Stepping timing truncation type centrifugal pump accelerated life test method based on three stresses

By using the three-stress step-timed tail-cut method in the centrifugal pump life test, combining the generalized Ailin model and virtual augmented samples, the problems of low life test efficiency and insufficient multi-stress handling capabilities in the existing technology are solved, and the problem of low life test efficiency and insufficient multi-stress handling capabilities are achieved. More accurate life assessment and test efficiency improvement.

CN118998074AInactive Publication Date: 2024-11-22AEROSPACE POWER RES INST (SUZHOU) CO LTD

Patent Information

Application Number
CN202411480800.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-23
Publication Date
2024-11-22
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing centrifugal pump life test methods are inefficient, making it difficult to accurately evaluate the acceleration life of long-life pumps. The existing models lack the ability to handle multi-stress conditions, and it is impossible to effectively simulate the failure process in complex environments.

Method used

The step-timed tail-cut acceleration life test method based on three stresses is adopted. By selecting the outlet pressure and impeller vibration signals as degradation signals, combining speed, temperature and flow as acceleration stress factors, a generalized Ailin model is constructed to generate an acceleration load spectrum. A step-timed cut-cut acceleration life test was implemented, and the life model was improved through virtual augmented samples and Weibull distribution fitting.

Benefits of technology

Improve the accuracy and test efficiency of life evaluation, enhance the applicability and reliability of test results, shorten the test time without changing the failure mechanism, generate a more complete life model, and accurately reflect the pumps under different stress conditions. Lifespan distribution characteristics.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118998074A_ABST
    Figure CN118998074A_ABST
Patent Text Reader

Abstract

The invention provides a stepping timing truncation type centrifugal pump accelerated life test method based on three stresses. According to the method, firstly, outlet pressure of a centrifugal pump is selected as a main degradation signal, and an impeller vibration signal is selected as an auxiliary degradation signal; according to abrasion, aging and fatigue failure mechanisms of the centrifugal pump, the rotating speed, the temperature and the flow are determined as accelerated stress factors, and an accelerated life model is built in combination with a generalized Ailin model. And generating an accelerated load spectrum, specifying the application sequence, level and duration of stress factors, and implementing a stepping timing truncation type accelerated life test. Pressure and vibration signal data recorded in the test process are subjected to curve fitting to generate a degradation track, and a failure threshold value is set. And fitting the relationship between the degradation data and the failure moment by adopting a least square method, estimating a pseudo-life value of the centrifugal pump under an accelerated stress condition, expanding pseudo-life data through a virtual augmented sample, fitting in combination with Weibull distribution, and finally estimating the life of the centrifugal pump under a conventional stress condition.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of centrifugal pumps, and in particular to a three-stress based accelerated life test method for a step-by-step timing truncated centrifugal pump. Background Art

[0002] In the life test of centrifugal pumps, since centrifugal pumps usually have a long life cycle, the existing life test efficiency may be difficult to meet the expected requirements. In order to improve the test efficiency, a larger acceleration ratio is often required, but this may lead to changes in the failure mechanism. In addition, the amount of data collected during the test is large, and the data processing is complex and time-consuming. Especially when a single degradation factor is used, the accuracy of the life-acceleration model is low. The existing life test methods still have many shortcomings.

[0003] For example, the Chinese invention patent with publication number CN115853767A discloses an accelerated life test system for a linear conjugate internal gear pump, but this method requires a full-cycle life test of the gear pump, which consumes a lot of manpower and material resources and is not suitable for life tests of long-life pumps. In the patent with publication number CN115758754A, an accelerated life test method for an axial piston pump based on constant pressure is proposed. Although the test scheme is optimized through random simulation, this method is only applicable to single stress conditions and is difficult to cope with life tests with variable stress or multiple stress combinations. The patent with publication number CN108470082B adopts an exponential inverse power law accelerated life model based on failure physics, but this model is mainly applicable to accelerated stresses such as temperature and electrical stress, and is insufficient in its ability to handle accelerated stresses such as load and speed.

[0004] Therefore, it is urgent to develop new accelerated life test methods for centrifugal pumps. Summary of the invention

[0005] The present application provides a three-stress based step-by-step timing truncated centrifugal pump accelerated life test method to improve the accuracy of life assessment.

[0006] The present application provides a three-stress based step-by-step timing truncated centrifugal pump accelerated life test method, comprising: Based on the operating performance of the centrifugal pump, the outlet pressure is selected as the main degradation signal and the impeller vibration signal is selected as the auxiliary degradation signal, wherein the main degradation signal and the auxiliary degradation signal are used to evaluate the performance degradation of the centrifugal pump; According to the wear, aging and fatigue failure mechanism of the centrifugal pump, the rotation speed, temperature and flow rate are determined as accelerated stress factors, wherein the accelerated stress factors are used as stress application conditions in the life test; Based on the generalized Eyring model, combined with the determined accelerated stress factor and the main degradation signal and the auxiliary degradation signal, an accelerated life model of the centrifugal pump is constructed; wherein the accelerated life model is used to describe the influence of stress on the life of the centrifugal pump; Based on the constructed accelerated life model, an accelerated load spectrum is generated, wherein the accelerated load spectrum specifies the application order, level and duration of different accelerated stress factors for the specific execution of the accelerated life test; According to the generated accelerated load spectrum, a step-by-step timing truncation accelerated life test is carried out; the outlet pressure and impeller vibration signal data of the centrifugal pump at different stress levels are recorded; Based on the recorded outlet pressure and impeller vibration signal data, the degradation trajectory of the centrifugal pump is calculated using a curve fitting method to generate a degradation curve reflecting the attenuation of pump performance; Based on the generated degradation curve, the failure threshold of the centrifugal pump is set, and the moment when the degradation curve reaches the failure threshold is determined; the mathematical relationship between the degradation data and the failure moment is fitted by the least squares method, and based on the fitted mathematical relationship, the pseudo life value of the centrifugal pump under accelerated stress conditions is estimated; Based on the obtained pseudo-life values, the pseudo-life data are expanded using virtual augmented samples; the expanded pseudo-life data are fitted with Weibull distribution to improve the accelerated life model; based on the improved accelerated life model, the life of the centrifugal pump under conventional stress conditions is estimated.

[0007] Furthermore, the outlet pressure is used as the main degradation signal and is collected in real time by a pressure sensor installed at the outlet of the centrifugal pump. The impeller vibration signal is used as the auxiliary degradation signal and is collected in real time by an acceleration sensor installed on the outside of the pump casing to monitor the mechanical vibration during the operation of the centrifugal pump.

[0008] Furthermore, the temperature in the accelerated stress factor is monitored in real time by multi-point temperature sensors installed in the pump body, and the temperature variation range is limited to between 10°C and 120°C to prevent premature failure of the pump seal due to excessive temperature.

[0009] Furthermore, the execution time of the stepped timing truncation accelerated life test is divided into multiple stages, the stress application time period of each stage increases, and the degradation state of the centrifugal pump is recorded at the end of each stage, and staged life data is generated.

[0010] Furthermore, the accelerated life model is constructed based on the generalized Eyring model and is implemented using the following formula 1: in, Indicates the centrifugal pump at time Accelerated life under is an empirical constant; are the stress values ​​of rotation speed, temperature and flow rate respectively; is the sensitivity coefficient, which reflects the influence of stress factors on life; The main degradation signal over time The change function of To assist the degradation signal over time The change function of and is a weighted index related to the degradation signal, which controls the weight of the impact of pressure and vibration on life respectively; is the nonlinear adjustment parameter in the accelerated life model; Among them, the function Use the following formula 2: in, It is the outlet pressure of the centrifugal pump when it is initially running, indicating the time Initial pressure value at ; is the running time of the experiment, which represents the time variable from the start of the experiment to the current moment; Indicates the design life of the centrifugal pump, indicating the expected operating time of the equipment under normal conditions; is the pressure decay rate parameter, which is used to control the speed and nonlinearity of pressure decay; is the amplitude of noise oscillation; is the frequency of noise oscillation; the initial magnitude of the correction term; is the decay constant.

[0011] function Use the following formula 3: in, It is the impeller vibration signal value of the centrifugal pump during initial operation, indicating the time The initial vibration amplitude at ; is the running time of the experiment, which represents the time variable from the start of the experiment to the current moment; It is the design life time of the centrifugal pump, indicating the expected operating time of the equipment under normal conditions; Add rate parameters to the vibration signal to control the nonlinear degree of vibration increase over time; the magnitude of the disturbance term; is the frequency of the disturbance term; is the increasing constant.

[0012] Furthermore, the degradation trajectory of the centrifugal pump is calculated by a curve fitting method based on the recorded outlet pressure and impeller vibration signal data to generate a degradation curve reflecting the attenuation of pump performance, including: Based on the outlet pressure and impeller vibration signal data collected during the operation of the centrifugal pump, the data at each time point are preprocessed, including noise removal, filtering and smoothing, to reduce the impact of measurement errors on curve fitting; The pre-processed outlet pressure and impeller vibration signals are used as input data, and a nonlinear curve fitting method is selected. The fitting method dynamically adjusts the fitting parameters according to the decay trend of the pump performance over time to ensure that the fitting results can accurately reflect the performance degradation process of the pump; In the curve fitting process, the attenuation trend of the outlet pressure and the growth trend of the impeller vibration signal are fitted separately, and then the two fitting results are combined to generate the overall degradation curve of the centrifugal pump, so as to accurately reflect the comprehensive performance degradation of the pump; The degradation curve generated is used to describe the progressive degradation of the centrifugal pump performance over time, where the drop in outlet pressure and the increase in impeller vibration are considered as the main indicators that the pump is gradually approaching failure.

[0013] Furthermore, the setting of the failure threshold of the centrifugal pump based on the generated degradation curve and determining the time when the degradation curve reaches the failure threshold includes: Based on the historical operation data of the centrifugal pump and its actual performance degradation characteristics, setting a failure threshold of the centrifugal pump, wherein the failure threshold is set by defining that the outlet pressure drops to a predetermined percentage of the initial value or the impeller vibration signal exceeds a specific limit; By analyzing the generated degradation curve point by point, the moment when the degradation signal reaches the set failure threshold is determined, and this moment is recorded as the failure time of the centrifugal pump; wherein the failure time is used to characterize the performance failure point of the centrifugal pump under accelerated stress conditions, reflecting the key degradation state of the pump.

[0014] Furthermore, the method of fitting the mathematical relationship between the degradation data and the failure time by the least square method and estimating the pseudo life value of the centrifugal pump under accelerated stress conditions based on the fitted mathematical relationship includes: The recorded degradation curve and failure time data are used as input, and the relationship between the degradation signal and the failure time is fitted using the least square method. The fitting results are gradually optimized to ensure that the degradation curve can accurately describe the degradation process of the pump. Based on the fitted mathematical relationship and combined with the influence weights of various stress factors under accelerated stress conditions, the pseudo life value of the centrifugal pump under accelerated stress conditions is estimated; wherein the pseudo life value is used to predict the expected failure time of the centrifugal pump in the accelerated test.

[0015] Furthermore, the method of expanding the pseudo-lifetime data using virtual augmented samples based on the obtained pseudo-lifetime value includes: Based on the pseudo life value obtained in the accelerated life test, it serves as the basis of life data and reflects the degradation characteristics of the centrifugal pump under accelerated stress conditions; Based on the pseudo life value, the historical life data or other relevant data are used to generate virtual augmented samples through data simulation technology; The data augmentation method of Monte Carlo simulation is used to randomly generate multiple pseudo-life data points to ensure that the augmented samples have a wide coverage under multiple stress conditions; By introducing randomness, error distribution and degradation model uncertainty, virtual augmented samples similar to actual conditions are generated to ensure that the generated virtual samples can reflect the possible life span change trend; The generated virtual augmented samples are combined with the initially obtained pseudo-lifetime values ​​to expand the pseudo-lifetime dataset and form a more complete pseudo-lifetime data distribution.

[0016] Furthermore, the Weibull distribution fitting adopts the following formula 4: in, For time The probability of failure when is a scale parameter, reflecting the characteristic time of life; is the shape parameter; is the amplitude of the sinusoidal fluctuation; is the frequency of the sinusoidal oscillation; is the amplitude of the cosine fluctuation; is the frequency of the cosine wave; are the rotation speed, temperature and flow rate in the stress factor respectively; is the sensitivity coefficient, reflecting the influence of different stress factors on life; is the stress adjustment coefficient, which is used to adjust the overall impact of the stress factor.

[0017] The beneficial effects of the technical solution provided by this application include: (1) By selecting the outlet pressure as the main degradation signal and the impeller vibration signal as the auxiliary degradation signal, this method can comprehensively reflect the performance degradation of the centrifugal pump. Compared with the evaluation method of a single degradation factor, this multi-signal combination method can more accurately capture the performance attenuation process of the centrifugal pump, thereby improving the accuracy of life prediction. (2) By determining the speed, temperature and flow rate as accelerated stress factors and constructing an accelerated life model based on the generalized Eyring model, this method can more realistically simulate the failure process of the centrifugal pump in a complex environment. Compared with the single stress accelerated test, the introduction of three stress factors makes the accelerated test more in line with the actual working conditions of the centrifugal pump, thereby enhancing the applicability and reliability of the test results. (3) By generating an accelerated load spectrum and implementing the accelerated life test in a step-by-step timing truncation method, the test time can be effectively shortened without changing the failure mechanism. This design method can reduce the test cycle of long-life pumps, improve test efficiency, and reduce time and resource consumption. (4) By using virtual augmented samples to expand pseudo-life data and combining them with Weibull distribution for fitting, this method can generate more data points based on limited test samples, enhancing the universality and prediction accuracy of the life model. At the same time, combined with Weibull distribution fitting, it can accurately reflect the life distribution characteristics of centrifugal pumps under different stress conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a flow chart of a method for accelerating life test of a centrifugal pump with step-by-step timing and truncated tail type based on three stresses provided in the first embodiment of the present application. DETAILED DESCRIPTION

[0019] Many specific details are described in the following description to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present application, so the present application is not limited by the specific implementation disclosed below.

[0020] The first embodiment of the present application provides a method for accelerating the life test of a centrifugal pump based on three stresses. Figure 1 , which is a schematic diagram of the first embodiment of the present application. Figure 1 A first embodiment of the present application provides a three-stress based step-by-step timing truncated centrifugal pump accelerated life test method which is described in detail.

[0021] Step S101: Based on the operating performance of the centrifugal pump, the outlet pressure is selected as the main degradation signal and the impeller vibration signal is selected as the auxiliary degradation signal, wherein the main degradation signal and the auxiliary degradation signal are used to evaluate the performance degradation of the centrifugal pump.

[0022] Furthermore, the outlet pressure is used as the main degradation signal and is collected in real time by a pressure sensor installed at the outlet of the centrifugal pump. The impeller vibration signal is used as the auxiliary degradation signal and is collected in real time by an acceleration sensor installed on the outside of the pump casing to monitor the mechanical vibration during the operation of the centrifugal pump.

[0023] In step S101, firstly, based on the operating performance of the centrifugal pump, a degradation signal suitable for life test is determined. The outlet pressure is mainly selected as the main degradation signal, mainly because the outlet pressure of the centrifugal pump directly reflects the liquid delivery capacity and overall operating status of the pump. As the pump wears and ages, the outlet pressure will gradually decrease, so it can more directly show the degradation process of the pump. At the same time, the vibration signal of the impeller is used as an auxiliary degradation signal, because the operation of the impeller inside the centrifugal pump will be affected by wear, bearing failure or degradation of other mechanical components, and the vibration signal can reflect these mechanical changes. As the centrifugal pump ages, the vibration signal of the impeller will gradually increase, so the vibration signal is also an important indicator of performance degradation.

[0024] In order to effectively capture these signals, a high-precision pressure sensor is installed at the outlet of the centrifugal pump to collect the outlet pressure data of the pump in real time. At the same time, an acceleration sensor is installed near the pump casing or impeller to capture changes in the impeller vibration signal. The acquisition frequency and sensitivity of the sensor need to be set according to the actual operating conditions of the pump to ensure that tiny pressure changes and vibration signals can be accurately captured during long-term operation. Usually, the sampling frequency of the pressure sensor is set to 1 Hz to 10 Hz to ensure that the pressure change trend can be effectively tracked, while the sampling frequency of the acceleration sensor is set between 1 kHz and 10 kHz to ensure high-precision monitoring of the vibration of high-speed rotating equipment.

[0025] Based on the collected outlet pressure and impeller vibration signals, they are used as the main degradation indicators of the life test for subsequent performance evaluation and life prediction. Changes in the outlet pressure will reflect the degradation of the hydraulic performance of the pump, while changes in the impeller vibration signal will reflect the degree of wear of the mechanical components. In order to ensure that these signals can accurately evaluate the performance degradation process of the centrifugal pump, it is necessary to perform necessary preprocessing on the signal data, including denoising, filtering and other operations to ensure the accuracy and validity of the data. Ultimately, selecting the outlet pressure and impeller vibration signals as degradation signals can provide a more comprehensive and accurate basis for evaluating the performance degradation of the pump, which is helpful for the construction of the life model and its accuracy in subsequent tests.

[0026] Step S102: According to the wear, aging and fatigue failure mechanism of the centrifugal pump, the rotation speed, temperature and flow rate are determined as accelerated stress factors, wherein the accelerated stress factors are used as stress application conditions in the life test.

[0027] In step S102, the rotation speed, temperature and flow rate are determined as accelerated stress factors according to the wear, aging and fatigue failure mechanisms of the centrifugal pump. The selection of these stress factors needs to comprehensively consider the working environment and failure modes of the centrifugal pump under actual working conditions. Wear, aging and fatigue are common degradation modes during the operation of centrifugal pumps, which affect different parts of the pump respectively. Therefore, the selection of accelerated stress factors is intended to accelerate these degradation processes to shorten the life test time and maintain the corresponding relationship with the actual failure mechanism.

[0028] First of all, the speed is the core factor that directly affects the operation of the centrifugal pump. The centrifugal pump drives the liquid from the inlet to the outlet through the high-speed rotating impeller. The higher the speed, the faster the pump runs and the greater the flow rate of the liquid. However, too high a speed will increase the wear of the impeller and bearings and shorten the life of these mechanical parts. Therefore, in the accelerated life test, the speed is used as one of the accelerated stress factors. In order to ensure that the speed stress is applied appropriately, the acceleration speed range can be set according to the rated speed of the pump. Usually, a speed level that is a certain percentage higher than the rated speed is selected, such as increasing the speed to 120%-150% of the rated value, thereby accelerating the wear process of the mechanical parts.

[0029] Secondly, temperature, as another accelerating stress factor, has a significant impact on the material aging of centrifugal pumps. The seals, bearings and other metal parts of centrifugal pumps will age faster in high temperature environments, and the physical properties of the materials will change, resulting in a decrease in the performance of the pump. In life tests, the temperature application needs to be set according to the operating environment and material properties of the pump. Generally, the test temperature can be raised to a level higher than the normal operating temperature, such as setting the temperature to 1.5 to 2 times the normal operating temperature of the pump, thereby accelerating the aging of seals, the impact of thermal expansion and contraction on bearings, and thermal stress changes in the overall structure. The test equipment should be equipped with a high-precision temperature control system to ensure the stability and consistency of the temperature stress application process.

[0030] Flow rate is a key parameter for centrifugal pumps to transport liquids, which directly determines the output performance of the pump. In accelerated life tests, flow rate is used as a stress factor to simulate the hydraulic load of the pump. When the flow rate increases, the pressure distribution inside the pump changes, and components such as the impeller, pump casing and seals will be subjected to a greater hydraulic load, resulting in faster degradation. In order to accelerate this degradation process, a liquid flow rate higher than the rated flow rate can be applied in the life test to shorten the life cycle of the pump by increasing the hydraulic load of the pump. Typically, the accelerated flow rate can be set to 1.2 to 1.5 times the rated flow rate to ensure that the hydraulic load of the pump increases within a reasonable range without causing instantaneous overload of the pump.

[0031] In actual operation, these three stress factors: speed, temperature and flow rate, should be applied simultaneously to simulate the degradation behavior of centrifugal pumps under complex working conditions. When applying these stresses, the application level of the acceleration factor must be adjusted according to the design parameters and failure mode of the pump to ensure that the stress application in the accelerated test is similar to the actual operating conditions. In addition, the application of stress must be carried out in a step-by-step manner to gradually increase the stress level while ensuring that the physical limit of the pump is not exceeded. In this way, the time of the life test can be effectively shortened, and reliable degradation data can be obtained for subsequent model construction and life prediction.

[0032] Furthermore, the temperature in the accelerated stress factor is monitored in real time by multi-point temperature sensors installed in the pump body, and the temperature variation range is limited to between 10°C and 120°C to prevent premature failure of the pump seal due to excessive temperature.

[0033] In this implementation step, the temperature is used as one of the accelerating stress factors and is monitored in real time by installing multiple temperature sensors in the centrifugal pump body. These multiple temperature sensors are installed at different key locations of the pump body, usually near seals, bearings and other components that are susceptible to temperature. These components are highly sensitive to temperature, and temperature changes can significantly affect their performance and life. Through multi-point monitoring, temperature data at different locations can be obtained to ensure that the impact of temperature changes on the overall performance of the pump can be fully evaluated.

[0034] Multi-point temperature sensors are implemented through high-precision thermocouples or other applicable temperature sensing devices to ensure that temperature changes can be continuously monitored during pump operation. These sensors are connected to the data acquisition system to feedback temperature data in real time and provide accurate temperature distribution information. At the same time, in order to ensure the safe operation of the system, the temperature range is strictly limited to between 10°C and 120°C. This temperature range is set based on the material properties of the pump body and the tolerance of the seal. Below 10°C may affect the flow of liquid and the mechanical properties of the pump, while above 120°C will cause premature aging or failure of seals and other heat-sensitive components.

[0035] In actual operation, the system processes the temperature signal in real time through the controller. When the temperature is detected to be close to or beyond the specified range, the control system can issue an alarm to remind the operator to adjust the operating parameters, or automatically adjust stress factors such as flow rate and speed to reduce temperature stress. This can effectively prevent the premature failure of components such as seals due to excessively high temperatures, and ensure that the life test of the pump is carried out within a safe temperature range.

[0036] At the same time, multi-point temperature monitoring provides accurate temperature change trend information, which helps to better understand the impact of temperature on the pump and can provide important reference for the optimization of the life model.

[0037] Step S103: Based on the generalized Eyring model, combined with the determined accelerated stress factor and the main degradation signal and the auxiliary degradation signal, an accelerated life model of the centrifugal pump is constructed; wherein the accelerated life model is used to describe the influence of stress on the life of the centrifugal pump.

[0038] In step S103, based on the generalized Eyring model, the accelerated life model of the centrifugal pump is constructed in combination with the previously determined accelerated stress factors and degradation signals. First, the generalized Eyring model is a common model used for accelerated life testing, which can quantify the impact of multiple stress factors on product life. This model assumes that the effect of stress factors will accelerate the degradation of the equipment, thereby shortening its life. Therefore, in order to construct the accelerated life model, it is first necessary to clarify the previously determined stress factors and their impact on the life of the centrifugal pump.

[0039] The construction of the model requires combining the three stress factors of speed, temperature and flow with the main degradation signal (outlet pressure) and the auxiliary degradation signal (impeller vibration). To achieve this goal, the weight of each stress factor and its influence coefficient must be determined first. These weights and coefficients are usually determined by historical experimental data or by the design and operating parameters of the equipment. Each stress factor will have a different degree of influence on the degradation process of the centrifugal pump, so it is necessary to determine their specific contribution to the degradation rate through experiments or calculations.

[0040] Once the effect of stress factors on degradation is determined, we can start to build an accelerated life model. This model is based on the generalized Eyring model, which takes the three stress factors of speed, temperature and flow as inputs, and the outlet pressure and impeller vibration signals as degradation outputs. The core idea of ​​the model is to establish a mathematical relationship between the stress factors and the degradation process of the pump, so that the model can reflect how the degradation rate of the pump accelerates under different stress levels. Specifically, the model describes the cumulative effect of stress factors through nonlinear relationships. Increases in speed, temperature and flow will lead to accelerated wear, aging and fatigue of the pump, which will affect the outlet pressure and impeller vibration signals.

[0041] During the model building process, appropriate parameters need to be set for each stress factor. For example, the parameters of the speed stress factor may include its contribution to the wear of the impeller and bearings, while the parameters of the temperature stress factor may affect the aging rate of seals and materials. The flow stress factor mainly affects the hydraulic load and the stress distribution of the pump body. Through the combination of these parameters, a mathematical model that reflects the relationship between stress and degradation processes can be generated.

[0042] At the same time, the processing of degradation signals is also crucial. The outlet pressure, as the main degradation signal, directly reflects the liquid delivery capacity of the pump, while the impeller vibration signal, as the auxiliary degradation signal, reflects the wear and vibration of the mechanical part. Both need to be accurately fitted in the model to ensure that the accelerated life model can accurately capture the performance degradation trend of the pump.

[0043] When constructing the model, the degradation curve also needs to be gradually optimized. By fitting the experimental data, the various parameters in the model are gradually adjusted to ensure that the model can reflect the actual degradation behavior under different stress conditions. Finally, by adjusting the influence weights of each stress factor of the generalized Eyring model and the processing method of the degradation signal, a mathematical model that can accurately reflect the impact of accelerated stress on the life of the centrifugal pump is constructed.

[0044] The model will be used to generate accelerated load spectra in subsequent steps and applied in accelerated life tests. Therefore, every detail in the construction process must be experimentally verified and optimized to ensure that the model has sufficient accuracy and operability.

[0045] Furthermore, the accelerated life model is constructed based on the generalized Eyring model and is implemented using the following formula 1: in, Indicates the centrifugal pump at time Accelerated life under is an empirical constant; are the stress values ​​of rotation speed, temperature and flow rate respectively; is the sensitivity coefficient, which reflects the influence of stress factors on life; The main degradation signal over time The change function of To assist the degradation signal over time The change function of and is a weighted index related to the degradation signal, which controls the weight of the impact of pressure and vibration on life respectively; is the nonlinear adjustment parameter in the accelerated life model; Among them, the function Use the following formula 2: in, It is the outlet pressure of the centrifugal pump when it is initially running, indicating the time Initial pressure value at ; is the running time of the experiment, which represents the time variable from the start of the experiment to the current moment; Indicates the design life of the centrifugal pump, indicating the expected operating time of the equipment under normal conditions; is the pressure decay rate parameter, which is used to control the speed and nonlinearity of pressure decay; is the amplitude of noise oscillation; is the frequency of noise oscillation; the initial magnitude of the correction term; is the decay constant.

[0046] function Use the following formula 3: in, It is the impeller vibration signal value of the centrifugal pump during initial operation, indicating the time The initial vibration amplitude at ; is the running time of the experiment, which represents the time variable from the start of the experiment to the current moment; It is the design life time of the centrifugal pump, indicating the expected operating time of the equipment under normal conditions; Add rate parameters to the vibration signal to control the nonlinear degree of vibration increase over time; the magnitude of the disturbance term; is the frequency of the disturbance term; is the increasing constant.

[0047] In this embodiment, the accelerated life model is constructed based on the generalized Eyring model, and the life prediction and analysis of the centrifugal pump is realized through formula (1). The formula is described in detail below.

[0048] In formula (1), Indicates the centrifugal pump at time The accelerated life under the condition of accelerated stress is the expected life of the pump. This life is the result of the combined effect of different stress factors and degradation signals. It is an empirical constant obtained through experiments or historical data, and reflects the basic life of the pump under specific working conditions.

[0049] The stress factor part of the formula and Represent the stress values ​​of the speed, temperature and flow of the centrifugal pump respectively. These stress factors are gradually applied during the test process. For each stress factor, there is a corresponding sensitivity coefficient and , these coefficients reflect the degree of influence of stress factors on life. A higher sensitivity coefficient indicates that the stress factor has a greater influence on life, while a lower coefficient indicates a smaller influence. These sensitivity coefficients are usually obtained through regression analysis or experimental data fitting.

[0050] The processing of the degradation signal part depends on the pressure signal and vibration signal , which are calculated using formulas (2) and (3) respectively. Formula (2) represents the outlet pressure over time. Initial outlet pressure In time is recorded as the initial value at , and decays gradually over time. The decay part is Control, parameters Determines the rate and nonlinear characteristics of pressure decay, is the design life time. On this basis, two correction items are added: one is , which models the noise or periodic fluctuations in the pressure signal, where is the amplitude of the fluctuation, is the frequency; the other correction term is , which is used to simulate the additional attenuation effect that gradually weakens over time, where is the decay constant.

[0051] Likewise, the vibration signal Formula (3) describes the change of the impeller vibration signal over time. Initial vibration signal Indicates the vibration value of the pump at the beginning of operation. As time goes by, the vibration signal increases according to Increment, parameter Control the rate and nonlinear characteristics of vibration growth, The same is the design life time. Similar to the pressure signal, the vibration signal also contains disturbance terms , used to simulate periodic fluctuations in vibration, parameters is the fluctuation range, is the disturbance frequency. In addition, the vibration signal also contains a logarithmic term , indicating the slow increase of vibration signal in long-term operation, is the increasing constant of the logarithmic term.

[0052] Weighted Index and They are used to control the weights of the pressure signal and the vibration signal on the life. These indices are used in the model to balance the influence of the two signals to ensure that the life model can fully reflect the overall degradation state of the pump. The specific weight values ​​can be obtained by fitting or analyzing experimental data.

[0053] The integral part of the formula Used for time The degradation process within is cumulatively calculated, and this integral describes the cumulative effect of the pressure and vibration signals. It is a nonlinear adjustment parameter used to control the nonlinear characteristics in the accelerated life model to make the model more consistent with the actual degradation behavior.

[0054] The specific values ​​of each parameter can be gradually determined through experimental data and historical records, and then the accelerated life of the centrifugal pump can be calculated using the above formula. By reasonably adjusting and optimizing the various parameters in the model, the calculation results can be ensured to have high accuracy and operability, thereby accurately predicting the life of the centrifugal pump.

[0055] Step S104: generating an accelerated load spectrum based on the constructed accelerated life model, wherein the accelerated load spectrum specifies the application sequence, level and duration of different accelerated stress factors for specific execution of the accelerated life test.

[0056] In step S104, based on the constructed accelerated life model, generating an accelerated load spectrum is one of the key steps in the accelerated life test of the centrifugal pump. The purpose of the accelerated load spectrum is to simulate the operation process of the centrifugal pump under different working conditions by reasonably arranging the application sequence, level and duration of the accelerated stress factor, thereby accelerating the degradation of the pump so as to predict its life in a shorter time.

[0057] The first step in generating an accelerated load spectrum is to analyze the effects of various stress factors in the accelerated life model, including speed, temperature, and flow. Since each stress factor has a different effect on the life of the pump, the order in which these stress factors are applied needs to be set in the load spectrum. Usually, the priority of stress factors can be determined based on the actual working environment and degradation mechanism of the pump. For example, in the actual application of some centrifugal pumps, the change in speed may be the most important influencing factor, so the change in speed can be given priority, supplemented by the adjustment of temperature and flow.

[0058] Next, after determining the order of stress factors, different levels need to be set for each stress factor. The core of accelerated life testing is to apply higher stress levels than actual operating conditions to shorten the test time. The stress levels of speed, temperature and flow rate usually need to be set within the upper limit of the rated operating conditions of the pump. For example, the speed can be set to 1.2 times the normal operating speed, the temperature can be increased by 20-30°C compared to the normal operating temperature, and the flow rate can be increased by 10-20%. These stress levels should be within the safe operating range of the pump, but sufficient to accelerate its degradation.

[0059] After setting the stress factor level, it is also necessary to make arrangements for the duration of each stress factor. The duration setting needs to take into account the strength of the stress factor and the pump's tolerance. If the stress level is set too high, the duration needs to be reduced accordingly to prevent the pump from failing excessively in a very short time. For example, when the speed increases to 120%, the duration can be set from a few hours to more than ten hours, depending on the design specifications of the pump and the purpose of the experiment. The stress application time of temperature and flow can also be adjusted according to actual needs.

[0060] In order to ensure the smooth progress of the entire accelerated life test, the accelerated load spectrum needs to apply stress in a step-by-step manner. This means that the application of different stress factors is gradually increased, rather than applying all stresses at once. In this way, the degradation of the pump can be observed step by step, and equipment failure caused by excessive instantaneous stress can be avoided. At the same time, step-by-step stress application helps to collect degradation data at different stages during the test, which is convenient for subsequent data analysis and life prediction.

[0061] When generating the accelerated load spectrum, the interaction between the various stress factors must also be considered. For example, an increase in speed will cause the temperature inside the pump to rise, so when applying speed stress, the temperature stress setting needs to adapt to this change. Similarly, an increase in flow will affect the hydraulic load of the pump, thereby indirectly affecting the stress level of speed and temperature. Therefore, when generating the accelerated load spectrum, the dynamic relationship between the various stress factors must be comprehensively considered to ensure that the load spectrum can truly reflect the actual working conditions of the pump.

[0062] Ultimately, the generated accelerated load spectrum will be used to guide the execution of the entire accelerated life test. By accurately setting the order, level and duration of stress factors, the test process can accelerate the degradation of the pump without exceeding its physical limits, thereby obtaining accurate life prediction data.

[0063] Furthermore, the execution time of the stepped timing truncation accelerated life test is divided into multiple stages, the stress application time period of each stage increases, and the degradation state of the centrifugal pump is recorded at the end of each stage, and staged life data is generated.

[0064] The execution time of the stepped time truncated accelerated life test is divided into multiple stages, each corresponding to a specific stress application cycle. In each stage, the applied stress level and duration are gradually increased to simulate the various stress changes that a centrifugal pump may encounter in long-term use. This staged approach ensures that the degradation process of the pump under different stress conditions can be fully and accurately recorded.

[0065] Specifically, the stress application period of each stage is incremental. For example, the initial stage can start with a lower stress level and last for a short period of time (such as a few hours). As the test progresses, the subsequent stages gradually increase the stress level and duration, which allows the pump to operate for a longer time under higher intensity stress, thereby accelerating its degradation process. The stress factors include the gradual increase of speed, temperature and flow rate, and must follow the provisions of the accelerated load spectrum to ensure that the test conditions meet the experimental design.

[0066] At the end of each stage, the current degradation state of the centrifugal pump is recorded. The degradation state is usually monitored by recording the pump outlet pressure, impeller vibration signal and other relevant operating parameters. These data reflect the gradual decay of the pump performance with the accumulation of time and stress. Through these timed records, technicians can obtain the degradation trend of the pump under different stress levels and identify key degradation nodes.

[0067] After recording the degradation state, the staged life data for each stage is generated. The staged life data includes the performance changes and degradation rate of the pump after running in each stress stage. These data provide continuous and staged inputs for the construction of the entire life model, enabling the model to more accurately predict the life of the pump in actual operation. The processing of these data usually involves fitting and analyzing the pressure and vibration signals to generate a degradation curve that reflects the performance attenuation trend of the pump.

[0068] This test method of gradual pressurization and staged recording can not only avoid premature failure under a single stress condition, but also provide technicians with sufficient data to analyze degradation modes and failure mechanisms under different stress levels. By implementing this method, the stress application time can be accurately controlled and staged life data can be collected, so as to comprehensively evaluate the life performance of the centrifugal pump within a limited test time.

[0069] Step S105: performing a step-by-step timing truncation accelerated life test according to the generated accelerated load spectrum; recording the outlet pressure and impeller vibration signal data of the centrifugal pump at different stress levels.

[0070] In step S105, a step-by-step timed truncation accelerated life test is performed according to the generated accelerated load spectrum. The purpose of the test is to accelerate the performance degradation of the centrifugal pump by gradually increasing its stress level, thereby simulating its long-term life process in a shorter time. In order to ensure that the test can be effectively implemented, the previously generated accelerated load spectrum must first be strictly followed, which specifies the order, level and duration of stress factors such as speed, temperature and flow rate.

[0071] At the beginning of the test, the pump is running in normal conditions, at which the stress level is low, and the speed, temperature and flow rate are gradually stressed. The stress level is gradually increased in a predetermined step-by-step manner, and the application time is determined by the duration in the acceleration load spectrum. Step timing means that at each stress level, the pump will run for a set period of time, and then in the next period of time, a higher stress level will be applied until the predetermined stress limit is reached. The application time of each stress level is set according to the load-bearing capacity of the pump and the expected degradation rate in order to balance the gradual accumulation of stress and the safe operation of the pump.

[0072] During the entire test process, the pump outlet pressure and impeller vibration signal data are recorded in real time. These two types of data are key indicators of the degradation process. The change in outlet pressure reflects the attenuation of the pump's hydraulic performance with the stress level, while the change in the impeller vibration signal reflects the wear and vibration of the mechanical parts inside the pump. Therefore, it is very important to install high-precision sensors for collecting pressure and vibration signals. The pressure sensor is usually installed at the outlet of the pump and can record the pressure value of the liquid flow in real time under different stress levels; the acceleration sensor is installed outside the pump housing to monitor the vibration of the impeller and record its vibration signal under high speed, temperature and flow conditions.

[0073] To ensure data accuracy, the sampling frequency of the sensor needs to adapt to the operating conditions of the pump. The sampling frequency of the pressure sensor is usually set to 1 to 10 Hz to capture the long-term trend of pressure changes. Since the acceleration sensor needs to monitor high-frequency vibrations, the sampling frequency should be set between 1 kHz and 10 kHz to ensure that subtle vibration changes can be captured. During the test, all data needs to be transmitted to the central control system in real time through the data acquisition system to continuously monitor and analyze the operating status of the pump.

[0074] At the same time, the test adopts a truncation design, which means that the test will end at a preset stress level or when a specific time is reached. The purpose of truncation is to terminate the test without the pump failing completely, thereby reducing unnecessary destructive losses and saving time and cost. The truncation point is set based on the design life of the pump and the actual requirements of the test, and usually ends when the pump reaches a certain stress level or continues to run for a predetermined period of time. When the truncation point is reached, although the pump has not completely failed, sufficient degradation data has been obtained for subsequent degradation trajectory calculation and life estimation.

[0075] Through this step-by-step timed truncation accelerated life test, the degradation process data of the centrifugal pump under different stress levels can be obtained in a relatively short time. These data provide an important basis for subsequent life prediction and model optimization.

[0076] Step S106: According to the recorded outlet pressure and impeller vibration signal data, a curve fitting method is used to calculate the degradation trajectory of the centrifugal pump, and a degradation curve reflecting the attenuation of pump performance is generated.

[0077] In step S106, the degradation trajectory of the centrifugal pump is calculated using a curve fitting method based on the previously recorded outlet pressure and impeller vibration signal data, in order to generate a degradation curve that can reflect the attenuation of pump performance. First of all, data processing is a key step. The pressure and vibration signal data collected in the experiment may contain noise and irregular fluctuations, so these data need to be preprocessed. Common preprocessing steps include denoising, smoothing, and filtering operations to remove external factors that may interfere with the analysis. A low-pass filter or other suitable signal processing method can be used to eliminate high-frequency noise to ensure that the data can accurately reflect the degradation trend of the pump.

[0078] Next, it is necessary to select a suitable curve fitting method to fit the preprocessed outlet pressure and impeller vibration signals. Generally, curve fitting can adopt a variety of different mathematical models, such as polynomial fitting, exponential fitting or other nonlinear fitting methods. The specific method to be selected should be determined according to the actual degradation behavior of the pump. For example, the pressure signal will show a gradual downward trend as the pump degrades, while the vibration signal may gradually increase with the increase of wear. Therefore, exponential or power function fitting may be more suitable for reflecting these nonlinear changes.

[0079] During the fitting process, the data of the outlet pressure and impeller vibration signal need to be processed separately. These two signals reflect the performance degradation of different aspects of the pump, so they cannot be simply mixed together, but must be fitted separately. The pressure signal data usually appears as a curve that gradually decreases over time, so the goal of fitting is to find a function that can accurately describe this attenuation trend. The vibration signal gradually increases over time, reflecting the wear and imbalance of mechanical parts, so it is also necessary to find a function that can reflect its growth trend. During the fitting process, commonly used fitting algorithms such as the least squares method can be used to optimize the fitting effect to ensure that the fitting curve can accurately follow the changing trend of the data points.

[0080] Once the fitting of the outlet pressure and impeller vibration signals is completed, the corresponding degradation trajectory can be generated. The degradation trajectory describes the performance attenuation process of the centrifugal pump under different stress levels, and these trajectories will be used for subsequent life prediction and model optimization. The degradation curve not only needs to show the degradation process of the pump under high stress conditions, but also needs to reflect the degradation differences under various stress levels, especially in the process of gradually increasing stress, the change of the curve can reflect the fatigue, wear and aging rate of the pump.

[0081] In actual operation, the generation of degradation curves is a dynamic process. After each stress stage, the fitting results are updated and the model is adjusted. By continuously collecting new data and updating the curve, the real-time degradation of the pump can be more accurately reflected. The generated degradation curve will serve as an important basis for subsequent analysis and life prediction, helping to evaluate the expected life of the pump under different stress conditions.

[0082] Therefore, by reasonably selecting the curve fitting method, combined with accurate preprocessing and algorithm optimization, the degradation trajectory calculation of pressure and vibration signals can be achieved, and accurate degradation curves can be generated for further life prediction and performance evaluation.

[0083] Furthermore, the degradation trajectory of the centrifugal pump is calculated by a curve fitting method based on the recorded outlet pressure and impeller vibration signal data to generate a degradation curve reflecting the attenuation of pump performance, including: Based on the outlet pressure and impeller vibration signal data collected during the operation of the centrifugal pump, the data at each time point are preprocessed, including noise removal, filtering and smoothing, to reduce the impact of measurement errors on curve fitting; The pre-processed outlet pressure and impeller vibration signals are used as input data, and a nonlinear curve fitting method is selected. The fitting method dynamically adjusts the fitting parameters according to the decay trend of the pump performance over time to ensure that the fitting results can accurately reflect the performance degradation process of the pump; In the curve fitting process, the attenuation trend of the outlet pressure and the growth trend of the impeller vibration signal are fitted separately, and then the two fitting results are combined to generate the overall degradation curve of the centrifugal pump, so as to accurately reflect the comprehensive performance degradation of the pump; The degradation curve generated is used to describe the progressive degradation of the centrifugal pump performance over time, where the drop in outlet pressure and the increase in impeller vibration are considered as the main indicators that the pump is gradually approaching failure.

[0084] In this embodiment, based on the outlet pressure and impeller vibration signal data collected during the operation of the centrifugal pump, the degradation trajectory of the centrifugal pump is calculated by a curve fitting method, and a degradation curve reflecting the attenuation of pump performance is generated. The process first requires preprocessing of the data to ensure the accuracy of the curve fitting result.

[0085] The preprocessing step includes denoising, filtering and smoothing the collected raw data. In actual operation, the outlet pressure and impeller vibration signals collected by the sensor may be interfered by various noises, such as environmental noise, electromagnetic interference, etc., which will affect the accuracy of the data. Therefore, filtering technology (such as low-pass filter or band-pass filter) is used to remove high-frequency noise, and smoothing is used to reduce the mutation of the data to ensure that the data is more continuous and smooth. The key to this step is to reduce measurement errors and ensure that the signal can truly reflect the operating status and degradation trend of the centrifugal pump.

[0086] After the data preprocessing is completed, the next step is to use the processed outlet pressure and impeller vibration signals as input data and select an appropriate nonlinear curve fitting method for fitting. Nonlinear curve fitting can capture the complex attenuation trend of pump performance over time, rather than a simple linear decline. This is because the performance of a centrifugal pump usually degrades over time, the pressure gradually decreases, and the vibration signal gradually increases with component wear and imbalance. In order to better reflect this complex change, the fitting method needs to be able to dynamically adjust the fitting parameters to ensure that the fitting curve can closely follow the actual changes in the data.

[0087] During the curve fitting process, the attenuation trend of the outlet pressure and the growth trend of the impeller vibration signal need to be fitted separately. The change in outlet pressure reflects the gradual decline in hydraulic performance of the pump during operation, while the increase in the impeller vibration signal reflects the wear and vibration increase of the mechanical components inside the pump. These two signals represent different aspects of pump degradation and therefore need to be processed and fitted separately. The fitting of the pressure signal is usually manifested as a curve that gradually decreases over time, while the vibration signal shows a trend that gradually increases over time. By fitting these two trends separately, the overall picture of the pump performance can be better captured.

[0088] After the individual fits are completed, the two fit results are combined to generate the overall degradation curve of the centrifugal pump. This degradation curve can accurately describe the comprehensive degradation of the performance of the centrifugal pump over time, where the drop in outlet pressure and the increase in impeller vibration are considered to be the two main indicators that the pump is gradually approaching failure. By combining the fitting results of the pressure and vibration signals, a complete degradation curve can be obtained, reflecting the entire process of the pump from the beginning of operation to the approach of failure.

[0089] The generated degradation curve is used to evaluate the performance attenuation of the centrifugal pump at different operating time points, and can provide a reference for subsequent life prediction and fault warning. This curve intuitively shows the health status of the pump at different time points. As the outlet pressure decreases and the impeller vibration increases, the performance of the pump gradually declines and eventually approaches the failure point. This method can be used to accurately simulate and evaluate the degradation process of the centrifugal pump, and the generated degradation curve can be used to effectively analyze and predict the operating status of the pump.

[0090] Step S107: Based on the generated degradation curve, a failure threshold of the centrifugal pump is set, and the moment when the degradation curve reaches the failure threshold is determined; the mathematical relationship between the degradation data and the failure moment is fitted by the least squares method, and based on the fitted mathematical relationship, the pseudo life value of the centrifugal pump under accelerated stress conditions is estimated.

[0091] In step S107, based on the previously generated degradation curve, the failure threshold of the centrifugal pump needs to be set first. The failure threshold is used to determine when the pump has reached an unacceptable degradation level, and is usually determined based on the design requirements and operating performance standards of the pump. The failure threshold can be set by analyzing degradation signals, such as the drop in outlet pressure and the growth rate of the impeller vibration signal, combined with historical data and operating specifications to determine a reasonable threshold. For example, when the outlet pressure may drop to a certain percentage of the original value (for example, 70%), or when the impeller vibration signal exceeds a certain limit, the pump is considered to have failed.

[0092] Once the failure threshold is determined, the next task is to determine the specific moment when the degradation curve reaches the failure threshold. This moment represents the failure time of the centrifugal pump under accelerated stress conditions, reflecting the time required for the pump to reach an unacceptable performance degradation state from the start of operation. In order to determine this moment, it is necessary to compare the set failure threshold with the degradation curve and analyze the curve to find the point where the degradation signal first reaches or exceeds the failure threshold. This moment corresponds to the failure time of the pump under specific accelerated stress conditions, and this data will be used for subsequent pseudo-life calculations.

[0093] Next, the least squares method is used to fit the relationship between the degradation data and the failure moment. The least squares method is a commonly used statistical method that is used to minimize the error between the fitting curve and the actual data points to ensure that the fitting curve can accurately reflect the changing trend of the data. Specifically, the least squares method continuously adjusts the parameters in the model to make the fitting curve as close as possible to the actual degradation data. During the fitting process, multiple failure moments need to be fitted with the corresponding degradation curve segments to ensure that the model can describe the degradation behavior of the centrifugal pump under different stress conditions.

[0094] Based on the mathematical relationship fitted by the least squares method, the pseudo life value of the centrifugal pump under accelerated stress conditions can be estimated. The pseudo life value refers to the expected life of the pump under accelerated stress conditions. Since the accelerated stress conditions are much higher than the actual working conditions, the pseudo life value is usually shorter than the actual life under conventional stress conditions. In order to calculate the pseudo life value, it is necessary to comprehensively consider the influence of the accelerated stress factor and use the fitted mathematical model to relate the degradation data to the failure time, so as to obtain the expected life of the pump under accelerated stress conditions.

[0095] The key to this process is to establish a mathematical relationship between the failure moment and the degradation trajectory through fitting, so that the subsequent pseudo-life estimation has a high degree of accuracy. The final pseudo-life value will be used for further analysis and play an important role in model optimization and life prediction under conventional stress conditions.

[0096] Furthermore, the setting of the failure threshold of the centrifugal pump based on the generated degradation curve and determining the time when the degradation curve reaches the failure threshold includes: Based on the historical operation data of the centrifugal pump and its actual performance degradation characteristics, setting a failure threshold of the centrifugal pump, wherein the failure threshold is set by defining that the outlet pressure drops to a predetermined percentage of the initial value or the impeller vibration signal exceeds a specific limit; By analyzing the generated degradation curve point by point, the moment when the degradation signal reaches the set failure threshold is determined, and this moment is recorded as the failure time of the centrifugal pump; wherein the failure time is used to characterize the performance failure point of the centrifugal pump under accelerated stress conditions, reflecting the key degradation state of the pump.

[0097] Setting the failure threshold of the centrifugal pump and determining the moment when the degradation curve reaches the threshold is intended to identify the moment when the pump approaches or reaches the failure state by analyzing the degradation process of the pump. First, the setting of the failure threshold is based on the historical operating data of the centrifugal pump and the characteristics of its performance degradation. This means that it is necessary to refer to the actual performance of the pump in long-term operation, including the pressure change and the trend of the vibration signal. These historical data provide key information about the various stages of the pump's normal operation and degradation process, laying the foundation for setting a reasonable failure threshold.

[0098] The failure threshold can be determined in two main ways: The first is based on the drop in outlet pressure. Outlet pressure is a core indicator of pump performance and will drop as the pump degrades. Therefore, the failure threshold can be set to a predetermined percentage of the pressure drop from its initial value. For example, if the initial pressure of the pump is When the pressure drops to When the vibration signal exceeds 70% or 50%, it may mean that the pump is close to failure. The second method is based on the increase of the impeller vibration signal. As the mechanical parts inside the pump wear and become unbalanced, the vibration signal will increase significantly. The failure threshold of the vibration signal can be determined through experiments and analysis. When the vibration signal exceeds a specific limit, it usually means that the mechanical system of the pump has been severely degraded and is close to failure.

[0099] Once the failure threshold is set, the next step is to analyze the generated degradation curve point by point to determine when the degradation signal reaches the set failure threshold. This process is achieved by comparing each data point in the degradation curve with the preset failure threshold. When a data point in the curve (whether it is the outlet pressure or the impeller vibration signal) reaches or exceeds the failure threshold, it is considered that the pump has reached or is close to failure at that moment. This moment is the failure time of the centrifugal pump, indicating that the performance degradation of the pump has reached an unacceptable level.

[0100] The determination of failure time is of great significance for the life prediction and operation and maintenance of the pump. This time reflects the degradation limit of the pump under accelerated stress conditions and is the key data for evaluating the life of the pump under harsh working conditions. By recording this failure time, we can gain a deep understanding of the degradation rate of the pump under different stress conditions, and optimize the operating parameters of the pump accordingly, or formulate a reasonable maintenance and replacement plan.

[0101] In this embodiment, the failure threshold can be reasonably set by referring to the historical data and actual performance of the centrifugal pump, and the failure time of the pump can be determined by analyzing the data points in the degradation curve. This method provides a comprehensive analysis of the degradation behavior of the centrifugal pump under accelerated stress conditions, which helps to accurately predict the life of the pump and effectively manage its operation in practical applications.

[0102] Furthermore, the method of fitting the mathematical relationship between the degradation data and the failure time by the least square method and estimating the pseudo life value of the centrifugal pump under accelerated stress conditions based on the fitted mathematical relationship includes: The recorded degradation curve and failure time data are used as input, and the relationship between the degradation signal and the failure time is fitted using the least square method. The fitting results are gradually optimized to ensure that the degradation curve can accurately describe the degradation process of the pump. Based on the fitted mathematical relationship and combined with the influence weights of various stress factors under accelerated stress conditions, the pseudo life value of the centrifugal pump under accelerated stress conditions is estimated; wherein the pseudo life value is used to predict the expected failure time of the centrifugal pump in the accelerated test.

[0103] In this embodiment, the mathematical relationship between the degradation data and the failure time is fitted based on the least squares method, and the pseudo life value of the centrifugal pump under accelerated stress conditions is finally estimated. This process first requires the previously generated degradation curve and the determined failure time as input data. The degradation curve contains the outlet pressure and impeller vibration signals that change over time during the operation of the pump. These signal data reflect the process of gradual decline in the performance of the pump. The failure time indicates that the pump has reached the set failure threshold at a certain moment, indicating that the performance of the pump at that point in time can no longer meet the expected requirements.

[0104] Next, the least squares method is used to fit the relationship between the degradation curve and the failure moment. The least squares method is a common mathematical method that optimizes the accuracy of the model by minimizing the error between the fitted curve and the actual data points. In this process, the goal of fitting is to find a mathematical relationship that best matches the change in the degradation signal over time with the failure moment. The process of gradually optimizing the fitting results requires adjusting the parameters in the fitting model to ensure that the final fitting curve can accurately describe the degradation behavior of the pump. For example, the gradual decrease in outlet pressure and the gradual increase in vibration signals over time are two key trends, so the fitting model must be able to fully capture these dynamic changes.

[0105] Once the least squares method successfully fits the relationship between the degradation curve and the failure time, the pseudo-life value of the centrifugal pump is further estimated based on this fitted mathematical relationship. The pseudo-life value refers to the expected failure time of the pump under accelerated stress conditions. Although the life under such accelerated conditions is different from the actual life of the pump under conventional stress, it can effectively reflect the degradation rate and failure trend of the pump under high stress conditions. In order to calculate the pseudo-life value, it is also necessary to combine the influence weights of various accelerated stress factors, including the acceleration effect of speed, temperature and flow. The weights of these stress factors reflect the impact of different stress conditions on the degradation rate of the pump. The fitting model will comprehensively consider these factors and introduce them into the pseudo-life calculation to more accurately estimate the accelerated life.

[0106] Ultimately, the pseudo-life value provides the expected failure time of the centrifugal pump in the accelerated life test. Since the life under accelerated stress conditions is short, the pseudo-life value helps to quickly evaluate the performance limit of the pump in a shorter time and provide a basis for life prediction under conventional stress conditions. In this way, the degradation pattern of the pump can be accurately captured, and reasonable maintenance and replacement decisions can be made based on the fitted pseudo-life value.

[0107] Step S108: Based on the obtained pseudo-life value, the pseudo-life data is expanded using virtual augmented samples; the expanded pseudo-life data is fitted by Weibull distribution to improve the accelerated life model; and the life of the centrifugal pump under conventional stress conditions is estimated according to the improved accelerated life model.

[0108] In step S108, based on the obtained pseudo-life value, it is necessary to use virtual augmented samples to expand the pseudo-life data. The purpose of this step is to enhance the accuracy and extensiveness of the life model by generating more virtual data points. The generation of virtual augmented samples is based on the actual pseudo-life data that has been collected, and more possible samples are expanded through simulation or statistical inference methods. This can be achieved through Monte Carlo simulation or other random generation methods to simulate pseudo-life data under different stress conditions to ensure that a wider range of stress change scenarios can be covered under limited experimental samples. Augmented samples not only make the data set more complete, but also effectively improve the robustness of the model, making it suitable for various possible operating conditions.

[0109] Once the virtual augmented samples are generated, the next step is to analyze them. By fitting the extended pseudo-life data with the Weibull distribution, the distribution characteristics of the life data can be better described. The Weibull distribution is a commonly used statistical distribution model, especially suitable for describing the failure behavior of equipment. It can simulate different types of failure modes and life distributions by adjusting the shape parameters and scale parameters. First, it is necessary to select appropriate Weibull distribution parameters, which are usually optimized by fitting the augmented sample data. The fitting process can be achieved through maximum likelihood estimation or other fitting methods to ensure that the model is as consistent as possible with the extended data set.

[0110] After fitting, the obtained Weibull distribution can more accurately reflect the degradation and life distribution characteristics of centrifugal pumps under various accelerated stress conditions. By fitting the extended pseudo-life data, the entire accelerated life model is further improved, enabling the model to predict life under more diversified stress conditions. This not only improves the prediction accuracy of the model, but also enhances its adaptability to different stress conditions.

[0111] Finally, based on the improved accelerated life model, the actual life of the centrifugal pump under conventional stress conditions can be estimated. Since the life under conventional stress conditions is difficult to obtain through direct experiments, it is necessary to rely on the accelerated life model for estimation. By converting the pseudo-life data under accelerated stress conditions into life data under conventional stress conditions, combined with the fitting results of the Weibull distribution, a reasonable life prediction value can be obtained. This prediction value is used to evaluate the expected service life of the centrifugal pump under its normal working environment, thereby providing a basis for subsequent maintenance plans and equipment management.

[0112] Furthermore, the method of expanding the pseudo-lifetime data using virtual augmented samples based on the obtained pseudo-lifetime value includes: Based on the pseudo life value obtained in the accelerated life test, it serves as the basis of life data and reflects the degradation characteristics of the centrifugal pump under accelerated stress conditions; Based on the pseudo life value, the historical life data or other relevant data are used to generate virtual augmented samples through data simulation technology; The data augmentation method of Monte Carlo simulation is used to randomly generate multiple pseudo-life data points to ensure that the augmented samples have a wide coverage under multiple stress conditions; By introducing randomness, error distribution and degradation model uncertainty, virtual augmented samples similar to actual conditions are generated to ensure that the generated virtual samples can reflect the possible life span change trend; The generated virtual augmented samples are combined with the initially obtained pseudo-lifetime values ​​to expand the pseudo-lifetime dataset and form a more complete pseudo-lifetime data distribution.

[0113] In this embodiment, based on the obtained pseudo-life values, the pseudo-life data is expanded by generating virtual augmented samples. This process starts with the pseudo-life values ​​obtained in the accelerated life test. The pseudo-life values ​​represent the expected failure time of the centrifugal pump under accelerated stress conditions and reflect the characteristics of the pump under accelerated degradation conditions. The pseudo-life values ​​are the basis for subsequent data expansion, indicating the degradation mode of the pump under the combined action of different stress factors in the test.

[0114] Next, based on these pseudo-life values, it is necessary to combine historical life data or other relevant data to generate virtual augmented samples through data simulation technology. Historical life data usually contains operating life records of pumps under normal stress conditions or data sets from previous tests, which provide a real reference basis. Through data simulation technology, the system can infer and generate additional virtual samples based on existing life data. Virtual samples are generated by introducing multiple variables and randomness based on existing data to create life data points similar to actual operating conditions.

[0115] In this process, Monte Carlo simulation technology is used for data enhancement. This method generates multiple pseudo-life data points through multiple random sampling to ensure that the generated virtual augmented samples can cover different stress conditions and multiple operating conditions. The advantage of Monte Carlo simulation is that it can simulate many possible life distributions and reflect the possible failure time of centrifugal pumps under different accelerated stress conditions. This randomly generated pseudo-life data point covers a wide range of stresses, ensuring that the augmented samples are not limited to the test results under a specific condition, so that a wider range of life change trends can be reflected.

[0116] In order to further improve the accuracy and diversity of virtual augmented samples, randomness, error distribution, and uncertainty of degradation models are also introduced in the simulation process. These factors help the generated virtual samples to be closer to the uncertain environment in the real world. For example, the randomness factor simulates the non-ideal conditions that may be encountered in actual operation, the error distribution reflects the error fluctuations under different measurement conditions, and the uncertainty of the degradation model explains the diverse degradation performance of the pump under different stress conditions by introducing different stress combinations. With the addition of these diverse factors, the generated virtual augmented samples can more accurately reflect the actual life changes.

[0117] Finally, these generated virtual augmented samples are combined with the initially obtained pseudo-life values ​​to form a more complete pseudo-life dataset. This expanded pseudo-life dataset has a wider coverage and richer diversity, and can provide reliable input data for subsequent life prediction and model optimization. By expanding the data, the degradation behavior of centrifugal pumps under various stress conditions can be more accurately analyzed, thereby better evaluating the life and performance of the pump.

[0118] Furthermore, the Weibull distribution fitting adopts the following formula 4: in, For time The probability of failure when is a scale parameter, reflecting the characteristic time of life; is the shape parameter; is the amplitude of the sinusoidal fluctuation; is the frequency of the sinusoidal oscillation; is the amplitude of the cosine fluctuation; is the frequency of the cosine wave; are the rotation speed, temperature and flow rate in the stress factor respectively; is the sensitivity coefficient, reflecting the influence of different stress factors on life; is the stress adjustment coefficient, which is used to adjust the overall impact of the stress factor.

[0119] In this embodiment, the Weibull distribution fitting method is used to estimate the failure probability of the centrifugal pump through formula (4): , to accurately describe the centrifugal pump in time Formula (4) is an extended form of Weibull distribution, which combines stress factor and fluctuation term to reflect the dynamic change of life in a more complete way.

[0120] First, the formula Is the pump at time The failure probability at time , indicates the possibility of the pump failing at that point in time. The value of gradually approaches 1, indicating that the failure probability increases.

[0121] In the formula It is a scale parameter, which indicates the characteristic time of life. This parameter is usually obtained by fitting experimental data and represents the average life time of the pump under specific stress conditions. The size directly affects The growth rate of When the value is larger, the pump life is longer; when When it is smaller, the life of the pump is shorter.

[0122] Shape parameters Controls the steepness of the failure probability curve, that is, how quickly the failure probability increases over time. , different types of failure modes can be simulated. For example, smaller Values ​​are usually used to simulate early failures (i.e., the failure probability is higher at the beginning), while larger It is used to simulate failure due to degradation over time.

[0123] Next, the formula also includes sine and cosine fluctuation terms, which are used to simulate the periodic fluctuation of life over time. By parameters Control the fluctuation range, Control fluctuation frequency. This term is introduced to simulate the fluctuation of failure probability caused by periodic stress (such as flow fluctuation or mechanical vibration) during the operation of the pump.

[0124] Cosine wave term It also plays a similar role, but its fluctuation pattern is expressed by the cosine function, and the parameter Control Range Control frequency. Compared with the sine wave, the cosine term increases the inverse decay of the square of time, which means that the fluctuation gradually decreases with the increase of time, reflecting the improvement of the stability of the pump in the long-term operation.

[0125] The stress factor is composed of and They represent speed, temperature and flow respectively. These stress factors reflect the main stresses that the pump is subjected to under different operating conditions. Each stress factor and its corresponding sensitivity coefficient and The sensitivity coefficient is determined by experiments or historical data and is usually used to quantify the accelerating effect of a specific stress on the life of the pump.

[0126] In the formula is the stress adjustment factor, which is used to adjust the overall impact of all stress factors. It acts as an adjustment parameter to adjust the weight of stress factors in life prediction according to different accelerated stress conditions. By combining all these factors, the formula can reflect the combined impact of different stresses and fluctuations on pump life, providing a more accurate life model.

[0127] In actual operation, the various parameters in the formula can be obtained through data fitting of accelerated life tests and historical data analysis. First, the experimental measurement and data fitting are used to determine and Then, the stress factor is determined based on the experimental data of the pump under different stress conditions. and its corresponding sensitivity coefficient Fluctuation parameters , and frequency , Fitting and optimization can be performed by observing the performance fluctuations of the pump in different time periods.

[0128] This formula can accurately predict the failure probability of the pump under accelerated stress conditions, providing a more reliable basis for life prediction.

[0129] A second embodiment of the present application provides an electronic device, the electronic device comprising: processor; The memory is used to store a program. When the program is read and executed by the processor, the program executes the accelerated life test method of a step-by-step timing truncated centrifugal pump based on three stresses provided in the first embodiment of the present application.

[0130] The third embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the accelerated life test method of a step-by-step timing truncated centrifugal pump based on three stresses provided in the first embodiment of the present application is executed.

[0131] Although the present application is disclosed as above in the form of a preferred embodiment, it is not intended to limit the present application. Any technical personnel in this field may make possible changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be based on the scope defined by the claims of the present application.

Claims

1. A three-stress based step-by-step timing truncated centrifugal pump accelerated life test method, characterized in that: include: Based on the operating performance of the centrifugal pump, the outlet pressure is selected as the main degradation signal and the impeller vibration signal is selected as the auxiliary degradation signal, wherein the main degradation signal and the auxiliary degradation signal are used to evaluate the performance degradation of the centrifugal pump; According to the wear, aging and fatigue failure mechanism of the centrifugal pump, the rotation speed, temperature and flow rate are determined as accelerated stress factors, wherein the accelerated stress factors are used as stress application conditions in the life test; Based on the generalized Eyring model, combined with the determined accelerated stress factor and the main degradation signal and the auxiliary degradation signal, an accelerated life model of the centrifugal pump is constructed; wherein the accelerated life model is used to describe the influence of stress on the life of the centrifugal pump; Based on the constructed accelerated life model, an accelerated load spectrum is generated, wherein the accelerated load spectrum specifies the application order, level and duration of different accelerated stress factors for the specific execution of the accelerated life test; According to the generated accelerated load spectrum, a step-by-step timing truncation accelerated life test is carried out; the outlet pressure and impeller vibration signal data of the centrifugal pump at different stress levels are recorded; Based on the recorded outlet pressure and impeller vibration signal data, the degradation trajectory of the centrifugal pump is calculated using a curve fitting method to generate a degradation curve reflecting the attenuation of pump performance; Based on the generated degradation curve, the failure threshold of the centrifugal pump is set, and the moment when the degradation curve reaches the failure threshold is determined; the mathematical relationship between the degradation data and the failure moment is fitted by the least squares method, and based on the fitted mathematical relationship, the pseudo life value of the centrifugal pump under accelerated stress conditions is estimated; Based on the obtained pseudo-life values, the pseudo-life data are expanded using virtual augmented samples; the expanded pseudo-life data are fitted with Weibull distribution to improve the accelerated life model; based on the improved accelerated life model, the life of the centrifugal pump under conventional stress conditions is estimated.

2. The accelerated life test method of a centrifugal pump based on three stresses according to claim 1 is characterized in that: The outlet pressure is used as the main degradation signal and is collected in real time by a pressure sensor installed at the outlet of the centrifugal pump. The impeller vibration signal is used as the auxiliary degradation signal and is collected in real time by an acceleration sensor installed outside the pump casing to monitor the mechanical vibration during the operation of the centrifugal pump.

3. The accelerated life test method of a centrifugal pump based on three stresses and step-by-step timing and tail-cutting according to claim 1 is characterized in that: The temperature in the accelerated stress factor is monitored in real time by multi-point temperature sensors installed in the pump body, and the temperature variation range is limited to between 10°C and 120°C to prevent premature failure of the pump seal due to excessive temperature.

4. The accelerated life test method of a centrifugal pump based on three stresses and step-by-step timing and tail-cutting according to claim 1 is characterized in that: The execution time of the step-by-step timing truncation accelerated life test is divided into multiple stages, the stress application time period of each stage increases, and the degradation state of the centrifugal pump is recorded at the end of each stage, and staged life data is generated.

5. The accelerated life test method of a centrifugal pump based on three stresses and step-by-step timing and tail-cutting according to claim 1 is characterized in that: The accelerated life model is constructed based on the generalized Eyring model and is implemented using the following formula 1: in, Indicates the centrifugal pump at time Accelerated life under is an empirical constant; are the stress values ​​of rotation speed, temperature and flow rate respectively; is the sensitivity coefficient, which reflects the influence of stress factors on life; The main degradation signal over time The change function of To assist the degradation signal over time The change function of and is a weighted index related to the degradation signal, which controls the weight of the impact of pressure and vibration on life respectively; is the nonlinear adjustment parameter in the accelerated life model; Among them, the function Use the following formula 2: in, It is the outlet pressure of the centrifugal pump when it is initially running, indicating the time Initial pressure value at ; is the running time of the experiment, which represents the time variable from the start of the experiment to the current moment; Indicates the design life of the centrifugal pump, indicating the expected operating time of the equipment under normal conditions; is the pressure decay rate parameter, which is used to control the speed and nonlinearity of pressure decay; is the amplitude of noise oscillation; is the frequency of noise oscillation; the initial magnitude of the correction term; is the decay constant; function Use the following formula 3: in, It is the impeller vibration signal value of the centrifugal pump during initial operation, indicating the time The initial vibration amplitude at ; is the running time of the experiment, which represents the time variable from the start of the experiment to the current moment; It is the design life time of the centrifugal pump, indicating the expected operating time of the equipment under normal conditions; Add rate parameters to the vibration signal to control the nonlinear degree of vibration increase over time; the magnitude of the disturbance term; is the frequency of the disturbance term; is the amplification constant.

6. The accelerated life test method of a centrifugal pump based on three stresses and step-by-step timing and tail-cutting according to claim 1 is characterized in that: The method of calculating the degradation trajectory of the centrifugal pump by using a curve fitting method based on the recorded outlet pressure and impeller vibration signal data to generate a degradation curve reflecting the attenuation of pump performance includes: Based on the outlet pressure and impeller vibration signal data collected during the operation of the centrifugal pump, the data at each time point are preprocessed, including noise removal, filtering and smoothing, to reduce the impact of measurement errors on curve fitting; The pre-processed outlet pressure and impeller vibration signals are used as input data, and a nonlinear curve fitting method is selected. The fitting method dynamically adjusts the fitting parameters according to the decay trend of the pump performance over time to ensure that the fitting results can accurately reflect the performance degradation process of the pump; In the curve fitting process, the attenuation trend of the outlet pressure and the growth trend of the impeller vibration signal are fitted separately, and then the two fitting results are combined to generate the overall degradation curve of the centrifugal pump, so as to accurately reflect the comprehensive performance degradation of the pump; The degradation curve generated is used to describe the progressive degradation of the centrifugal pump performance over time, where the drop in outlet pressure and the increase in impeller vibration are considered as the main indicators that the pump is gradually approaching failure.

7. The accelerated life test method of a centrifugal pump based on three stresses and step-by-step timing and tail-cutting according to claim 1 is characterized in that: The step of setting a failure threshold of the centrifugal pump based on the generated degradation curve and determining the time when the degradation curve reaches the failure threshold comprises: Based on the historical operation data of the centrifugal pump and its actual performance degradation characteristics, setting a failure threshold of the centrifugal pump, wherein the failure threshold is set by defining that the outlet pressure drops to a predetermined percentage of the initial value or the impeller vibration signal exceeds a specific limit; By analyzing the generated degradation curve point by point, the moment when the degradation signal reaches the set failure threshold is determined, and this moment is recorded as the failure time of the centrifugal pump; wherein the failure time is used to characterize the performance failure point of the centrifugal pump under accelerated stress conditions, reflecting the key degradation state of the pump.

8. The accelerated life test method of a centrifugal pump based on three stresses and step-by-step timing and tail-cutting according to claim 1 is characterized in that: The method of fitting the mathematical relationship between the degradation data and the failure time by the least square method and estimating the pseudo life value of the centrifugal pump under accelerated stress conditions based on the fitted mathematical relationship includes: The recorded degradation curve and failure time data are used as input, and the relationship between the degradation signal and the failure time is fitted using the least squares method. The fitting results are gradually optimized to ensure that the degradation curve can accurately describe the degradation process of the pump. Based on the fitted mathematical relationship and combined with the influence weights of various stress factors under accelerated stress conditions, the pseudo life value of the centrifugal pump under accelerated stress conditions is estimated; wherein the pseudo life value is used to predict the expected failure time of the centrifugal pump in the accelerated test.

9. The accelerated life test method of a centrifugal pump based on three stresses and step-by-step timing and tail-cutting according to claim 1 is characterized in that: The method of expanding the pseudo-lifetime data by using virtual augmented samples based on the obtained pseudo-lifetime value comprises: Based on the pseudo life value obtained in the accelerated life test, it serves as the basis of life data and reflects the degradation characteristics of the centrifugal pump under accelerated stress conditions; Based on the pseudo life value, the historical life data or other relevant data are used to generate virtual augmented samples through data simulation technology; The data augmentation method of Monte Carlo simulation is used to randomly generate multiple pseudo-life data points to ensure that the augmented samples have a wide coverage under multiple stress conditions; By introducing randomness, error distribution and degradation model uncertainty, virtual augmented samples similar to actual conditions are generated to ensure that the generated virtual samples can reflect the possible life span change trend; The generated virtual augmented samples are combined with the initially obtained pseudo-lifetime values ​​to expand the pseudo-lifetime dataset and form a more complete pseudo-lifetime data distribution.

10. The accelerated life test method of a centrifugal pump based on three stresses and step-by-step timing and tail-cutting according to claim 1 is characterized in that: The Weibull distribution is fitted using the following formula 4: in, For time The probability of failure when is a scale parameter, reflecting the characteristic time of life; is the shape parameter; is the amplitude of the sinusoidal fluctuation; is the frequency of the sinusoidal oscillation; is the amplitude of the cosine fluctuation; is the frequency of the cosine wave; are the rotation speed, temperature and flow rate in the stress factor respectively; is the sensitivity coefficient, reflecting the influence of different stress factors on life; is the stress adjustment coefficient, which is used to adjust the overall impact of the stress factor.

Citation Information

Patent Citations

  • A Modeling Method for Accelerated Life Testing of Aviation Hydraulic Pumps Based on Exponential-Inverse Power Law

    CN108470082B

  • Energy-saving optimization method for accelerated life test of axial plunger pump

    CN115758754A

  • Accelerated life test system and method for linear conjugate internal gear pump

    CN115853767A

  • Plunger-pump acceleration factor determining method based on theoretical models

    CN104156621A

  • Reliability evaluation method for long-service-life and high-reliability electronic product

    CN104182603A

Cited By

  • Reliability analysis and maintenance method based on wind power generation multi-component degradation modeling

    CN119226760A

  • Reliability analysis and maintenance method based on multi-component degradation modeling of wind power generation

    CN119226760B

  • Accelerated aging evaluation method for friction performance of surface coating of sliding bearing

    CN122409400A