynamically adjusted based on the real-time rotational speed of the bearing.

6. The method for real-time monitoring of bearing lubrication status based on fluid dynamics model correction according to claim 1, characterized in that, The temperature rise rate of the bearing is obtained by extracting the slope through least squares linear fitting of continuously collected bearing temperature data within a sliding time window.

7. The method for real-time monitoring of bearing lubrication status based on fluid dynamics model correction according to claim 1, characterized in that, When the lubrication safety margin index is lower than the set threshold, a system alarm is triggered, indicating that the current bearing lubrication status is in a high-risk range.

Images