A method for evaluating the sealing performance degradation of a speed reducer based on an intelligent sensor

Abstract

This invention relates to the field of equipment condition monitoring technology and discloses a method for evaluating the degradation of reducer sealing performance based on intelligent sensors. This method simultaneously collects reducer air pressure, oil temperature, and external acoustic vibration signals; calculates a two-phase distortion correction factor based on temperature changes; calculates the vapor-liquid coupling pressure fluctuation rate after removing thermal expansion interference by combining the air pressure and temperature time change rates; calculates the aeroacoustic energy gradient by calculating the acoustic signal amplitude change; calculates the viscosity-temperature acoustic distortion compensation factor by combining a preset damping coefficient; when the air pressure fluctuates negatively, multiplies its absolute value by the acoustic energy gradient and the compensation factor to obtain a cross-domain leakage mapping index; performs logarithmic weighting and global average accumulation on this index to output the severity of seal degradation, effectively filtering out mechanical noise and temperature drift distortion interference.

Description

Technical Field

[0001] This invention relates to the field of equipment condition monitoring technology, specifically to a method for evaluating the degradation of reducer sealing performance based on intelligent sensors. Background Technology

[0002] As a core component of a mechanical transmission system, the reducer is usually filled with lubricating oil to ensure the normal operation of the gear set. In order to prevent lubricating oil leakage and the intrusion of external impurities, the reducer output shaft and other positions are equipped with sealing structures. With the long-term high-load operation of the equipment, the sealing lip will inevitably wear and degrade, which will lead to micro-leakage.

[0003] Currently, conventional technologies for monitoring the sealing performance of speed reducers mostly rely on single internal air pressure threshold alarms or external acoustic vibration inspections. However, in actual high-speed, heavy-load operating scenarios, the meshing of gears inside the speed reducer generates strong structural high-frequency noise. This background mechanical noise can easily completely mask the high-frequency acoustic signals of aerodynamic micro-leakage generated in the early stages of minor degradation of the sealing lip, resulting in extremely low sensitivity and easy failure of simple acoustic monitoring methods. To help isolate the masked signals, some existing technologies attempt to introduce internal air pressure changes for joint judgment. However, in actual operating conditions, the internal temperature rise of the speed reducer causes the lubricating oil to undergo a vapor-liquid two-phase transformation, and the resulting nonlinear thermal expansion effect leads to internal cavities. The air pressure rises significantly; this air pressure change caused by thermal expansion overlaps with the air pressure drop caused by micro-leakage, making it impossible for traditional static signal correlation decomposition methods to accurately separate the real air pressure fluctuations caused purely by mass loss, which easily leads to false alarms or missed alarms; in addition, a deeper problem that is often overlooked by existing technologies is that the dynamic temperature rise of the lubricating oil inside the reducer will cause its kinematic viscosity to drop significantly, thereby changing the medium damping attenuation rate when the sound waves from the internal micro-leakage are transmitted to the external casing; this dynamic change in medium damping will cause physical amplification or transmission distortion of the acoustic signal collected under high temperature and high operating load, which seriously destroys the physical mapping relationship between acoustic characteristics and the actual severity of leakage;

[0004] Therefore, existing assessment methods cannot effectively eliminate the interference of gear operation heat generation on air pressure judgment, cannot accurately quantify the change in sound wave transmission damping caused by dynamic temperature rise of lubricating oil, and cannot achieve pure extraction of cross-physical domain signal synchronous noise reduction and micro-leakage characteristics under working conditions with strong mechanical noise and complex temperature drift distortion. Thus, it is difficult to accurately and quantitatively assess the degradation state of reducer sealing performance. Summary of the Invention

[0005] This invention provides a method for evaluating the degradation of reducer sealing performance based on intelligent sensors, which helps to solve the problems mentioned in the background art.

[0006] This invention provides the following technical solution: a method for evaluating the degradation of reducer sealing performance based on intelligent sensors, comprising:

[0007] During the monitoring period, the internal air pressure, lubricating oil temperature and external acoustic vibration signals of the reducer are collected synchronously at a set frequency.

[0008] Based on the initial and real-time lubricating oil temperatures, combined with the preset thermal expansion characteristic coefficient, the two-phase distortion correction factor is calculated.

[0009] Calculate the time rate of change of internal air pressure and the time rate of change of lubricating oil temperature at adjacent sampling times;

[0010] The vapor-liquid coupling pressure fluctuation rate is calculated by combining the initial internal air pressure, the initial lubricating oil temperature, the two-phase distortion correction factor, and the above two change rates.

[0011] The aeroacoustic energy gradient is obtained by calculating the absolute value of the time-varying rate of change of the squared difference of the acoustic vibration signal amplitude between adjacent sampling times.

[0012] Based on the initial and real-time lubricating oil temperatures, and combined with the preset viscosity-temperature acoustic damping coefficient, the viscosity-temperature acoustic distortion compensation factor is calculated.

[0013] The sign of the vapor-liquid coupling pressure fluctuation rate is determined. When it is a negative fluctuation, its absolute value is multiplied by the aeroacoustic energy gradient and the viscosity-temperature acoustic distortion compensation factor to obtain the cross-domain leakage mapping index.

[0014] Logarithmically weighted the cross-domain leakage mapping index at each sampling time within the monitoring period, and then globally averaged and accumulated the results to output the severity of reducer seal degradation.

[0015] Optionally, the step of synchronously collecting the internal air pressure, lubricating oil temperature, and external acoustic vibration signals of the reducer at a set frequency during the monitoring period includes:

[0016] A dynamic air pressure sensor is installed in the cavity inside the reducer to measure the internal air pressure; a temperature sensor is installed in the lubricating oil sump inside the reducer to measure the lubricating oil temperature; a high-frequency acoustic emission sensor is installed on the outer housing of the sealing lip of the reducer output shaft to measure the acoustic vibration signal at the seal.

[0017] Set the total number of time node samples and the preset acquisition frequency; calculate and determine the time sampling interval based on the reciprocal relationship of the preset acquisition frequency, and establish the corresponding discrete time node sequence;

[0018] Based on the set discrete time node sequence and time sampling interval, the internal air pressure sequence, lubricating oil temperature sequence and acoustic vibration signal sequence are synchronously acquired through the sensor at each time node.

[0019] Optionally, the step of calculating the two-phase distortion correction factor based on the initial time and the real-time collected lubricating oil temperature, combined with a preset thermal expansion characteristic coefficient, includes:

[0020] Before the monitoring cycle begins, the standard volume expansion coefficient of the lubricating oil inside the reducer and the volume ratio of the lubricating oil volume to the upper air cavity volume when the reducer is in a static state are obtained; the standard volume expansion coefficient and the volume ratio are multiplied to obtain the preset thermal expansion characteristic coefficient.

[0021] Calculate the difference between the real-time collected lubricating oil temperature and the lubricating oil temperature measured at the initial time; divide the difference by the lubricating oil temperature measured at the initial time to obtain the relative temperature change.

[0022] The relative temperature change is multiplied by the preset thermal expansion characteristic coefficient, and the product is used as the exponent of the natural constant for exponential operation to calculate the two-phase distortion correction factor.

[0023] Optionally, the calculation of the time change rate of internal air pressure and the time change rate of lubricating oil temperature at adjacent sampling times includes:

[0024] Calculate the difference in internal air pressure between the current time node and the previous time node, and divide the difference in internal air pressure by the time sampling interval to obtain the rate of change of internal air pressure over time.

[0025] Calculate the difference in lubricating oil temperature between the current time node and the previous time node, and divide the difference in lubricating oil temperature by the time sampling interval to obtain the lubricating oil temperature change rate over time.

[0026] Optionally, the calculation of the vapor-liquid coupled pressure fluctuation rate by combining the initial internal pressure, the initial lubricating oil temperature, the two-phase distortion correction factor, and the aforementioned two rates of change includes:

[0027] The initial state ratio is obtained by dividing the internal air pressure measured at the initial moment by the lubricating oil temperature measured at the initial moment.

[0028] Multiply the initial state ratio, the two-phase distortion correction factor, and the lubricating oil temperature-time change rate together to obtain the theoretical value of thermal expansion pressure change.

[0029] The vapor-liquid coupling pressure fluctuation rate is calculated by subtracting the theoretical value of the thermal expansion pressure change from the internal pressure time change rate.

[0030] Optionally, the step of calculating the absolute value of the time-varying rate of change of the squared difference in the amplitude of the acoustic vibration signal between adjacent sampling times to obtain the aeroacoustic energy gradient includes:

[0031] Calculate the square of the acoustic vibration signal amplitude at the current time point and the previous time point respectively;

[0032] Calculate the difference between the squared value at the current time point and the squared value at the previous time point;

[0033] Divide the difference by the time sampling interval to obtain the acoustic energy time-varying rate;

[0034] The aeroacoustic energy gradient is calculated by taking the absolute value of the time-varying rate of acoustic energy.

[0035] Optionally, the step of calculating the viscosity-temperature acoustic distortion compensation factor based on the initial time and the real-time collected lubricating oil temperature, combined with a preset viscosity-temperature acoustic damping coefficient, includes:

[0036] Before the monitoring cycle begins, the standard kinematic viscosity values ​​of the lubricating oil at different standard temperatures are obtained, and the viscosity-temperature characteristic slope of the lubricating oil is calculated according to the viscosity-temperature characteristic formula; the acoustic transmission loss coefficient of the reducer housing material is obtained; the absolute value of the viscosity-temperature characteristic slope is multiplied by the acoustic transmission loss coefficient to obtain the preset viscosity-temperature acoustic damping coefficient.

[0037] Calculate the difference between the lubricating oil temperature measured at the initial moment and the lubricating oil temperature collected in real time, and divide the difference by the lubricating oil temperature collected in real time to obtain the reverse relative temperature deviation.

[0038] The viscosity-temperature acoustic distortion compensation factor is calculated by multiplying the reverse relative temperature deviation by the preset viscosity-temperature acoustic damping coefficient and using the product as the exponent of the natural constant.

[0039] Optionally, the sign of the vapor-liquid coupling pressure fluctuation rate is determined. When it is a negative fluctuation, its absolute value is multiplied by the aeroacoustic energy gradient and the viscosity-temperature acoustic distortion compensation factor to obtain the transdomain leakage mapping index, including:

[0040] The sign direction value of the vapor-liquid coupled pressure fluctuation rate is extracted using a sign function;

[0041] Subtract the sign direction value from one and divide the result by two to construct a retention coefficient, which is such that when the vapor-liquid coupling pressure fluctuation rate is negative, it takes a value of one, and otherwise takes a value of zero.

[0042] The cross-domain leakage mapping index is calculated by sequentially multiplying the retention coefficient, the absolute value of the vapor-liquid coupling pressure fluctuation rate, the aeroacoustic energy gradient, and the viscosity-temperature acoustic distortion compensation factor.

[0043] Optionally, the step of logarithmically weighting the cross-domain leakage mapping index at each sampling time within the monitoring period and performing a global average cumulative calculation to output the severity of reducer seal degradation includes:

[0044] The cross-domain leakage mapping exponent at each valid sampling time is added to the natural constant to obtain the sum value;

[0045] Calculate the natural logarithm of the sum to obtain the logarithmic weighting coefficients;

[0046] Multiply the cross-domain leakage mapping index at the corresponding sampling time with the logarithmic weighting coefficient to obtain the weighted mapping feature value at each sampling time;

[0047] The weighted mapping feature values ​​of all sampling times excluding the initial time within the monitoring period are summed to obtain the global accumulated value;

[0048] The global cumulative value is averaged by the total number of time node samples, and the calculated severity of the reducer seal degradation is output.

[0049] The present invention has the following beneficial effects:

[0050] 1. In the specific and complex operating environment of the reducer, the heat generated by gear operation not only causes nonlinear thermal expansion of the lubricating oil, thus interfering with the internal air pressure judgment, but also causes temperature drift distortion of the acoustic propagation medium. In addition, the complex internal mechanical structure noise can easily mask the high-frequency aerodynamic step signal generated by early high-pressure gas micro-leakage. Therefore, this solution simultaneously collects internal air pressure, oil temperature and external acoustic signals. By calculating the two-phase distortion correction factor and viscosity-temperature acoustic distortion compensation factor, cross-domain mapping feature extraction is performed only when the air pressure shows a real negative fluctuation. This design can effectively eliminate the physical interference of gear operation heat generation and completely filter out asynchronous mechanical structure noise under the specific working conditions of strong noise and high dynamic temperature change. It can accurately capture the transient change characteristics of micro-leakage at the seal and finally realize the reliable quantification and output of the severity of reducer seal degradation.

[0051] 2. By setting discrete time node numbers and time sampling intervals according to preset acquisition frequencies, continuous analog physical signals are converted into standardized digital time series. Setting a reasonable total number of time node samples and acquisition frequency can cover a sufficiently long physical running time to include more sporadic micro-leakage pulse events, making the global evaluation results more statistically smooth and representative, and can also keenly capture ultra-high frequency aeroacoustic transient pulses, effectively improving the perception limit of early micro-leakage physical characteristics.

[0052] 3. By pre-obtaining the standard volumetric expansion coefficient of the lubricating oil and the ratio of the static lubricating oil volume inside the reducer to the upper air cavity volume, the thermal expansion characteristic coefficient is calculated. This calculation process organically combines the inherent thermodynamic properties of the fluid with the geometric proportions of the specific cavity space of the reducer, accurately quantifying the degree of compression of the internal sealed air chamber space by the heated volume expansion of the lubricating oil under specific operating conditions, and establishing an objective physical benchmark for accurately assessing the nonlinear influence of temperature fluctuations on air pressure. By utilizing the relative change amplitude of lubricating oil temperature at the initial moment and in real time, and combining it with the thermal expansion characteristic coefficient for exponential calculation, a two-phase distortion correction factor is obtained. This process dynamically quantifies the nonlinear thermal expansion effect of the lubricating oil inside the reducer due to the operating temperature rise, accurately characterizing the theoretical change law of air pressure under complex thermodynamic conditions, thus effectively eliminating the interference of natural pressure inflation caused by system temperature rise in air pressure characteristic analysis.

[0053] 4. By performing first-order backward difference calculations on the internal air pressure and lubricating oil temperature at adjacent sampling times, the time change rate of air pressure and the time change rate of temperature are obtained. This processing method, which transforms absolute static measurement into relative dynamic rate, can keenly capture the evolution trend of physical signals within a very short time window, effectively reduce the measurement error caused by the static baseline drift of the sensor, highlight the dynamic fluctuation characteristics of the signal, and provide a direct rate characterization for deconstructing the transient thermo-pressure coupling changes in the sealed cavity.

[0054] 5. By combining the initial internal pressure-temperature ratio, two-phase distortion correction factor, and temperature change rate, the theoretical influence of thermal expansion is accurately deducted from the overall pressure-time change rate, and the pure vapor-liquid coupling pressure fluctuation rate is calculated. This process completely eliminates the serious physical interference of gear operation heat generation on the judgment of air pressure in the sealed air chamber. It can restore the real air pressure drop caused only by the loss of gas mass at the sealing lip from the complex temperature and pressure aliasing signal, greatly improving the objective accuracy of air pressure monitoring and judgment of micro-leakage.

[0055] 6. By calculating the squared difference of the absolute amplitude of the acoustic vibration signal at adjacent sampling times and dividing it by the time sampling interval, the aeroacoustic energy gradient is obtained. This calculation step converts the ordinary acoustic amplitude signal into the mutation rate of the transient energy pulse, which can significantly amplify the aeroacoustic high-frequency energy step characteristics generated by the instantaneous micro-leakage of high-pressure gas. It effectively overcomes the masking effect of smooth mechanical background noise on small abnormal signals and realizes high-sensitivity extraction of the transient characteristics of weak impact of high-frequency airflow.

[0056] 7. By obtaining the standard kinematic viscosity values ​​of lubricating oil at different standard temperatures to calculate the viscosity-temperature characteristic slope, and combining it with the acoustic transmission loss coefficient of the reducer housing material, a viscosity-temperature acoustic distortion compensation factor for the variation of medium damping was constructed. This design fully considers the viscosity decrease phenomenon of lubricating oil caused by dynamic temperature rise, accurately quantifies the damping attenuation law of sound wave transmission medium caused by temperature change, and effectively eliminates the false amplification or excessive attenuation interference of medium viscosity attenuation on the transmission intensity of high-frequency micro-leakage acoustic signals under high temperature conditions.

[0057] 8. By determining the sign direction of the gas pressure fluctuation rate of the vapor-liquid coupling, a nonlinear filtering mapping mechanism is constructed using the sign function. Only when the gas pressure shows a true negative fluctuation is the absolute value of the gas pressure fluctuation multiplied and fused with the acoustic energy gradient and distortion compensation factor. This step utilizes the unique physical synchronization of the instantaneous drop in internal gas pressure and the sudden increase in external acoustic energy that inevitably accompanies the occurrence of micro-leakage, realizing joint signal verification across physical domains and completely filtering out internal mechanical structure noise interference such as gear meshing that is asynchronous with the leak.

[0058] 9. By logarithmically weighting the cross-domain leakage mapping index extracted at each effective sampling time within the monitoring period and performing global average accumulation calculation, the overall sealing degradation severity of the system is finally output. Nonlinear logarithmic weighting effectively smooths the discrete and extreme individual leakage pulse spikes and reasonably amplifies the cumulative effect of continuous micro-degradation, while global average accumulation integrates the characteristics of all micro-leakage events within the entire period, so that the final output degradation severity index not only has high statistical stability, but also reflects the macroscopic physical degradation state of the equipment intuitively and quantitatively. Attached Figure Description

[0059] Figure 1 This is a schematic diagram of the basic process of the present invention. Detailed Implementation

[0060] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0061] Example 1, refer to Figure 1 A method for evaluating the degradation of reducer sealing performance based on intelligent sensors, comprising:

[0062] During the monitoring period, the internal air pressure, lubricating oil temperature and external acoustic vibration signals of the reducer are collected synchronously at a set frequency.

[0063] Based on the initial and real-time lubricating oil temperatures, combined with the preset thermal expansion characteristic coefficient, the two-phase distortion correction factor is calculated.

[0064] Calculate the time rate of change of internal air pressure and the time rate of change of lubricating oil temperature at adjacent sampling times;

[0065] The vapor-liquid coupling pressure fluctuation rate is calculated by combining the initial internal air pressure, the initial lubricating oil temperature, the two-phase distortion correction factor, and the above two change rates.

[0066] The aeroacoustic energy gradient is obtained by calculating the absolute value of the time-varying rate of change of the squared difference of the acoustic vibration signal amplitude between adjacent sampling times.

[0067] Based on the initial and real-time lubricating oil temperatures, and combined with the preset viscosity-temperature acoustic damping coefficient, the viscosity-temperature acoustic distortion compensation factor is calculated.

[0068] The sign of the vapor-liquid coupling pressure fluctuation rate is determined. When it is a negative fluctuation, its absolute value is multiplied by the aeroacoustic energy gradient and the viscosity-temperature acoustic distortion compensation factor to obtain the cross-domain leakage mapping index.

[0069] Logarithmically weighted the cross-domain leakage mapping index at each sampling time within the monitoring period, and then globally averaged and accumulated the results to output the severity of reducer seal degradation.

[0070] The monitoring cycle includes synchronously collecting the internal air pressure, lubricating oil temperature, and external acoustic vibration signals of the reducer at a set frequency, including:

[0071] A dynamic air pressure sensor is installed in the cavity inside the reducer to measure the internal air pressure; a temperature sensor is installed in the lubricating oil sump inside the reducer to measure the lubricating oil temperature; a high-frequency acoustic emission sensor is installed on the outer housing of the sealing lip of the reducer output shaft to measure the acoustic vibration signal at the seal.

[0072] Set the total number of time node samples and the preset acquisition frequency; calculate and determine the time sampling interval based on the reciprocal relationship of the preset acquisition frequency, and establish the corresponding discrete time node sequence;

[0073] Based on the set discrete time node sequence and time sampling interval, the internal air pressure sequence, lubricating oil temperature sequence and acoustic vibration signal sequence are synchronously acquired through the sensor at each time node.

[0074] The calculation of the two-phase distortion correction factor based on the initial and real-time lubricating oil temperatures, combined with a preset thermal expansion characteristic coefficient, includes:

[0075] Before the monitoring cycle begins, the standard volume expansion coefficient of the lubricating oil inside the reducer and the volume ratio of the lubricating oil volume to the upper air cavity volume when the reducer is in a static state are obtained; the standard volume expansion coefficient and the volume ratio are multiplied to obtain the preset thermal expansion characteristic coefficient.

[0076] Calculate the difference between the real-time collected lubricating oil temperature and the lubricating oil temperature measured at the initial time; divide the difference by the lubricating oil temperature measured at the initial time to obtain the relative temperature change.

[0077] The relative temperature change is multiplied by the preset thermal expansion characteristic coefficient, and the product is used as the exponent of the natural constant for exponential operation to calculate the two-phase distortion correction factor.

[0078] The calculation of the time change rate of internal air pressure and the time change rate of lubricating oil temperature at adjacent sampling times includes:

[0079] Calculate the difference in internal air pressure between the current time node and the previous time node, and divide the difference in internal air pressure by the time sampling interval to obtain the rate of change of internal air pressure over time.

[0080] Calculate the difference in lubricating oil temperature between the current time node and the previous time node, and divide the difference in lubricating oil temperature by the time sampling interval to obtain the lubricating oil temperature change rate over time.

[0081] The calculation of the vapor-liquid coupled pressure fluctuation rate, combining the initial internal air pressure, initial lubricating oil temperature, the two-phase distortion correction factor, and the aforementioned two rates of change, includes:

[0082] The initial state ratio is obtained by dividing the internal air pressure measured at the initial moment by the lubricating oil temperature measured at the initial moment.

[0083] Multiply the initial state ratio, the two-phase distortion correction factor, and the lubricating oil temperature-time change rate together to obtain the theoretical value of thermal expansion pressure change.

[0084] The vapor-liquid coupling pressure fluctuation rate is calculated by subtracting the theoretical value of the thermal expansion pressure change from the internal pressure time change rate.

[0085] The process of calculating the absolute value of the time-varying rate of change of the squared difference in the amplitude of the acoustic vibration signal between adjacent sampling times to obtain the aeroacoustic energy gradient includes:

[0086] Calculate the square of the acoustic vibration signal amplitude at the current time point and the previous time point respectively;

[0087] Calculate the difference between the squared value at the current time point and the squared value at the previous time point;

[0088] Divide the difference by the time sampling interval to obtain the acoustic energy time-varying rate;

[0089] The aeroacoustic energy gradient is calculated by taking the absolute value of the time-varying rate of acoustic energy.

[0090] The step of calculating the viscosity-temperature acoustic distortion compensation factor based on the initial and real-time lubricating oil temperatures, combined with a preset viscosity-temperature acoustic damping coefficient, includes:

[0091] Before the monitoring cycle begins, the standard kinematic viscosity values ​​of the lubricating oil at different standard temperatures are obtained, and the viscosity-temperature characteristic slope of the lubricating oil is calculated according to the viscosity-temperature characteristic formula; the acoustic transmission loss coefficient of the reducer housing material is obtained; the absolute value of the viscosity-temperature characteristic slope is multiplied by the acoustic transmission loss coefficient to obtain the preset viscosity-temperature acoustic damping coefficient.

[0092] Calculate the difference between the lubricating oil temperature measured at the initial moment and the lubricating oil temperature collected in real time, and divide the difference by the lubricating oil temperature collected in real time to obtain the reverse relative temperature deviation.

[0093] The viscosity-temperature acoustic distortion compensation factor is calculated by multiplying the reverse relative temperature deviation by the preset viscosity-temperature acoustic damping coefficient and using the product as the exponent of the natural constant.

[0094] The sign of the vapor-liquid coupling pressure fluctuation rate is determined by multiplying its absolute value by the aeroacoustic energy gradient and the viscosity-temperature acoustic distortion compensation factor when it is negative, to obtain the transdomain leakage mapping index, including:

[0095] The sign direction value of the vapor-liquid coupled pressure fluctuation rate is extracted using a sign function;

[0096] Subtract the sign direction value from one and divide the result by two to construct a retention coefficient, which is such that when the vapor-liquid coupling pressure fluctuation rate is negative, it takes a value of one, and otherwise takes a value of zero.

[0097] The cross-domain leakage mapping index is calculated by sequentially multiplying the retention coefficient, the absolute value of the vapor-liquid coupling pressure fluctuation rate, the aeroacoustic energy gradient, and the viscosity-temperature acoustic distortion compensation factor.

[0098] The cross-domain leakage mapping index at each sampling time within the monitoring period is logarithmically weighted and then globally averaged and accumulated to output the severity of reducer seal degradation, including:

[0099] The cross-domain leakage mapping exponent at each valid sampling time is added to the natural constant to obtain the sum value;

[0100] Calculate the natural logarithm of the sum to obtain the logarithmic weighting coefficients;

[0101] Multiply the cross-domain leakage mapping index at the corresponding sampling time with the logarithmic weighting coefficient to obtain the weighted mapping feature value at each sampling time;

[0102] The weighted mapping feature values ​​of all sampling times excluding the initial time within the monitoring period are summed to obtain the global accumulated value;

[0103] The global accumulated value is averaged by the total number of time node samples to output the calculated severity of reducer seal degradation. In the specific complex environment of reducer operation, the heat generated by gear operation not only causes nonlinear thermal expansion of lubricating oil, thus interfering with the internal air pressure judgment, but also causes temperature drift distortion of the acoustic propagation medium. Coupled with the complex internal mechanical structure noise, these factors can easily mask the high-frequency aerodynamic step signal generated by early high-pressure gas micro-leakage. Therefore, this scheme simultaneously collects internal air pressure, oil temperature and external acoustic signals. By calculating the two-phase distortion correction factor and viscosity-temperature acoustic distortion compensation factor, cross-domain mapping feature extraction is performed only when the air pressure shows a real negative fluctuation. This design can effectively eliminate the physical interference of gear operation heat generation and completely filter out asynchronous mechanical structure noise under the specific working conditions of strong noise and high dynamic temperature change, accurately capture the transient change characteristics of micro-leakage at the seal, and finally realize the reliable quantification and output of the severity of reducer seal degradation.

[0104] Example 2: A method for evaluating the degradation of reducer sealing performance based on intelligent sensors, further comprising:

[0105] The monitoring cycle includes synchronously collecting the internal air pressure, lubricating oil temperature, and external acoustic vibration signals of the reducer at a set frequency, including:

[0106] A dynamic air pressure sensor is installed in the cavity inside the reducer to measure the internal air pressure. A temperature sensor is installed in the lubricating oil sump inside the reducer to measure the lubricating oil temperature. A high-frequency acoustic emission sensor is installed on the outer housing of the sealing lip of the reducer output shaft to measure the acoustic vibration signal at the seal.

[0107] Set discrete time node numbers Its value range is ,in This represents the total number of time points sampled within the monitoring period, with a value of [value to be filled in]. to , A larger value can cover a longer physical operating time, including more possible micro-leakage pulse events, making the global average cumulative result more statistically smooth and representative. However, an excessively large value will lead to an excessively long monitoring period, and irreversible macroscopic drift in the base air pressure and base temperature inside the reducer at the beginning and end of the period. The smaller the value, the faster the calculation speed and the higher the real-time response of the system. However, if the value is too small, the monitoring period will be too short. In a single period, it may be impossible to capture the occasional micro-leakage pulses with statistical significance, resulting in a large random fluctuation error in the calculated severity of degradation.

[0108] Set time sampling interval The preset sampling frequency calculate, , Values to , A larger value allows for the capture of finer ultra-high frequency aeroacoustic transient pulses, improving the physical sensing limit for early, minute leaks. However, excessively large values ​​can lead to a dramatic increase in the system's hardware sampling load, generating massive amounts of redundant data. Furthermore, they may introduce electronic background noise beyond the actual leak frequency band into the calculation, interfering with the energy gradient solution. The smaller the value, the less pressure on data storage and computation. However, if the value is too small, such as below the lower limit of the acoustic emission characteristic frequency, it will cause Nyquist aliasing, completely lose the high-frequency transient energy step characteristics at the moment of micro-leakage, make the aeroacoustic energy gradient fail, and thus cause the entire mapping scheme to collapse.

[0109] Based on the node sequence number and time sampling interval set above, the internal air pressure sequence is synchronously acquired through the set sensors at each time node. Lubricating oil temperature sequence and acoustic vibration signal sequences By setting discrete time node numbers and time sampling intervals according to preset acquisition frequencies, continuous analog physical signals are converted into standardized digital time series. Setting a reasonable total number of time node samples and acquisition frequency can cover a sufficiently long physical running time to include more sporadic micro-leakage pulse events, making the global evaluation results more statistically smooth and representative, and can also keenly capture ultra-high frequency aeroacoustic transient pulses, effectively improving the perception limit of early micro-leakage physical characteristics.

[0110] The calculation of the two-phase distortion correction factor based on the initial and real-time lubricating oil temperatures, combined with a preset thermal expansion characteristic coefficient, includes:

[0111] Calculate the two-phase distortion correction factor and quantify the nonlinear thermal expansion effect of the lubricating oil inside the reducer due to temperature rise:

[0112] in: : No. Two-phase distortion correction factor at each time point; An exponential function with the natural constant e as its base; No. The lubricating oil temperature measured at each time point; The lubricating oil temperature measured at the first time point; The coefficient of thermal expansion of the lubricating oil can be obtained directly from the lubricating oil's physical property table or the manufacturer's technical specifications before the monitoring cycle, specifying the standard volumetric expansion coefficient of the lubricating oil used in the reducer. ,unit: Simultaneously, the volume of lubricating oil inside the reducer at static conditions is obtained. Volume of the upper air cavity ratio The thermal expansion characteristic coefficient of the lubricating oil is calculated as follows: By pre-obtaining the standard volumetric expansion coefficient of the lubricating oil and the ratio of the static lubricating oil volume inside the reducer to the upper air cavity volume, the thermal expansion characteristic coefficient is jointly calculated. This calculation process organically combines the inherent thermodynamic properties of the fluid itself with the geometric proportions of the specific cavity space of the reducer, accurately quantifying the degree of compression of the internal sealed air chamber space by the heated volume expansion of the lubricating oil under specific operating conditions, and establishing an objective physical benchmark for accurately assessing the nonlinear influence of temperature fluctuations on air pressure. By utilizing the relative change amplitude of lubricating oil temperature at the initial moment and in real time, and combining it with the thermal expansion characteristic coefficient for exponential calculation, a two-phase distortion correction factor is obtained. This processing dynamically quantifies the nonlinear thermal expansion effect of the lubricating oil inside the reducer due to the operating temperature rise, accurately characterizing the theoretical change law of air pressure under complex thermodynamic conditions, thus effectively eliminating the interference of natural pressure inflation caused by system temperature rise in air pressure characteristic analysis.

[0113] The calculation of the time change rate of internal air pressure and the time change rate of lubricating oil temperature at adjacent sampling times includes:

[0114] The rate of change of internal air pressure over time is:

[0115] Calculate the rate of change of lubricating oil temperature over time:

[0116] in: No. The rate of change of internal air pressure over time at each time point; No. The rate of change of lubricating oil temperature over time at each time point; , : No. , No. Internal air pressure measured by sensors at various time points; , : No. , No. The lubricating oil temperature measured by sensors at various time points; Time sampling interval. By performing first-order backward difference calculations on the internal air pressure and lubricating oil temperature at adjacent sampling times, the time change rate of air pressure and the time change rate of temperature are obtained. This processing method, which transforms absolute static measurement into relative dynamic rate, can keenly capture the evolution trend of physical signals within an extremely short time window, effectively reduce the measurement error caused by the static baseline drift of the sensor, highlight the dynamic fluctuation characteristics of the signal, and provide a direct rate characterization for deconstructing the transient thermo-pressure coupling changes in the sealed cavity.

[0117] The calculation of the vapor-liquid coupled pressure fluctuation rate, combining the initial internal air pressure, initial lubricating oil temperature, the two-phase distortion correction factor, and the aforementioned two rates of change, includes:

[0118] To eliminate the interference of gear operation heat generation on air pressure judgment, the vapor-liquid coupling air pressure fluctuation rate is calculated as follows:

[0119] in: No. Vapor-liquid coupling pressure fluctuation rate at each time point; No. The rate of change of internal air pressure over time at each time point; No. Two-phase distortion correction factor at each time point; The internal air pressure measured by the sensor at the first time point; The lubricating oil temperature measured by the sensor at the first time point; No. The lubricating oil temperature change rate over time at each time point. By combining the initial internal pressure-temperature ratio, the two-phase distortion correction factor, and the temperature change rate, the theoretical influence of thermal expansion is accurately subtracted from the overall pressure change rate over time, and the pure vapor-liquid coupling pressure fluctuation rate is calculated. This processing completely eliminates the serious physical interference of gear operation heat generation on the judgment of air pressure in the sealed air chamber. It can restore the real air pressure drop caused only by the loss of gas mass at the sealing lip from the complex temperature and pressure aliasing signal, which greatly improves the objective accuracy of air pressure monitoring in judging micro-leakage.

[0120] The process of calculating the absolute value of the time-varying rate of change of the squared difference in the amplitude of the acoustic vibration signal between adjacent sampling times to obtain the aeroacoustic energy gradient includes:

[0121] Extract the transient change characteristics of the acoustic signal at the seal, and calculate the aeroacoustic energy gradient as follows:

[0122] in: No. Aeroacoustic energy gradient at each time point; , : No. , No. Acoustic vibration signals obtained by sensors at various time points; Time sampling interval; Absolute value calculation symbol. By calculating the squared difference of the absolute amplitude of the acoustic vibration signal at adjacent sampling times and dividing it by the time sampling interval, the aeroacoustic energy gradient is obtained. This calculation step converts the ordinary acoustic amplitude signal into the mutation rate of the transient energy pulse, which can significantly amplify the aeroacoustic high-frequency energy step characteristics generated by the instantaneous micro-leakage of high-pressure gas, effectively overcome the masking effect of smooth mechanical background noise on small abnormal signals, and realize high-sensitivity extraction of the transient characteristics of weak impact of high-frequency airflow.

[0123] The step of calculating the viscosity-temperature acoustic distortion compensation factor based on the initial and real-time lubricating oil temperatures, combined with a preset viscosity-temperature acoustic damping coefficient, includes:

[0124] Calculate the viscosity-temperature acoustic distortion compensation factor as a physical reference for correcting acoustic characteristics:

[0125] in: No. Viscosity-temperature acoustic distortion compensation factor at each time point; The lubricating oil temperature measured by the sensor at the first time point; No. The lubricating oil temperature measured by sensors at various time points; The viscosity-temperature acoustic damping coefficient is determined by consulting the lubricating oil's technical specifications before each monitoring cycle. and Based on the standard kinematic viscosity value, the viscosity-temperature characteristic slope of the lubricating oil was calculated using the Wolter viscosity-temperature formula. Meanwhile, the acoustic transmission loss coefficient of the reducer housing material was checked. Calculate the viscosity-temperature acoustic damping coefficient. By obtaining the standard kinematic viscosity values ​​of lubricating oil at different standard temperatures to calculate the viscosity-temperature characteristic slope, and combining it with the acoustic transmission loss coefficient of the reducer housing material, a viscosity-temperature acoustic distortion compensation factor for the variation of medium damping was constructed. This design fully considers the viscosity decrease phenomenon of lubricating oil caused by dynamic temperature rise, accurately quantifies the damping attenuation law of sound wave transmission medium caused by temperature change, and effectively eliminates the false amplification or excessive attenuation interference of medium viscosity attenuation on the transmission intensity of high-frequency micro-leakage acoustic signals under high temperature conditions.

[0126] The sign of the vapor-liquid coupling pressure fluctuation rate is determined by multiplying its absolute value by the aeroacoustic energy gradient and the viscosity-temperature acoustic distortion compensation factor when it is negative, to obtain the transdomain leakage mapping index, including:

[0127] After filtering out mechanical noise interference and compensating for temperature drift in the acoustic propagation medium, the cross-domain leakage mapping index is calculated as follows:

[0128] in: No. Cross-domain leakage mapping index at each time point; No. Vapor-liquid coupling pressure fluctuation rate at each time point; Sign function, when The internal value is set to 1 when it is greater than 0, 0 when it is equal to 0, and -1 when it is less than 0. No. Aeroacoustic energy gradient at each time point; No. The viscosity-temperature acoustic distortion compensation factor at each time point. By determining the sign direction of the vapor-liquid coupling pressure fluctuation rate, a nonlinear filtering mapping mechanism is constructed using a sign function. Only when the pressure shows a true negative fluctuation is the absolute value of the pressure fluctuation multiplied and fused with the acoustic energy gradient and the distortion compensation factor. This step utilizes the unique physical synchronization of the instantaneous drop in internal pressure and the sudden increase in external acoustic energy that inevitably occurs when a micro-leak occurs, realizing joint signal verification across physical domains and completely filtering out internal mechanical structure noise interference such as gear meshing that is asynchronous with the leak.

[0129] The cross-domain leakage mapping index at each sampling time within the monitoring period is logarithmically weighted and then globally averaged and accumulated to output the severity of reducer seal degradation, including:

[0130] Based on all valid micro-leakage events monitored during the monitoring period, the severity of reducer seal degradation was calculated as follows:

[0131] in: Severity of seal degradation in the output reducer; The total number of time points sampled during the monitoring period; No. The cross-domain leakage mapping index at each time point. By logarithmically weighting the cross-domain leakage mapping index extracted at each effective sampling time within the monitoring period and performing global average accumulation calculation, the overall sealing degradation severity of the system is finally output. Nonlinear logarithmic weighting effectively smooths the discrete and extreme individual leakage pulse spikes and reasonably amplifies the cumulative effect of continuous small degradation, while global average accumulation integrates the characteristics of all micro-leakage events within the entire period, so that the final output degradation severity index not only has high statistical stability, but also reflects the macroscopic physical degradation state of the equipment intuitively and quantitatively.

[0132] 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, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0133] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for evaluating the degradation of reducer sealing performance based on intelligent sensors, characterized in that, include: During the monitoring period, the internal air pressure, lubricating oil temperature and external acoustic vibration signals of the reducer are collected synchronously at a set frequency. Based on the initial and real-time lubricating oil temperatures, combined with the preset thermal expansion characteristic coefficient, the two-phase distortion correction factor is calculated. Calculate the time rate of change of internal air pressure and the time rate of change of lubricating oil temperature at adjacent sampling times; The vapor-liquid coupling pressure fluctuation rate is calculated by combining the initial internal air pressure, the initial lubricating oil temperature, the two-phase distortion correction factor, and the above two change rates. The aeroacoustic energy gradient is obtained by calculating the absolute value of the time-varying rate of change of the squared difference of the acoustic vibration signal amplitude between adjacent sampling times. Based on the initial and real-time lubricating oil temperatures, and combined with the preset viscosity-temperature acoustic damping coefficient, the viscosity-temperature acoustic distortion compensation factor is calculated. The sign of the vapor-liquid coupling pressure fluctuation rate is determined. When it is a negative fluctuation, its absolute value is multiplied by the aeroacoustic energy gradient and the viscosity-temperature acoustic distortion compensation factor to obtain the cross-domain leakage mapping index. Logarithmically weighted the cross-domain leakage mapping index at each sampling time within the monitoring period, and then globally averaged and accumulated the results to output the severity of reducer seal degradation.

2. The method for evaluating the degradation of reducer sealing performance based on intelligent sensors according to claim 1, characterized in that, The monitoring cycle includes synchronously collecting the internal air pressure, lubricating oil temperature, and external acoustic vibration signals of the reducer at a set frequency, including: A dynamic air pressure sensor is installed in the cavity inside the reducer to measure the internal air pressure; a temperature sensor is installed in the lubricating oil sump inside the reducer to measure the lubricating oil temperature; a high-frequency acoustic emission sensor is installed on the outer housing of the sealing lip of the reducer output shaft to measure the acoustic vibration signal at the seal. Set the total number of time node samples and the preset acquisition frequency; calculate and determine the time sampling interval based on the reciprocal relationship of the preset acquisition frequency, and establish the corresponding discrete time node sequence; Based on the set discrete time node sequence and time sampling interval, the internal air pressure sequence, lubricating oil temperature sequence and acoustic vibration signal sequence are synchronously acquired through the sensor at each time node.

3. The method for evaluating the degradation of reducer sealing performance based on intelligent sensors according to claim 1, characterized in that, The calculation of the two-phase distortion correction factor based on the initial and real-time lubricating oil temperatures, combined with a preset thermal expansion characteristic coefficient, includes: Before the monitoring cycle begins, the standard volume expansion coefficient of the lubricating oil inside the reducer and the volume ratio of the lubricating oil volume to the upper air cavity volume when the reducer is in a static state are obtained; the standard volume expansion coefficient is multiplied by the volume ratio to obtain the preset thermal expansion characteristic coefficient. Calculate the difference between the real-time collected lubricating oil temperature and the lubricating oil temperature measured at the initial time; divide the difference by the lubricating oil temperature measured at the initial time to obtain the relative temperature change. The relative temperature change is multiplied by the preset thermal expansion characteristic coefficient, and the product is used as the exponent of the natural constant for exponential operation to calculate the two-phase distortion correction factor.

4. The method for evaluating the degradation of reducer sealing performance based on intelligent sensors according to claim 1, characterized in that, The calculation of the time change rate of internal air pressure and the time change rate of lubricating oil temperature at adjacent sampling times includes: Calculate the difference in internal air pressure between the current time node and the previous time node, and divide the difference in internal air pressure by the time sampling interval to obtain the rate of change of internal air pressure over time. Calculate the difference in lubricating oil temperature between the current time node and the previous time node, and divide the difference in lubricating oil temperature by the time sampling interval to obtain the lubricating oil temperature change rate over time.

5. The method for evaluating the degradation of reducer sealing performance based on intelligent sensors according to claim 1, characterized in that, The calculation of the vapor-liquid coupled pressure fluctuation rate, combining the initial internal air pressure, initial lubricating oil temperature, the two-phase distortion correction factor, and the aforementioned two rates of change, includes: The initial state ratio is obtained by dividing the internal air pressure measured at the initial moment by the lubricating oil temperature measured at the initial moment. Multiply the initial state ratio, the two-phase distortion correction factor, and the lubricating oil temperature-time change rate together to obtain the theoretical value of thermal expansion pressure change. The vapor-liquid coupling pressure fluctuation rate is calculated by subtracting the theoretical value of the thermal expansion pressure change from the internal pressure time change rate.

6. The method for evaluating the degradation of reducer sealing performance based on intelligent sensors according to claim 1, characterized in that, The process of calculating the absolute value of the time-varying rate of change of the squared difference in the amplitude of the acoustic vibration signal between adjacent sampling times to obtain the aeroacoustic energy gradient includes: Calculate the square of the acoustic vibration signal amplitude at the current time point and the previous time point respectively; Calculate the difference between the squared value at the current time point and the squared value at the previous time point; Divide the difference by the time sampling interval to obtain the acoustic energy time-varying rate; The aeroacoustic energy gradient is calculated by taking the absolute value of the time-varying rate of acoustic energy.

7. The method for evaluating the degradation of reducer sealing performance based on intelligent sensors according to claim 1, characterized in that, The step of calculating the viscosity-temperature acoustic distortion compensation factor based on the initial and real-time lubricating oil temperatures, combined with a preset viscosity-temperature acoustic damping coefficient, includes: Before the monitoring cycle begins, the standard kinematic viscosity values ​​of the lubricating oil at different standard temperatures are obtained, and the viscosity-temperature characteristic slope of the lubricating oil is calculated according to the viscosity-temperature characteristic formula; the acoustic transmission loss coefficient of the reducer housing material is obtained; the absolute value of the viscosity-temperature characteristic slope is multiplied by the acoustic transmission loss coefficient to obtain the preset viscosity-temperature acoustic damping coefficient. Calculate the difference between the lubricating oil temperature measured at the initial moment and the lubricating oil temperature collected in real time, and divide the difference by the lubricating oil temperature collected in real time to obtain the reverse relative temperature deviation. The viscosity-temperature acoustic distortion compensation factor is calculated by multiplying the reverse relative temperature deviation by the preset viscosity-temperature acoustic damping coefficient and using the product as the exponent of the natural constant.

8. The method for evaluating the degradation of reducer sealing performance based on intelligent sensors according to claim 1, characterized in that, The sign of the vapor-liquid coupling pressure fluctuation rate is determined by multiplying its absolute value by the aeroacoustic energy gradient and the viscosity-temperature acoustic distortion compensation factor when it is negative, to obtain the transdomain leakage mapping index, including: The sign direction value of the vapor-liquid coupled pressure fluctuation rate is extracted using a sign function; Subtract the sign direction value from one and divide the result by two to construct a retention coefficient, which is such that when the vapor-liquid coupling pressure fluctuation rate is negative, it takes a value of one, and otherwise takes a value of zero. The cross-domain leakage mapping index is calculated by sequentially multiplying the retention coefficient, the absolute value of the vapor-liquid coupling pressure fluctuation rate, the aeroacoustic energy gradient, and the viscosity-temperature acoustic distortion compensation factor.

9. The method for evaluating the degradation of reducer sealing performance based on intelligent sensors according to claim 1, characterized in that, The cross-domain leakage mapping index at each sampling time within the monitoring period is logarithmically weighted and then globally averaged and accumulated to output the severity of reducer seal degradation, including: The cross-domain leakage mapping exponent at each valid sampling time is added to the natural constant to obtain the sum value; Calculate the natural logarithm of the sum to obtain the logarithmic weighting coefficients; Multiply the cross-domain leakage mapping index at the corresponding sampling time with the logarithmic weighting coefficient to obtain the weighted mapping feature value at each sampling time; The weighted mapping feature values ​​of all sampling times excluding the initial time within the monitoring period are summed to obtain the global accumulated value; The global cumulative value is averaged by the total number of time node samples, and the calculated severity of the reducer seal degradation is output.

Images