Mechanical seal leakage detection and life prediction system and prediction method thereof

Abstract

The application relates to a mechanical seal leakage detection and life prediction system and a prediction method thereof, and belongs to the technical field of mechanical seal performance testing and fault diagnosis, which comprises a base, a multi-scene simulation test bed, an environmental parameter adjusting module, a leakage detection module, a control module and a life prediction module. The multi-scene simulation test bed is integrated with a variable-frequency speed-regulating motor, an axial loading mechanism, a radial vibration generator and a medium circulating system, can independently adjust the main shaft rotating speed, the axial load, the radial vibration, the medium pressure, the temperature, the type and the cleanliness, simulates multiple preset operation scenes, collects leakage rate, friction torque and sealing cavity temperature signals through the leakage detection module, and is internally provided with an algorithm based on performance degradation data fitting in the life prediction module. The application can quickly and accurately evaluate the degradation trend of the mechanical seal under complex working conditions, significantly shortens the test cycle, and the prediction result is closer to the actual application.

Description

Technical Field

[0001] This invention relates to the field of mechanical seal performance testing and fault diagnosis technology, and in particular to a mechanical seal leakage detection and life prediction system and prediction method. Background Technology

[0002] Mechanical seals are key components in rotating equipment (such as pumps, compressors, and reactors) to prevent fluid leakage. Their performance is directly related to the safety, environmental protection, and economic operation of the equipment. In practical applications, mechanical seals face the coupled effects of various complex working conditions such as start-up and shutdown, pressure fluctuations, temperature changes, media contamination, and vibration. Their failure is often the result of accelerated degradation due to the combined effects of multiple stresses.

[0003] Most existing mechanical seal testing technologies operate under a single steady-state condition (such as constant temperature, constant pressure, and constant speed), judging failure by monitoring whether the leakage rate exceeds a certain fixed threshold. However, this single-condition, threshold-based testing method has significant shortcomings: First, it cannot assess the impact of alternating or superimposed multiple operating conditions on the cumulative damage to the seal, resulting in a large deviation between the predicted results and the actual lifespan in the field. Second, traditional life tests usually require running until the seal completely fails, which is time-consuming and cannot meet the needs of rapid assessment. Finally, there is a lack of a systematic solution that can integrate multi-source signals (such as leakage rate, friction torque, and acoustic emission) and perform quantitative life prediction based on a damage accumulation model.

[0004] Therefore, there is an urgent need for a system and method that can simulate real-world complex working conditions, quickly and accurately detect the performance degradation of mechanical seals, and predict their remaining life. Summary of the Invention

[0005] This invention provides a mechanical seal leakage detection and life prediction system and prediction method, which solves the defects of existing technologies such as single test conditions for mechanical seals, difficulty in simulating multi-stress coupled service environments, reliance on single index threshold judgment for life prediction, and insufficient engineering applicability due to the complexity of multi-performance index fusion methods.

[0006] The solution to the above-mentioned technical problems of the present invention is as follows: a mechanical seal leakage detection and life prediction system and prediction method, including a base, a multi-scenario simulation test bench, an environmental parameter adjustment module, a leakage detection module, a control module and a life prediction module.

[0007] The multi-scenario simulation test bench is mounted on a base for mounting the mechanical seal under test. It includes a sealing cavity, a main shaft, a variable frequency speed-regulating motor, an axial loading mechanism, a radial vibration generator, and a media circulation system. The sealing cavity contains a stationary ring seat for fixing the stationary ring. The main shaft passes through the sealing cavity and is used to mount and rotate the rotating ring. The variable frequency speed-regulating motor is connected to the main shaft via a coupling and is configured to steplessly adjust the main shaft's speed. The axial loading mechanism is used to adjust the axial load applied to the sealing end face. The radial vibration generator is used to apply controllable radial vibration to the main shaft. The media circulation system is connected to the sealing cavity and is used to independently adjust the media pressure, temperature, type, and cleanliness within the sealing cavity. Through this configuration, the multi-scenario simulation test bench can independently control the rotational speed, axial load, radial vibration, and the pressure, temperature, type, and cleanliness of the media, thereby providing the mechanical seal with various stress inputs close to the real service environment.

[0008] The environmental parameter adjustment module is connected to the multi-scenario simulation test bench and is used to independently adjust the spindle speed, axial load, amplitude and frequency of radial vibration, medium pressure, medium temperature, medium type and medium cleanliness according to the preset scenario sequence to simulate multiple different operating scenarios. This module enables the system to automatically switch operating conditions according to the program and realize multi-scenario serialized testing, which significantly improves the automation and reproducibility of the test.

[0009] The leakage detection module is installed on the multi-scenario simulation test bench and is used to collect the leakage rate, friction torque and sealing cavity temperature signals of the mechanical seal under test in real time under various operating scenarios. Through the synchronous acquisition of multi-source signals, it can capture the performance degradation characteristics of the seal under different working conditions more comprehensively and sensitively.

[0010] The control module is electrically connected to the environmental parameter adjustment module and the leakage detection module, respectively. It is used to control the execution order and duration of the scene sequence, and to receive and store detection data. This module is the control core of the entire system, ensuring that each parameter is executed accurately according to the predetermined trajectory, and at the same time completing the centralized management of data.

[0011] The lifetime prediction module is connected to the control module. The lifetime prediction module has a built-in algorithm based on performance degradation data fitting. It is used to comprehensively predict the remaining lifetime based on the degradation trajectory of leakage rate and other auxiliary signals collected in multiple scenarios. The module converts the collected multi-dimensional degradation data into quantitative lifetime prediction results, realizing an integrated function from detection to evaluation.

[0012] Furthermore, the radial vibration generator is either an eccentric cam mechanism or a piezoelectric exciter. The eccentric cam mechanism drives the eccentric cam to rotate via a servo motor to change the eccentricity. The piezoelectric exciter inputs an excitation signal through a power amplifier to adjust the vibration amplitude and frequency. The functions and advantages of these two implementation methods are as follows: the eccentric cam mechanism has a simple structure and large output force, and can generate a large radial displacement, making it suitable for simulating low-frequency, large-amplitude radial vibrations caused by bearing wear and shaft misalignment; the piezoelectric exciter has a fast response speed and high control precision, and can generate high-frequency vibrations up to several kilohertz, making it suitable for simulating high-frequency, small-amplitude radial vibrations caused by fluid excitation and rotor imbalance. By selecting different radial vibration generators according to the test requirements, various radial vibration sources in actual working conditions can be flexibly covered, improving the adaptability and simulation accuracy of the test device.

[0013] Furthermore, the media circulation system includes a storage tank, a heater, a cooler, a pressure pump, a filter, and a media replacement valve. The storage tank is used to store the test media. The suction end of the pressure pump is connected to the storage tank, and the output end is connected to the media inlet of the sealed cavity, used to regulate the media pressure within the sealed cavity. The heater and the cooler are installed in series on the circulation pipeline between the sealed cavity and the storage tank, used to regulate the media temperature. The filter is installed at the suction end of the pressure pump or the outlet pipeline of the sealed cavity, used to regulate the media cleanliness. The media replacement valve is installed on the supply pipeline between the storage tank and the sealed cavity, used to switch between different media types. The function and advantages of this media circulation system are: through independently configured pressure pump, heater, cooler, filter, and media replacement valve, decoupled control of media pressure, temperature, cleanliness, and type is achieved. Any parameter can be adjusted independently without interfering with other parameters, thereby realistically simulating the coupled influence of on-site media conditions (such as pressure fluctuations, temperature changes, particulate contamination, media switching, etc.) on sealing performance, greatly improving the realism and repeatability of the test.

[0014] Furthermore, the leakage detection module includes a weighing sensor, a torque sensor, and an acoustic emission sensor. The weighing sensor is installed below the leakage liquid collection container of the sealed cavity or on the leakage liquid discharge pipeline to measure the leakage rate. The torque sensor is installed between the main shaft and the coupling of the variable frequency speed control motor to measure the friction torque. The acoustic emission sensor is attached to the stationary ring or the outer wall of the sealed cavity with a coupling agent to capture the contact state signal of the sealing end face. The function and advantages of this leakage detection module are: the weighing sensor provides high-precision quantitative measurement of the leakage rate; the torque sensor reflects the change in frictional power consumption of the sealing end face; and the acoustic emission sensor can sense the microscopic contact state of the sealing end face (such as friction, wear, tear, dry friction, etc.) in real time. The three sensors complement each other to form a multi-dimensional fusion detection system, which can capture the slight degradation of sealing performance in the early stage before the leakage rate has increased significantly, realize early warning, and avoid misjudgment of a single signal, thereby improving the reliability and sensitivity of detection.

[0015] Furthermore, the control module is electrically connected via fieldbus to the environmental parameter adjustment module, the frequency converter of the variable frequency speed control motor, the servo driver of the axial loading mechanism, the excitation signal source of the radial vibration generator, and the pressure pump, heater, cooler, and medium replacement valve of the medium circulation system, forming a closed-loop control circuit. The control module is also electrically connected via data acquisition card to the weighing sensor, torque sensor, and acoustic emission sensor of the leakage detection module. The function and advantages of this connection method are: using fieldbus (such as EtherCAT, PROFINET) to build a distributed closed-loop control system can realize the synchronous and precise control of all actuators, ensuring that the parameters in the complex scene sequence change in a coordinated manner according to the preset trajectory, and the control cycle can reach the millisecond level, ensuring the authenticity and repeatability of the working condition switching; at the same time, the high-precision data acquisition card synchronously acquires multiple sensor signals to ensure the consistency of the time axis, providing a high-quality data foundation for subsequent degradation trajectory analysis and life prediction.

[0016] Furthermore, the multiple different operating scenarios include at least one combination of the following: low temperature and low pressure slow start-up scenario, high temperature and high pressure high speed steady-state scenario, instantaneous pressure impact scenario, instantaneous speed rise and fall scenario, dry friction start-up scenario, operating scenario under medium containing solid particles, radial vibration scenario with preset amplitude and frequency, and shutdown cooling scenario. The function and advantage of these multiple operating scenarios are that: the above scenarios comprehensively cover the typical working conditions that mechanical seals may experience in actual service, including normal steady-state operation, start-stop transient process, abnormal impact conditions, and accelerated degradation conditions. In particular, they include radial vibration scenario and particulate medium scenario. These two types of scenarios are factors that are easily ignored in the prior art but have a significant impact on the seal life. By combining these scenarios in a preset sequence, the cumulative damage process of the seal under multiple stress alternation / superposition can be truly reflected, making the collected degradation data closer to reality, thereby improving the accuracy of subsequent life prediction.

[0017] This invention also provides a method for detecting and predicting the lifespan of mechanical seals, using the mechanical seal leakage detection and lifespan prediction system described in any of the above claims, and includes the following steps: Step 1: Install the mechanical seal to be tested on the multi-scenario simulation test bench.

[0018] Step 2: Load a predefined sequence of running scenarios through the control module. This sequence contains at least three different running scenarios, each with a preset duration and parameter change trajectory.

[0019] Step 3: During the execution of the running scenario sequence, the leakage rate, friction torque and sealing cavity temperature are continuously collected by the leakage detection module and recorded as time series data by the control module.

[0020] Step 4: Segment the leakage rate data for each operating scenario and extract the steady-state leakage rate, transient peak leakage rate, and friction torque rise rate for each scenario as performance degradation indicators in multiple dimensions.

[0021] Step 5: The life prediction module uses a weighted summation method to compare the performance degradation indicators extracted from multiple dimensions under different scenarios with the corresponding failure thresholds based on the pre-stored failure threshold parameter set corresponding to each operating scenario, and calculates the current comprehensive damage degree.

[0022] Step 6: Based on the comprehensive damage sequence obtained from previous tests, use regression analysis or exponential fitting extrapolation to predict the time point when the comprehensive damage reaches the preset failure threshold, and use this as the remaining life.

[0023] The function and advantages of this method are as follows: by loading sequences containing multiple different operating scenarios and extracting multi-dimensional performance degradation indicators from them, and then using a weighted summation method to comprehensively evaluate the current damage state, and finally using historical damage degree sequences for regression or exponential fitting extrapolation, the remaining life of mechanical seals under complex service conditions can be quantitatively predicted. Compared with the traditional single-condition threshold method, this method fully considers the different contributions of different operating conditions to the fatigue and wear of seals, and greatly improves the accuracy of prediction and engineering practicality. At the same time, by introducing accelerated degradation scenarios, a complete degradation trajectory can be obtained in a short time, significantly shortening the test cycle.

[0024] Furthermore, the predefined sequence of operating scenarios includes at least: a steady-state scenario for simulating normal operation, a transient scenario for simulating start-stop processes, and an accelerated degradation scenario for simulating abnormal operating conditions. The function and advantages of this combination of three scenarios are as follows: the steady-state scenario is used to obtain the baseline degradation rate of the seal under normal operating conditions; the transient scenario (such as start-stop and pressure shock) is used to capture the instantaneous response peak of the sealing end face when parameters change abruptly, and these peaks are often the key factors leading to early seal failure; the accelerated degradation scenario accelerates the typical failure mode of the seal in a short time by simultaneously increasing the rotational speed, temperature, load, and superimposing radial vibration and other multi-stress coupling acceleration. The organic combination of the three ensures the consistency between the degradation mechanism and actual operation, effectively shortens the test time, and achieves the best balance between test efficiency and prediction accuracy.

[0025] Furthermore, the weighting coefficients for each scenario used in the weighted summation method are pre-set according to the frequency or severity of the scenario in actual application conditions. The function and advantage of this feature are that the contribution of various operating conditions to the damage to the seal life is different under different application scenarios. For example, for pumps that are frequently started and stopped, the weight of the start-stop scenario should be higher; for compressors that run continuously, the weight of the steady-state scenario should be dominant; for harsh environments that may have media contamination or vibration, the weight of the accelerated degradation scenario needs to be increased accordingly. By pre-calibrating the weighting coefficients according to the actual operating conditions, the comprehensive damage degree can truly reflect the cumulative damage pattern under the specific application, avoiding the prediction error caused by the one-size-fits-all averaging treatment, and significantly improving the engineering adaptability and personalized prediction capability of the method.

[0026] Furthermore, the overall damage degree D in step 5 is calculated using the following formula: D =(wi(Ii / Ti)); Wherein, Ii is the performance degradation index extracted under the i-th operating scenario, Ti is the pre-stored failure threshold corresponding to the scenario, and wi is the weight coefficient of the i-th scenario, and wi = 1; the comprehensive damage degree D ranges from 0 to 1, and the seal is judged to fail when D reaches or exceeds 1.0; the life prediction module uses the exponential fitting model D(t)=ae^(bt) to fit the D value sequence obtained from previous detections, where a and b are fitting parameters, t is the running time, and the predicted total life is obtained by solving D(t)=1.0, and the remaining life is obtained by subtracting the running time. The function and advantage of this formula and fitting model are: by using the ratio Ii / Ti, performance indicators of different dimensions (such as leakage rate, torque rise rate, etc.) are normalized to the dimensionless relative damage degree. Then, a comprehensive damage degree D is obtained by weighted summation, which intuitively reflects the overall health status of the seal (0 for brand new, 1 for failed). The exponential fitting model D(t)=ae^(bt) can accurately describe the slow-accelerated nonlinear characteristics of seal performance degradation, that is, the initial degradation is slow, the wear is accelerated in the later stage, and the leakage rate increases exponentially. By performing exponential fitting on the D value sequence obtained from each test, the time point of D=1.0 can be extrapolated, thereby accurately predicting the remaining life. This method avoids the large error caused by the linear assumption and is more in line with the actual physical process of mechanical seal failure.

[0027] The beneficial effects of this invention are as follows: This invention provides a mechanical seal leakage detection and life prediction system and method, which has the following advantages: 1. By integrating a multi-scenario simulation test bench, an environmental parameter adjustment module, and a control module, this invention enables independent adjustment and sequential automatic switching of spindle speed, axial load, radial vibration, medium pressure, medium temperature, medium type, and medium cleanliness. It can simulate various typical operating scenarios, including steady-state operation, start-stop transients, pressure shocks, radial vibration, and particulate media. Compared with existing test devices with single or limited working condition combinations, this invention can realistically reproduce the alternating and superimposed effects of multiple stresses on mechanical seals in actual service, significantly improving the correlation between test results and actual on-site operating conditions.

[0028] 2. By integrating a weighing sensor, a torque sensor, and an acoustic emission sensor into the leakage detection module, multi-dimensional synchronous acquisition of leakage rate, friction torque, and sealing end face contact state is achieved. Compared with existing detection methods that rely on a single leakage rate or vibration signal, this invention can capture the slight degradation of sealing performance in the early stage before the leakage rate has increased significantly, thus achieving early warning. At the same time, the complementary use of multi-source information avoids misjudgment of a single signal, improving the reliability and sensitivity of detection.

[0029] 3. Regarding the life prediction method, this invention proposes a model for calculating the comprehensive damage degree based on weighted summation of multiple scenarios. By normalizing and comparing the performance degradation indicators extracted under each scenario, such as steady-state leakage rate, transient peak leakage rate, and friction torque rise rate, with the corresponding failure threshold, and assigning weight coefficients according to the occurrence frequency or hazard level of the scenario in actual working conditions, a dimensionless comprehensive damage degree is obtained. This method overcomes the shortcomings of traditional single-index threshold judgment, which cannot reasonably quantify the cumulative damage under multiple working conditions, and makes the prediction results more in line with actual applications.

[0030] 4. An exponential fitting model is used to extrapolate the comprehensive damage sequence obtained from each test to predict the time point when the damage reaches the failure threshold, thereby obtaining the remaining life. This fitting model can accurately characterize the slow-accelerated nonlinear characteristics that are common in the performance degradation of mechanical seals, avoiding the large errors caused by linear assumptions. At the same time, by introducing accelerated degradation scenarios, the test cycle is significantly shortened, achieving a good balance between test efficiency and prediction accuracy.

[0031] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it according to the contents of the specification, the preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings. Specific embodiments of the present invention are given in detail below with reference to the accompanying drawings. Attached Figure Description

[0032] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings: Figure 1 This is a schematic diagram of a mechanical seal leakage detection and life prediction system and prediction method provided in an embodiment of the present invention; Figure 2 A front view of a mechanical seal leakage detection and life prediction system and prediction method provided in an embodiment of the present invention; Figure 3 This is a system architecture diagram of a mechanical seal leakage detection and life prediction system and prediction method provided in an embodiment of the present invention; Figure 4 This is a flowchart of a mechanical seal leakage detection and life prediction system and prediction method provided in an embodiment of the present invention.

[0033] The attached diagram lists the components represented by each number as follows: 1. Multi-scenario simulation test bench; 101. Sealed cavity; 1011. Static ring seat; 102. Spindle; 103. Variable frequency speed control motor; 104. Axial loading mechanism; 105. Radial vibration generator; 106. Media circulation system; 1061. Liquid storage tank; 1062. Heater; 1063. Cooler; 1064. Pressure pump; 1065. Filter; 1066. Media replacement valve; 2. Environmental parameter adjustment module; 3. Leakage detection module; 301. Weighing sensor; 302. Torque sensor; 303. Acoustic emission sensor; 4. Control module; 5. Life prediction module; 6. Base. Detailed Implementation

[0034] The following is in conjunction with the appendix Figure 1-4 The principles and features of the present invention are described below. The examples given are for illustrative purposes only and are not intended to limit the scope of the invention. The invention is described more specifically in the following paragraphs by way of example with reference to the accompanying drawings. The advantages and features of the invention will become clearer from the following description. It should be noted that the drawings are in a very simplified form and use non-precise proportions, and are only used to facilitate and clarify the illustration of the embodiments of the invention.

[0035] It should be noted that when a component is said to be fixed to another component, it can be directly on the other component or it may have a component in between. When a component is said to be connected to another component, it can be directly connected to the other component or it may have a component in between. When a component is said to be set to another component, it can be directly set to the other component or it may have a component in between. The terms vertical, horizontal, left, right, and similar expressions used in this document are for illustrative purposes only.

[0036] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The terminology used herein includes, and / or encompasses, any and all combinations of one or more of the associated listed items.

[0037] Example 1, please refer to Figures 1 to 3 This embodiment provides a mechanical seal leakage detection and life prediction system, including a base 6, a multi-scenario simulation test bench 1, an environmental parameter adjustment module 2, a leakage detection module 3, a control module 4, and a life prediction module 5.

[0038] The base 6 is a cast iron platform with vibration reduction and leveling functions, used to support and fix other modules.

[0039] The multi-scenario simulation test bench 1 is installed on the base 6 and is used to install the mechanical seal to be tested. Specifically, it includes a sealing cavity 101, a main shaft 102, a variable frequency speed control motor 103, an axial loading mechanism 104, a radial vibration generator 105, and a media circulation system 106.

[0040] The sealed cavity 101 is a cylindrical pressure vessel, and a stationary ring seat 1011 for fixing the stationary ring is provided inside. The stationary ring seat 1011 is provided with a positioning pin and a sealing ring groove to ensure that the stationary ring does not rotate circumferentially or move axially after installation.

[0041] The spindle 102 passes through the sealed cavity 101 and is supported in the bearing seats on both sides of the cavity by double-row angular contact ball bearings. The part of the spindle 102 located in the sealed cavity is machined with steps and keyways for installing the rotating ring and driving it to rotate. The spindle 102 and the cavity are dynamically sealed by an oil seal or a mechanical seal.

[0042] The variable frequency speed control motor 103 is a three-phase asynchronous motor with a rated power of 11kW. It is connected to the main shaft 102 through a flexible diaphragm coupling and is equipped with a vector frequency converter to achieve stepless speed regulation and constant torque output from 0 to 6000 rpm.

[0043] The axial loading mechanism 104 is driven by a servo electric cylinder and is installed on the rear end cover of the sealed cavity 101. The push rod of the servo electric cylinder passes through the cavity end cover and directly or through a spring buffer against the back of the stationary ring seat 1011. The thrust is controlled by a servo driver, and the axial load applied to the sealing end face can be continuously adjusted in the range of 0 to 5 kN with a loading accuracy of 1%FS.

[0044] The radial vibration generator 105 uses a piezoelectric vibrator, which is mounted on the bearing seat of the main shaft 102 near the motor side via a bracket. The output push rod of the piezoelectric vibrator is in radial contact with the main shaft 102 through a rolling bearing. By inputting a 0-2000Hz sinusoidal, random, or swept frequency excitation signal through a power amplifier, controllable radial vibration with an amplitude of 0-50m can be generated on the main shaft 102.

[0045] The media circulation system 106 includes a storage tank 1061, a heater 1062, a cooler 1063, a pressure pump 1064, a filter 1065, and a media replacement valve 1066. The storage tank 1061 is a stainless steel sealed container with a volume of 50L. It has a filling port and an exhaust valve at the top and a drain valve at the bottom. The pressure pump 1064 is a variable frequency centrifugal pump. Its suction end is connected to the storage tank 1061 through a pipeline, and its output end is connected to the media inlet of the sealed cavity 101 after passing through the media replacement valve 1066. The cavity pressure can be controlled within the range of 0 to 2.5MPa by adjusting the pump speed. The heater 1062 is a flange-type electric heating tube installed between the sealed cavity 101 and the storage tank. Between 1061, in conjunction with a PT100 temperature sensor and a PID controller, the medium temperature can be raised from room temperature to 180℃. Cooler 1063 is a shell-and-tube water cooler, installed in series with heater 1062. Active cooling of the medium temperature is achieved by adjusting the cooling water flow rate. Filter 1065 is installed at the suction end of pressure pump 1064 and adopts a replaceable filter element structure. The cleanliness of the medium can be adjusted by replacing filter elements of different precision (5m, 20m, 50m). Medium replacement valve 1066 is a three-way solenoid valve. One end is connected to pressure pump 1064, and the other two ends are connected to two storage tanks of different media, respectively, to achieve rapid switching between clean water, oil or slurry containing particles.

[0046] The environmental parameter adjustment module 2 consists of multiple independent PID controllers and signal conversion units. The signals are connected to the various actuators (frequency converter, servo driver, power amplifier, pressure pump frequency converter, heater, cooling water regulating valve, medium replacement valve, etc.) of the multi-scenario simulation test bench 1. This module receives the scenario parameter set values ​​issued by the control module 4, outputs the corresponding analog quantity or pulse signal to drive the actuator, and provides real-time feedback of the actual value to form a local closed loop. The environmental parameter adjustment module 2 is used to independently adjust the spindle speed, axial load, amplitude and frequency of radial vibration, medium pressure, medium temperature, medium type and medium cleanliness according to the preset scenario sequence to simulate multiple different operating scenarios.

[0047] The leakage detection module 3 is installed on the multi-scenario simulation test bench 1 and includes a weighing sensor 301, a torque sensor 302 and an acoustic emission sensor 303.

[0048] The weighing sensor 301 is a cantilever beam strain sensor with a range of 1 kg and an accuracy of 0.01 g. It is installed at the bottom of the leakage liquid collection container below the leakage port of the sealed cavity 101. The real-time leakage rate (mL / h) is obtained by continuously weighing the change in mass of the leakage liquid and calculating the differential.

[0049] The torque sensor 302 is a strain-type rotary torque sensor with a range of 200 Nm and an accuracy of 0.1% FS. It is installed between the coupling of the spindle 102 and the variable frequency speed control motor 103 and adopts non-contact signal transmission to measure the friction torque of the sealing end face in real time.

[0050] The acoustic emission sensor 303 is a piezoelectric ceramic sensor with a resonant frequency of 150kHz. It is attached to the outer wall of the stationary ring seat 1011 by vacuum silicone grease coupling to capture AE signals (such as friction, tearing, dry grinding, etc.) generated by micro-contact of the sealed end face. The sampling frequency is 2MHz.

[0051] The control module 4 adopts a PC-based soft PLC (such as Beckhoff TwinCAT), and is electrically connected to the environmental parameter adjustment module 2, the frequency converter of the variable frequency speed control motor 103, the servo driver of the axial loading mechanism 104, the excitation signal source (signal generator) of the radial vibration generator 105, the frequency converter of the pressure pump 1064 of the medium circulation system 106, the solid-state relay of the heater 1062, the electric regulating valve of the cooler 1063, and the medium replacement valve 1066 via the EtherCAT fieldbus, forming a high-speed real-time closed-loop control loop with a control cycle of 1ms. At the same time, the control module 4 is electrically connected to the weighing sensor 301 (via amplifier), the torque sensor 302 (via transmitter), and the acoustic emission sensor 303 (via preamplifier) ​​via the NI USB-6363 high-speed data acquisition card, and acquires detection data at a synchronous sampling rate of 1kHz and stores it to the local SSD hard disk.

[0052] The lifetime prediction module 5 is a software program (such as a prediction engine developed based on MATLAB Runtime or Python) running on the same PC. It is connected to the control module 4 through the OPC UA interface, reads the detection data in real time and calls the built-in algorithm. The lifetime prediction module 5 has a built-in algorithm based on performance degradation data fitting, which is used to comprehensively predict the remaining lifetime based on the degradation trajectory of leakage rate and other auxiliary signals collected in multiple scenarios.

[0053] Example 2: Definition of running scenario sequence; In this embodiment, for a mechanical seal (end face material: silicon carbide to silicon carbide, auxiliary seal: fluororubber) of model CM1Bn-40 used in petrochemical process pumps, a sequence of 5 scenarios is predefined in the human-machine interface of control module 4, with a total simulation time of 48 hours, to accelerate the simulation of 6 months of actual operation in the field. The settings of each scenario are as follows: Scenario S1 (steady-state operation, duration 24h): Simulates rated operating conditions, medium is clean water, temperature is 80℃, pressure is 1.0MPa, speed is 2900rpm, axial load is the design value (end face specific pressure 1.2MPa), radial vibration amplitude is 0m. This scenario is used to obtain the baseline degradation rate of the seal under normal steady-state operating conditions.

[0054] Scenario S2 (Cryogenic Start-up / Shutdown Transient, 10 min each time, repeated every 2 h, cumulative equivalent time 2 h): Simulates the start-up and shutdown process, with the medium temperature linearly increasing from 20℃ to 80℃, the speed linearly increasing from 0 to 2900 rpm, and the pressure linearly increasing from 0 to 1.0 MPa. Other parameters are the same as in steady state. This scenario is used to capture the peak leakage rate during the start-up and shutdown process.

[0055] Scenario S3 (Pressure shock, lasting 2 hours): Based on steady-state conditions, a pressure step is performed every 15 minutes: the medium pressure is increased from 1.0 MPa to 2.5 MPa within 3 seconds, held for 10 seconds, and then reduced to 1.0 MPa within 3 seconds. This scenario simulates the pressure shock caused by water hammer in pipelines or rapid opening and closing of valves.

[0056] Scenario S4 (containing solid particulate media, lasting 4 hours): The medium is switched to oil slurry containing 5wt% quartz sand (particle size 20-50m) through the medium replacement valve 1066. The pressure is reduced to 0.8MPa and the rotation speed is reduced to 2000rpm. Other parameters are the same as in steady state. This scenario simulates the working condition of medium contamination.

[0057] Scenario S5 (Accelerated Degradation, lasting 16 hours): Simultaneously, multiple stress acceleration measures are taken to increase the rotational speed to 3500 rpm, the medium temperature to 120°C, the axial load to 20%, and apply radial vibration with an amplitude of 20 m and a frequency of 800 Hz. This scenario is used to accelerate the typical wear and fatigue failure modes of the seal.

[0058] The above-mentioned multiple different operating scenarios cover low temperature and low pressure slow start-up scenario (S2), high temperature and high pressure high speed steady state scenario (S1), instantaneous pressure impact scenario (S3), operating scenario under medium containing solid particles (S4), and radial vibration scenario with preset amplitude and frequency (S5), comprehensively simulating the typical working conditions that mechanical seals may experience in actual service.

[0059] Example 3: Detection and Lifetime Prediction Method Steps; Please combine Figure 4 In this embodiment, the above system is used to periodically inspect three sets of mechanical seals from the same batch (three inspections are performed at 0, 24, and 48 hours respectively) to predict the remaining life. The specific steps are as follows: Step 1: Installation. Install the dynamic ring of the mechanical seal to be tested on the main shaft 102 and the stationary ring on the stationary ring seat 1011. Tighten the cover bolts according to the torque specified in the product manual. Inject clean water into the liquid storage tank 1061 and start the medium circulation system 106 to vent and pre-lubricate.

[0060] Step 2: Load the running scenario sequence. The operator selects the preset scenario sequence (S1S2S3S4S5) on the touch screen of the control module 4, sets the total number of cycles to 1 (48 hours in total), and clicks the start button. The control module 4 automatically executes each scenario in sequence. The parameter change trajectory (such as the temperature rise curve, the acceleration curve, and the pressure step waveform) in each scenario is strictly implemented by closed-loop control according to the preset values.

[0061] Step 3: Data Acquisition. During the entire 48-hour test, the leakage detection module 3 continuously acquires and records the following data at a frequency of 1Hz: leakage rate Q(t) (mL / h), friction torque M(t) (Nm), and acoustic emission signal RMS value AE(t) (V). At the same time, the control module 4 synchronously records the actual values ​​of each environmental parameter. All data is stored in a CSV file with a timestamp.

[0062] Step 4: Extract performance degradation indicators. After the test, the lifetime prediction module 5 automatically reads the data file, segments it according to the scenario-based timestamps, and extracts the following multi-dimensional degradation indicators: For steady-state scenario S1: take the average leakage rate of the last 2 hours (22h~24h) as the steady-state leakage rate I1. In this example, I1=2.5 mL / h when detected at the 48th hour.

[0063] For transient scenario S2: extract the maximum value of the leakage rate during each startup process, and take the average value of all repetitions as the transient peak leakage rate I2. In this example, I2 = 8.0 mL / h when detected at the 48th hour.

[0064] For accelerated degradation scenario S5: Perform a piecewise linear regression on the friction torque M(t) hourly to calculate the torque increase rate (slope) per hour, and take the average increase rate of the last 4 hours as I3. In this example, I3 = 0.2 Nm / h.

[0065] (Using the same method, extract the corresponding index values ​​for hour 0 and hour 24: hour 0: I1 = 0.1 mL / h, I2 = 0.5 mL / h, I3 = 0.01 Nm / h; hour 24: I1 = 1.2 mL / h, I2 = 4.0 mL / h, I3 = 0.08 Nm / h.) Step 5: Calculate the overall damage degree. The control module 4 pre-stores the failure thresholds of this type of seal in various scenarios: steady-state scenario T1=10 mL / h (leakage rate), transient scenario T2=15 mL / h (peak leakage rate), and accelerated degradation scenario T3=0.5 Nm / h (torque rise rate). Based on the actual application conditions of this seal in petrochemical process pumps, the occurrence frequency and hazard level of each scenario are obtained by statistically analyzing the field operation data. Weighting coefficients are set: steady-state scenario w1=0.6, transient scenario w2=0.1, and accelerated degradation scenario w3=0.3 (satisfying wi=1). The current overall damage degree is calculated using the formula D =(wi(Ii / Ti)).

[0066] Taking the test data from the 48th hour as an example: D48 = 0.6(2.5 / 10) + 0.1(8.0 / 15) + 0.3(0.2 / 0.5) = 0.60.25 + 0.10.533 +0.30.4 = 0.15 + 0.0533 + 0.12 = 0.3233.

[0067] Similarly, we can calculate that: D0 = 0.01 (initial new seal), D24 = 0.15, the comprehensive damage degree D ranges from 0 to 1, and the seal is judged to be in failure when D1.0.

[0068] Step 6: Predict remaining lifetime. The lifetime prediction module 5 extracts the three detection times t=[0, 24, 48] hours and the corresponding D value sequence [0.01, 0.15, 0.3233]. An exponential fitting model D(t)=ae^(bt) is used for nonlinear least squares fitting (the initial value at t=0 is removed to reduce fitting error; in practice, the parameters are determined using t=24 and 48). The fitting parameters are calculated as a=0.085 and b=0.028. Solving the equation 0.085e^(0.028t) = 1.0, the predicted total lifetime t_total= ln(1 / 0.085) / 0.02888.2 hours is obtained. The system has been running for 48 hours, therefore the predicted remaining lifetime is 88.2 - 48 = 40.2 hours. This result is displayed through the human-machine interface and a report is generated. It can also be transmitted via email or OPC. UA pushes the message to the upper-level equipment management system, instructing maintenance personnel to arrange seal replacement 88 hours in advance.

[0069] If the seal continues to operate, steps 5 to 6 above can be repeated every 24 hours. After each test, the fitting parameters and remaining life prediction values ​​are dynamically updated. As the D value sequence increases, the prediction accuracy gradually improves.

[0070] Example 4: Examples of weight coefficient adjustment in various scenarios; In different application scenarios, the weighting coefficients can be adjusted according to the actual working conditions, for example: For cooling water pumps that are frequently started and stopped (dozens of times a day), transient scenarios contribute the most to seal damage, and the weights can be adjusted to: w1=0.3, w2=0.6, w3=0.1.

[0071] For large compressors that operate continuously (starting and stopping only a few times a year), steady-state wear is dominant, and the cleanliness of the medium is relatively high. The weights can be adjusted to: w1=0.8, w2=0.05, w3=0.15.

[0072] For mine slurry pumps with strong vibrations and sand-containing media, the accelerated degradation scenario should have the highest weight, which can be set as: w1=0.2, w2=0.1, w3=0.7.

[0073] Operators only need to modify the weight coefficients of each scenario in the configuration interface of control module 4, and the system will automatically recalculate the comprehensive damage degree D and the remaining life prediction value without changing the test process or hardware configuration, which demonstrates the good engineering adaptability of the present invention.

[0074] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Content not described in detail in this specification is prior art known to those skilled in the art.

[0075] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Those skilled in the art can readily implement the present invention based on the accompanying drawings and the above description. However, any modifications, alterations, or variations made by those skilled in the art without departing from the scope of the present invention, utilizing the disclosed technical content, are equivalent embodiments of the present invention. Furthermore, any equivalent changes, alterations, or variations made to the above embodiments based on the essential technology of the present invention are still within the protection scope of the present invention.

Claims

1. A mechanical seal leakage detection and life prediction system, comprising a base (6), a multi-scenario simulation test bench (1), an environmental parameter adjustment module (2), a leakage detection module (3), a control module (4), and a life prediction module (5), characterized in that, The multi-scenario simulation test bench (1) is mounted on the base (6) and includes a sealed cavity (101), a main shaft (102), a variable frequency speed control motor (103), an axial loading mechanism (104), a radial vibration generator (105), and a media circulation system (106). The sealed cavity (101) is equipped with a stationary ring seat (1011) for fixing the stationary ring. The main shaft (102) passes through the sealed cavity (101) and is used to mount the rotating ring to drive its rotation; The variable frequency speed control motor (103) is connected to the main shaft (102) via a coupling; The axial loading mechanism (104) is used to adjust the axial load applied to the sealing end face; The radial vibration generator (105) is used to apply controllable radial vibration to the spindle (102); The medium circulation system (106) is connected to the sealed cavity (101) and is used to independently adjust the medium pressure, medium temperature, medium type and medium cleanliness in the sealed cavity; The environmental parameter adjustment module (2) is connected to the multi-scenario simulation test bench (1) to independently adjust the spindle speed, axial load, amplitude and frequency of radial vibration, medium pressure, medium temperature, medium type and medium cleanliness according to the preset scenario sequence, so as to simulate multiple different operating scenarios. The leakage detection module (3) is installed on the multi-scenario simulation test bench (1) and is used to collect the leakage rate, friction torque and sealing cavity temperature signals of the mechanical seal under test in real time under various operating scenarios. The control module (4) is electrically connected to the environmental parameter adjustment module (2) and the leakage detection module (3) respectively, and is used to control the execution order and duration of the scene sequence, and to receive and store detection data; The lifetime prediction module (5) is connected to the control module (4). The lifetime prediction module (5) has a built-in algorithm based on performance degradation data fitting, which is used to comprehensively predict the remaining lifetime based on the degradation trajectory of leakage rate and other auxiliary signals collected in multiple scenarios.

2. The mechanical seal leakage detection and life prediction system according to claim 1, characterized in that, The radial vibration generator (105) is an eccentric cam mechanism or a piezoelectric vibrator. The eccentric cam mechanism drives the eccentric cam to rotate through a servo motor to change the eccentricity. The piezoelectric vibrator inputs an excitation signal through a power amplifier to adjust the vibration amplitude and frequency.

3. The mechanical seal leakage detection and life prediction system according to claim 1, characterized in that, The media circulation system (106) includes a storage tank (1061), a heater (1062), a cooler (1063), a pressure pump (1064), a filter (1065), and a media replacement valve (1066). The storage tank (1061) is used to store the test medium; The suction end of the pressure pump (1064) is connected to the liquid storage tank (1061), and the output end is connected to the medium inlet of the sealed cavity (101) for adjusting the medium pressure in the sealed cavity; The heater (1062) and cooler (1063) are connected in series on the circulation pipeline between the sealed cavity (101) and the storage tank (1061) to regulate the temperature of the medium; The filter (1065) is installed at the suction end of the pressure pump (1064) or the outlet pipeline of the sealed cavity (101) to adjust the cleanliness of the medium. The medium replacement valve (1066) is installed on the liquid supply pipeline between the liquid storage tank (1061) and the sealed cavity (101) for switching different types of media.

4. The mechanical seal leakage detection and life prediction system according to claim 1, characterized in that, The leakage detection module (3) includes a weighing sensor (301), a torque sensor (302), and an acoustic emission sensor (303). The weighing sensor (301) is installed below the leakage liquid collection container of the sealed cavity (101) or on the leakage liquid discharge pipeline to measure the leakage rate; The torque sensor (302) is installed between the main shaft (102) and the coupling of the variable frequency speed control motor (103) for measuring friction torque; The acoustic emission sensor (303) is attached to the outer wall of the stationary ring or the sealed cavity (101) by means of a coupling agent, and is used to capture the contact status signal of the sealed end face.

5. The mechanical seal leakage detection and life prediction system according to claim 1, characterized in that, The control module (4) is electrically connected to the environmental parameter adjustment module (2), the frequency converter of the variable frequency speed control motor (103), the servo driver of the axial loading mechanism (104), the excitation signal source of the radial vibration generator (105), the pressure pump (1064), heater (1062), cooler (1063) and medium replacement valve (1066) of the medium circulation system (106) via fieldbus to form a closed-loop control circuit; the control module (4) is also electrically connected to the weighing sensor (301), torque sensor (302) and acoustic emission sensor (303) of the leakage detection module (3) via data acquisition card.

6. The mechanical seal leakage detection and life prediction system according to claim 1, characterized in that, The various operating scenarios include at least one combination of the following: low temperature and low pressure slow start-up scenario, high temperature and high pressure high speed steady state scenario, instantaneous pressure impact scenario, instantaneous speed increase and decrease scenario, dry friction start-up scenario, operating scenario under medium containing solid particles, radial vibration scenario with preset amplitude and frequency, and shutdown cooling scenario.

7. A method for detecting and predicting the lifespan of mechanical seals, comprising using the mechanical seal leakage detection and lifespan prediction system as described in any one of claims 1 to 6, characterized in that, Includes the following steps: Step 1: Install the mechanical seal to be tested onto the multi-scenario simulation test bench (1); Step 2: Load a predefined sequence of running scenarios through the control module (4). The sequence contains at least three different running scenarios in sequence, each scenario having a preset duration and parameter change trajectory. Step 3: During the execution of the running scenario sequence, the leakage rate, friction torque and sealing cavity temperature are continuously collected by the leakage detection module (3) and recorded as time series data by the control module (4); Step 4: Segment the leakage rate data for each operating scenario and extract the steady-state leakage rate, transient peak leakage rate and friction torque rise rate for each scenario as performance degradation indicators in multiple dimensions. Step 5, the life prediction module (5) calculates the current comprehensive damage degree by comparing the performance degradation indicators of multiple dimensions extracted in different scenarios with the corresponding failure thresholds in a weighted summation method according to the pre-stored failure threshold parameter set corresponding to each operating scenario. as well as Step 6: Based on the comprehensive damage sequence obtained from previous tests, use regression analysis or exponential fitting extrapolation to predict the time point when the comprehensive damage reaches the preset failure threshold, which is then used as the remaining lifespan.

8. The method for detecting leakage and predicting lifespan of mechanical seals according to claim 7, characterized in that, The predefined sequence of operating scenarios includes at least: a steady-state scenario for simulating normal operation, a transient scenario for simulating start-stop process, and an accelerated degradation scenario for simulating abnormal operating conditions.

9. The method for detecting leakage and predicting lifespan of mechanical seals according to claim 7, characterized in that, The weighting coefficients for each scenario used in the weighted summation method are preset based on the frequency of occurrence or hazard level of the scenario in actual application conditions.

10. The method for detecting leakage and predicting lifespan of mechanical seals according to claim 7, characterized in that, The overall damage degree D in step 5 is calculated using the following formula: D = (wi (Ii / Ti)) Wherein, Ii is the performance degradation index extracted under the i-th running scenario, Ti is the failure threshold corresponding to the scenario that is stored in advance, wi is the weight coefficient of the i-th scenario, and wi = 1; the value range of the comprehensive damage degree D is 0 to 1, and the seal is judged to fail when D reaches or exceeds 1.0; the life prediction module (5) is based on the D value sequence obtained from each detection, and uses the exponential fitting model D(t)=ae^(bt) to fit, where a and b are fitting parameters, t is the running time, and the predicted total life is obtained by solving D(t)=1.0, and then the remaining life is obtained by subtracting the running time.

Images