A digital twin-driven clutch performance monitoring and evaluation system

Abstract

The present invention discloses a clutch performance monitoring and evaluation system driven by a digital twin, which belongs to the technical field of clutch performance monitoring and evaluation. The system includes: a clutch entity and a clutch twin system; the clutch entity is used to collect actual operation data, and transmit the actual operation data to the clutch twin system as test data of the clutch twin system; the clutch twin system is used to monitor and evaluate the clutch performance based on the test data to obtain an evaluation result. Through the cross-platform collaboration and communication between the clutch entity and multiple software systems, the full life cycle working characteristics of the clutch entity are obtained, the degree of degradation of the clutch working characteristics is quantified, monitored and evaluated, the clutch performance test efficiency is improved, and the clutch test cost is reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of clutch performance monitoring and evaluation, and specifically relates to a clutch performance monitoring and evaluation system driven by a digital twin. Background Art

[0002] The wet shift clutch is one of the core components of the integrated transmission device. Its performance directly affects the transmission efficiency and working reliability of the transmission device. The gradual degradation of the clutch friction characteristics leads to the gradual change of its dynamic response and thermal load characteristics, which causes the unstable working quality and failure of the clutch. Therefore, the monitoring and evaluation of the friction characteristic degradation is the key issue for the wet shift clutch to achieve high reliability operation and healthy service.

[0003] At present, the commonly used performance prediction method is to use historical data, numerical models and expert experience to predict the degradation trend of performance and parameters. However, the built-in monitoring sensors of wet shift clutches are limited, and it is difficult to collect and extract effective data; the numerical model cannot perform real-time optimization and feedback based on the actual wear characteristics of the friction elements; in addition, the evaluation of degradation characteristics based on expert experience is not real-time. In recent years, the rapid development of digital twin technology has provided a new idea for solving the above problems. The use of digital twin technology can be used to dynamically monitor the clutch, trace the history, and dynamically control it, which has good application prospects. Summary of the invention

[0004] The present invention aims to solve the deficiencies of the prior art and proposes a clutch performance monitoring and evaluation system driven by digital twins. The clutch digital twin system established by the present invention obtains multivariate clutch performance parameters through a small number of sensors and obtains the real-time performance characteristics of the clutch through an evaluation method, solving the problem of difficult clutch performance monitoring and evaluation in real vehicle state. Through cross-platform collaboration and communication between the clutch entity and multiple software systems, the full life cycle working characteristics of the clutch entity are obtained, the degree of degradation of the clutch working characteristics is quantified, monitored and evaluated, the efficiency of clutch performance testing is improved, and the cost of clutch testing is reduced.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A digital twin-driven clutch performance monitoring and evaluation system, comprising: a clutch entity and a clutch twin system;

[0007] The clutch entity is used to collect actual operation data, and transmit the actual operation data to the clutch twin system as test data of the clutch twin system;

[0008] The clutch twin system is used to monitor and evaluate the clutch performance based on the test data to obtain an evaluation result.

[0009] Further preferably, the clutch twin system comprises: a clutch twin, a prediction module, a logic module, a signal acquisition and software control platform;

[0010] The signal acquisition and software control platform is used to receive test data collected by the clutch entity;

[0011] The clutch twin is used to simulate the clutch entity;

[0012] The prediction module is used to obtain twin data based on the test data;

[0013] The logic module is used to obtain a clutch failure threshold and a health index based on the twin data, and to evaluate the clutch degradation process based on the clutch failure threshold and the health index to obtain the evaluation result.

[0014] Further preferably, the twin data includes: friction coefficient, friction torque, rotation speed, engagement time and temperature;

[0015] The prediction module includes: a dynamic simulation layer and a data quantization layer;

[0016] The dynamic simulation layer is used to obtain the twin data based on the test data;

[0017] The data quantization layer is used to display the real-time change characteristics of the twin data.

[0018] Further preferably, the logic module comprises: a feature extraction layer and a data evaluation layer;

[0019] The feature extraction layer is used to construct a performance degradation model based on the twin data, and update and reconstruct the performance degradation model to obtain a weight coefficient;

[0020] The data evaluation layer is used to obtain the failure threshold and the health index of the clutch based on the weight coefficient, and evaluate the clutch state based on the failure threshold and the health index to obtain the evaluation result.

[0021] Further preferably, the performance degradation model includes:

[0022]

[0023] Where, L i,t represents the degradation data of clutch i at time t; η represents the functional form of the degradation model; represents the random effect of the degradation model of clutch i; ε i,t represents the random noise of clutch i at time t.

[0024] Further preferably, the conditions satisfied by the weight coefficient include:

[0025]

[0026] In the formula, ω i,j Represents the weight coefficient; n j represents the number of degradation data collected from the operating clutch j; represents the degree of degradation obtained using the update method; represents the degree of degradation obtained using the reconstruction method; It means that the reconstructed degradation trajectory is the weighted average of the historical failure trajectories.

[0027] Further preferably, the health index includes:

[0028] d j =X i,j ω i,j ,

[0029] Where, d j Indicates health index; X i,j represents the value of twin data i at monitoring time j; ω i,j Represents the weight coefficient.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] 1. The present invention can import a small amount of sensor test data in the clutch entity into the clutch twin system to obtain more twin data on the clutch performance evolution trend, complete the clutch monitoring and evaluation work, and provide situation-based maintenance strategies. Not only can it obtain performance parameters that cannot be measured by clutch testing, but it can also significantly save test costs and provide technical and methodological support for the digital development of clutches.

[0032] 2. The clutch performance prediction model proposed in the present invention can quantify the degradation characteristics of the clutch friction characteristics.

[0033] 3. The clutch performance logic model proposed in the present invention obtains the clutch health index and failure threshold by constructing a clutch degradation model and a degradation weight coefficient.

[0034] 4. The clutch digital twin system established by the present invention obtains the change pattern of the engagement time, friction torque, surface temperature and friction coefficient of the clutch entity throughout its life cycle through cross-platform collaboration and communication between the clutch entity and multiple software systems, thereby quantifying, monitoring and evaluating the performance characteristics of the clutch. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0036] Figure 1 This is a schematic diagram of the structure of a clutch performance monitoring and evaluation system driven by a digital twin according to an embodiment of the present invention;

[0037] Figure 2 This is a schematic diagram of the physical structure of a clutch according to an embodiment of the present invention;

[0038] Figure 3 This is a schematic diagram of the prediction module structure according to an embodiment of the present invention;

[0039] Figure 4 This is a schematic diagram of the workflow of the dynamic simulation layer according to an embodiment of the present invention;

[0040] Figure 5 This is a schematic diagram of the logic module structure of an embodiment of the present invention;

[0041] Figure 6 It is a signal acquisition and software control platform for the embodiment of the present invention. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0044] Embodiment 1:

[0045] This embodiment provides a clutch performance monitoring and evaluation system driven by a digital twin. Figure 1 As shown, the system includes: a clutch entity and a clutch twin system; the clutch entity is used to collect actual operating data and transmit the actual operating data to the clutch twin system as test data for the clutch twin system; the clutch twin system is used to monitor and evaluate the clutch performance based on the test data to obtain evaluation results.

[0046] The system also includes a twin database for storing and calling data generated by the clutch entity and the clutch twin system. Specifically including: clutch entity input parameters and sensor test parameters, clutch twin system twin data; at the same time, it also includes real-time working data and historical storage data of the clutch entity and the twin system. In this embodiment, the clutch input parameters include: structural parameters, material parameters, oil parameters, working parameters; sensor test data include: speed and torque of the active and passive ends of the clutch, oil pressure of the oil supply system control oil circuit, oil temperature and flow of the lubrication oil circuit and oil outlet of the oil supply system; twin data includes the engagement time, friction torque, surface temperature and friction coefficient of the clutch twin.

[0047] Specifically, Figure 2 As shown in the figure, the clutch entity includes: drive motor, speed torque sensor, clutch box, clutch oil supply system, load motor, oil flow, temperature and pressure sensor, etc. The drive motor provides drive speed for the clutch active section; the load motor provides load torque for the clutch passive end. The oil supply system has two oil inlet lines, one is the pressure oil line, which fills or drains the clutch piston to complete the engagement and separation of the clutch, and is equipped with an oil pressure sensor; the other is the lubrication oil line, which cools and lubricates the clutch friction element, and is equipped with oil flow and temperature sensors; the oil outlet of the clutch box returns the oil to the oil supply system, and is equipped with oil flow and temperature sensors. Speed ​​and torque sensors are arranged at both the active and passive ends of the clutch to collect the speed and torque of the active and passive ends of the clutch. The above collected test data is transmitted to the signal acquisition and software control platform in the form of voltage, and converted into actual characteristic parameter values ​​and stored in the twin database.

[0048] The clutch twin system includes: clutch twin, prediction module, logic module, signal acquisition and software control platform; the signal acquisition and software control platform is used to receive the test data collected by the clutch entity. LabVIEW is used to design the signal acquisition and control platform, and the TCP / IP communication protocol is used to complete the acquisition and storage of input parameters, sensor test data, and twin data in the clutch entity and twin system. The clutch twin is used to simulate the clutch entity, import the input parameters into the clutch twin system, and use Pro / E and ADAMS to complete the three-dimensional modeling and dynamic evolution of the clutch twin. The prediction module is used to obtain twin data based on test data. The logic module is used to obtain the clutch health index based on the twin data, and evaluate the clutch degradation process based on the clutch health index to obtain the evaluation results.

[0049] Among them, Figure 3As shown, the prediction module includes: a dynamic simulation layer and a data quantization layer; the dynamic simulation layer is used to obtain twin data based on test data; the data quantization layer is used to display the real-time change characteristics of the twin data.

[0050] Specifically, the dynamic simulation layer includes two prediction models, namely, the asperity contact and wear model and the friction element thermal-fluid-solid coupling numerical model, such as Figure 4 First, the clutch entity input parameters are input into the asperity contact and wear model to obtain the friction coefficient; then the friction coefficient is introduced into the friction element thermal-fluid-solid coupling numerical model to solve the changes in friction torque, speed, temperature and engagement time.

[0051] The specific working process of the micro-asperity contact and wear model includes: first, inputting material parameters and working parameters; quantifying the actual morphological features of the friction element through the micro-asperity height features. The fast Fourier transform algorithm (FFT) is used to solve the elastic-plastic deformation of the surface, without distinguishing the fluid lubrication area and the micro-asperity contact area of ​​the friction pair. The local pressure is solved by the load balance equation, and the pressure continuity condition is automatically satisfied during the solution process. The moving point heat source integral method is used to solve the surface temperature rise, and finally the friction coefficient and wear amount are obtained, thereby updating the actual morphological features.

[0052] The specific working process of the friction element thermal-fluid-solid coupling numerical model includes: according to the input working parameters, material parameters, and friction coefficient, the Reynolds equation and the load balance equation are jointly solved to obtain the friction torque. Then, the axial motion equation of the friction pair and the heat conduction equation are combined to obtain the speed, temperature, and engagement time changes within a bonding cycle.

[0053] like Figure 5 As shown, the logic module includes: a feature extraction layer and a data evaluation layer; the feature extraction layer is used to build a performance degradation model based on twin data, and update and reconstruct the performance degradation model to obtain a weight coefficient; the data evaluation layer is used to obtain a failure threshold of the clutch based on the weight coefficient, and build a health index based on the failure threshold.

[0054] In the feature extraction layer, the twin data is called to form a clutch performance degradation model:

[0055]

[0056] Where, L i,t represents the degradation data of clutch i at time t; η represents the functional form of the degradation model; represents the random effect of the degradation model of clutch i; ε i,t represents the random noise of clutch i at time t.

[0057] Then, the weighted average method is used to reconstruct the above performance degradation model to obtain a reconstructed model, including:

[0058]

[0059] In the formula, represents the reconstructed degradation model of the running clutch j at time t, represents the state monitoring data collected from the initial state to the current time t; w i，j ≥0 indicates weight coefficient, and there is represents the estimated degradation model parameters, and m represents the number of historically failed clutches.

[0060] Afterwards, the Bayesian method is used to update the above performance degradation model to obtain an updated model, including:

[0061]

[0062] In the formula, represents the degradation state of the running clutch j at time t obtained by using the Bayesian updating method, L j，. represents the state monitoring data collected from the initial state to the current time t, represents the degradation model parameters updated using the Bayesian updating method, that is, the random effects updated at time t.

[0063] In order to maximize the fit between the above reconstruction model and the updated model, the weight coefficient needs to satisfy the following formula:

[0064]

[0065] In the formula, ω i,j Represents the weight coefficient; n j represents the number of degradation data collected from the operating clutch j; represents the degree of degradation obtained using the update method; represents the degree of degradation obtained using the reconstruction method; It means that the reconstructed degradation trajectory is the weighted average of the historical failure trajectories.

[0066] At the data evaluation layer, the failure threshold of the in-service clutch is estimated and the health index is constructed.

[0067] After obtaining the weight coefficient, the failure threshold of the running clutch j can be estimated, and the formula is:

[0068]

[0069] In the formula, ω i,j≥0 holds for i=1, 2, ..., m, and represents the degradation data of the historically failed clutch i observed at the last failure moment.

[0070] Health indicators include:

[0071] d j =X i,j ω i,j ,

[0072] Where, d j Indicates health index; X i,j represents the value of twin data i at monitoring time j; ω i,j Represents the weight coefficient.

[0073] If the health index d j Exceeding the failure threshold E[D j ], the clutch enters a failure state and a situation-based maintenance strategy needs to be formulated immediately.

[0074] The signal acquisition and software control platform is used for data collection, storage and monitoring of the clutch entity and twin system operation, and can display parameter changes in real time, extract and monitor characteristic parameters, etc. Figure 6 As shown in the figure, it includes input parameter module, environmental parameter module, sensor data module, clutch twin visualization module, twin data module, and situation-based maintenance module. The clutch input parameters can be input in the input parameter module; the environmental parameters of the clutch, including ambient temperature, air pressure, humidity, etc., can be input in the environmental parameter module; the dynamic changes of sensor data can be seen in the sensor data module; the clutch twin visualization module can watch the three-dimensional dynamic process of clutch engagement and separation; the twin data module can monitor the dynamic changes of twin data in real time; in the situation-based maintenance module, the health index of the twin data is evaluated in real time. If it exceeds the failure threshold, this module will display the alarm function and give the corresponding maintenance strategy, specifically: considering the cost of clutch monitoring data extraction and analysis, formulate a situation-based maintenance strategy, including the minimum maintenance cost strategy and the maximum availability strategy.

[0075] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A digital twin driven clutch performance monitoring and evaluation system, characterized in that: include: Clutch entity and clutch twin system; The clutch entity is used to collect actual operation data, and transmit the actual operation data to the clutch twin system as test data of the clutch twin system; The clutch twin system includes: a clutch twin, a prediction module, a logic module, a signal acquisition and software control platform; the signal acquisition and software control platform is used to receive test data collected by the clutch entity; the clutch twin is used to simulate the clutch entity; the prediction module is used to obtain twin data based on the test data, and the twin data includes: friction coefficient, friction torque, rotation speed, engagement time and temperature; the logic module is used to obtain a clutch failure threshold and a health index based on the twin data, and evaluate the clutch degradation process based on the clutch failure threshold and the health index to obtain the evaluation result; Among them, the prediction module includes: a dynamic simulation layer and a data quantization layer; the dynamic simulation layer is used to obtain the twin data based on the test data; the data quantization layer is used to display the real-time change characteristics of the twin data; specifically, the dynamic simulation layer includes two prediction models, namely the micro-protrusion contact and wear model and the friction element thermal-fluid-solid coupling numerical model; first, the clutch entity input parameters are input into the micro-protrusion contact and wear model to obtain the friction coefficient; then the friction coefficient is imported into the friction element thermal-fluid-solid coupling numerical model to solve the changes in friction torque, speed, temperature and engagement time; the specific working process of the micro-protrusion contact and wear model includes: first, inputting material parameters and working parameters; the actual morphological characteristics of the friction element The micro-convex height characteristics are quantified, and the fast Fourier transform algorithm is used to solve the elastic-plastic deformation of the surface. The fluid lubrication area and the micro-convex contact area of ​​the friction pair are not distinguished. The local pressure is solved by the load balance equation. The pressure continuity condition is automatically satisfied during the solution process. The moving point heat source integral method is used to solve the surface temperature rise, and finally the friction coefficient and wear amount are obtained, thereby updating the actual morphological characteristics; the specific working process of the thermal-fluid-solid coupling numerical model of the friction element includes: according to the input working parameters, material parameters, and friction coefficient, the Reynolds equation and the load balance equation are jointly solved to obtain the friction torque; then, the axial motion equation of the friction pair and the heat conduction equation are combined to obtain the changes in speed, temperature, and engagement time within a bonding cycle; The logic module includes: a feature extraction layer and a data evaluation layer; the feature extraction layer is used to construct a performance degradation model based on the twin data, and update and reconstruct the performance degradation model to obtain a weight coefficient; the data evaluation layer is used to obtain the failure threshold and the health index of the clutch based on the weight coefficient, and evaluate the clutch state based on the failure threshold and the health index to obtain the evaluation result; the performance degradation model includes: Where, L i,t represents the degradation data of clutch i at time t; η represents the functional form of the degradation model; represents the random effect of the degradation model of clutch i; ε i,t represents the random noise of clutch i at time t; then the weighted average method is used to reconstruct the above performance degradation model to obtain the reconstructed model, including: In the formula, represents the reconstructed degradation model of the running clutch j at time t, represents the state monitoring data collected from the initial state to the current time t; ω i，j ≥0 indicates weight coefficient, and there is represents the estimated degradation model parameters, and m represents the number of historically failed clutches; then, the Bayesian method is used to update the above performance degradation model to obtain an updated model, including: In the formula, represents the degradation state of the running clutch j at time t obtained by using the Bayesian updating method, L j，. represents the state monitoring data collected from the initial state to the current time t, represents the degradation model parameters updated using the Bayesian updating method, that is, the random effects updated at time t; The conditions satisfied by the weight coefficient include: In the formula, ω i,j Represents the weight coefficient; n j represents the number of degradation data collected from the operating clutch j; represents the degree of degradation obtained using the update method; represents the degree of degradation obtained using the reconstruction method; It means that the reconstructed degradation trajectory is the weighted average of the historical failure trajectory. After obtaining the weight coefficient, the failure threshold of the running clutch j can be estimated, and the formula is: In the formula, ω i,j ≥0 holds for i=1, 2, ..., m, and represents the degradation data of the historically failed clutch i observed at the last failure moment; the health index includes: d j =X i,j ω i,j , where d j Indicates health index; X i,j represents the value of twin data i at monitoring time j; ω i,j Represents the weight coefficient; if the health index d j Exceeding the failure threshold E[D j ], the clutch enters a failure state and a situation-based maintenance strategy needs to be formulated immediately; The clutch twin system is used to monitor and evaluate the clutch performance based on the test data to obtain an evaluation result.

Images