A servo feeding system fault diagnosis method and device and electronic equipment

Abstract

The application provides a servo feeding system fault diagnosis method, device and electronic equipment, wherein the servo feeding system fault diagnosis method comprises the following steps: constructing a high-fidelity dynamic model for a target feeding system; collecting vibration signals of different faults under different working conditions and constructing a multi-working-condition fault database; performing a fault simulation experiment by using the high-fidelity dynamic model, acquiring simulation signals, and comparing the simulation signals with the vibration signals to optimize and verify the high-fidelity dynamic model; collecting actual vibration data of the target feeding system, comparing the actual vibration data with twin data, and determining a fault type of the target feeding system. Through the application, the precision fault diagnosis and prediction of the ball screw feeding system are realized, and the problems of poor precision and real-time performance in the prior art are solved.

Description

A method, apparatus and electronic device for servo feed system fault diagnosis Technical Field

[0001] This invention relates to the field of CNC machine tool technology, and in particular to a method, device and electronic equipment for servo feed system fault diagnosis. Background Technology

[0002] CNC machine tools, short for numerical control machine tools, are automated machine tools equipped with a program control system. This control system logically processes programs with control codes or other symbolic instructions, decodes them, represents them with coded numbers, and inputs them into the CNC device via an information carrier. After processing, the CNC device sends various control signals to control the machine tool's movements, automatically machining parts according to the shape and dimensions required by the drawings. The feed system is a crucial component of CNC machine tools, enabling the movement of the load.

[0003] With the development of intelligent manufacturing, CNC machine tools are placing increasingly higher demands on the stability and precision of their servo feed systems. As a core transmission component, the health of the ball screw directly affects machining quality. However, existing ball screw fault diagnosis methods mainly rely on manual experience or simple signal processing techniques, resulting in problems such as low diagnostic accuracy, delayed response, and difficulty in predictive maintenance.

[0004] There is currently no effective solution to the problems of poor accuracy and real-time performance in existing related technologies. Summary of the Invention

[0005] This invention provides a fault diagnosis method, device, and electronic device for a servo feed system, which addresses the shortcomings of poor accuracy and real-time performance in existing related technologies and enables intelligent health management.

[0006] In a first aspect, the present invention provides a method for diagnosing faults in a servo feed system, comprising:

[0007] A high-fidelity dynamic model for the target feed system is constructed; the high-fidelity dynamic model includes a ball screw pair dynamic model based on a six-degree-of-freedom differential equation system and a five-degree-of-freedom rolling bearing vibration model;

[0008] Vibration signals of different faults under different working conditions were collected, and a multi-working-condition fault database was constructed.

[0009] The high-fidelity dynamic model was used to conduct a fault simulation experiment, and the simulation signal was obtained and compared with the vibration signal to optimize and verify the high-fidelity dynamic model.

[0010] The actual vibration data of the target feed system is collected, and the actual vibration data is compared with twin data to determine the fault type of the target feed system; the twin data consists of the collected vibration signal and the simulation signal of the optimized high-fidelity dynamic model.

[0011] According to a fault diagnosis method for a servo feed system provided by the present invention, a dynamic model of a ball screw pair for the target feed system is constructed, including:

[0012] Analyze the mechanical properties of the balls in the raceway and construct the dynamic equations of the target feed system.

[0013] Based on the dynamic equations of the target feed system, the coupling relationship between the rotation and linear motion of the ball screw is determined;

[0014] By combining the coupling relationship between the rotation and linear motion of the ball screw, the differential equations of the dynamic behavior of the ball screw pair in the target feed system are constructed, and the dynamic model of the ball screw pair is generated.

[0015] According to the fault diagnosis method for a servo feed system provided by the present invention, the mechanical properties of the balls in the raceway are analyzed, and the dynamic equations of the target feed system are constructed, including:

[0016] Analyze the contact forces between the balls and raceways, as well as between the two raceways, and construct the Lagrange equations.

[0017] Based on the displacement of the nut on the lead screw, the Lagrange equation is calculated to obtain the linear motion equation of the nut and the linear kinetic energy of the nut is determined.

[0018] Based on the rotation angle of the ball screw, the Lagrange equation is calculated to obtain the rotational motion equation of the ball screw, and the rotational kinetic energy of the ball screw is determined.

[0019] According to a fault diagnosis method for a servo feed system provided by the present invention, a five-degree-of-freedom rolling bearing vibration model for the target feed system is constructed, including:

[0020] Based on the angular position of each rolling element and the angular velocity of the cage, the total deformation of the corresponding rolling element is determined;

[0021] By combining Hertzian contact theory, the contact force between each rolling element and the raceway is determined, and the vibration model of the five-degree-of-freedom rolling bearing is constructed.

[0022] According to the fault diagnosis method for a servo feed system provided by the present invention, vibration signals of different faults under different operating conditions are collected, and a multi-operating-condition fault database is constructed, including:

[0023] The ball screw pair sample was driven to move by a pre-built experimental test platform, and the vibration signal of the ball screw pair sample was collected.

[0024] The system simulates normal operating conditions and different fault conditions, and organizes and preprocesses the vibration signals collected under different operating conditions to build a multi-condition fault database.

[0025] According to the fault diagnosis method for a servo feed system provided by the present invention, the high-fidelity dynamic model is optimized by comparing the vibration signal, including:

[0026] Based on Kalman filtering and Bayesian estimation, the vibration signal and the simulation signal are dynamically matched.

[0027] By analyzing probability density distribution and comparing frequency domain features, the similarity between the simulated signal and the vibration signal is quantified, and a transferability evaluation index is established.

[0028] Domain adversarial network optimization is employed to optimize fault feature transfer, aiming to minimize the distribution difference between the simulated signal and the vibration signal, thereby adjusting the model parameters of the high-fidelity dynamic model.

[0029] According to a fault diagnosis method for a servo feed system provided by the present invention, the high-fidelity dynamic model is verified, including:

[0030] Feature extraction and envelope spectrum analysis are performed on the simulated signal and the vibration signal to obtain key features;

[0031] The key features are compared with the theoretical values ​​of the features, and the effectiveness of the high-fidelity dynamic model is verified based on the comparison results.

[0032] According to a fault diagnosis method for a servo feed system provided by the present invention, the method involves collecting actual vibration data of the target feed system and comparing the actual vibration data with twin data to determine the fault type of the target feed system, including:

[0033] The target feeding system collects data through multiple pre-deployed sensors, and the collected results are fused based on a support vector machine or deep learning model to generate actual vibration data.

[0034] The actual vibration data is compared with the twin data to determine the fault type of the target feed system.

[0035] Secondly, the present invention also provides a fault diagnosis device for a servo feed system, comprising:

[0036] A construction module is used to build a high-fidelity dynamic model for the target feed system; the high-fidelity dynamic model includes a ball screw pair dynamic model based on a six-degree-of-freedom differential equation system and a five-degree-of-freedom rolling bearing vibration model;

[0037] The acquisition module is used to collect vibration signals of different faults under different working conditions and to build a multi-working-condition fault database.

[0038] The optimization module is used to conduct fault simulation experiments using the high-fidelity dynamic model, obtain simulation signals, and compare them with the vibration signals to optimize and verify the high-fidelity dynamic model.

[0039] The diagnostic module is used to collect the actual vibration data of the target feed system and compare the actual vibration data with twin data to determine the fault type of the target feed system; the twin data consists of the collected vibration signal and the simulation signal of the optimized high-fidelity dynamic model.

[0040] Thirdly, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the servo feed system fault diagnosis method as described in the first aspect above.

[0041] Fourthly, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the servo feed system fault diagnosis method as described in the first aspect above.

[0042] Fifthly, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the servo feed system fault diagnosis method as described in the first aspect above.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] The fault diagnosis method for servo feed systems provided by this invention constructs a high-fidelity dynamic model of the ball screw feed system based on digital twins and combines multi-sensor data fusion to integrate high-precision modeling with data-driven approaches. This achieves real-time mapping and intelligent optimization of the physical system, enabling accurate fault diagnosis and prediction of the ball screw feed system. It solves the problems of poor accuracy and real-time performance in existing related technologies. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0046] Figure 1 is a flowchart of the fault diagnosis method for the servo feed system provided by the present invention;

[0047] Figure 2 is a schematic diagram of the dynamic model of the ball screw pair in an embodiment of the present invention;

[0048] Figure 3 is a schematic diagram of the five-degree-of-freedom rolling bearing vibration model in an embodiment of the present invention;

[0049] Figure 4 is a schematic diagram of the simulation experimental platform in an embodiment of the present invention;

[0050] Figure 5 is a schematic diagram of envelope spectrum analysis under pitting fault of the lead screw in an embodiment of the present invention.

[0051] Figure 6 is a schematic diagram of envelope spectrum analysis under bearing outer ring failure in an embodiment of the present invention;

[0052] Figure 7 is a probability density distribution diagram of pitting failure of the lead screw in an embodiment of the present invention;

[0053] Figure 8 is a probability density distribution diagram of the outer ring fault signal of the support bearing in an embodiment of the present invention;

[0054] Figure 9 is a structural block diagram of the servo feed system fault diagnosis device provided by the present invention.

[0055] Figure 10 is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0057] This invention provides a fault diagnosis method for a servo feed system. Figure 1 is a flowchart of the fault diagnosis method for a servo feed system provided by this invention. As shown in Figure 1, the method includes the following steps:

[0058] Step S101: Construct a high-fidelity dynamic model for the target feed system; the high-fidelity dynamic model includes a ball screw pair dynamic model based on a six-degree-of-freedom differential equation system and a five-degree-of-freedom rolling bearing vibration model;

[0059] Step S102: Collect vibration signals of different faults under different working conditions and construct a multi-working-condition fault database;

[0060] Step S103: Use a high-fidelity dynamic model to conduct a fault simulation experiment, obtain simulation signals, and compare them with vibration signals to optimize and verify the high-fidelity dynamic model.

[0061] Step S104: Collect actual vibration data of the target feed system and compare the actual vibration data with twin data to determine the fault type of the target feed system; the twin data consists of the collected vibration signal and the simulation signal of the optimized high-fidelity dynamic model.

[0062] In this method, a high-fidelity dynamic model is first constructed. This model serves as a substitute for the target feed system during fault diagnosis, simulating its operation under different parameters or conditions. Then, vibration signals under different conditions and faults are experimentally collected. Next, simulation signals from the high-fidelity dynamic model during faults are acquired, and these signals are dynamically compared with the experimentally collected signals to further optimize the model and improve its accuracy. Finally, by monitoring the actual operation of the target feed system, vibration signal data detected by multiple sensors are compared in real-time with digital twin data to diagnose faults in the target feed system. In this process, a high-fidelity dynamic model of the ball screw feed system is constructed based on digital twins. Combined with multi-sensor data fusion, high-precision modeling and data-driven approaches are integrated, achieving real-time mapping and intelligent optimization of the physical system. This enables accurate fault diagnosis and prediction of the ball screw feed system, solving the problems of poor accuracy and real-time performance in existing related technologies.

[0063] In some embodiments, step S101, constructing a dynamic model of the ball screw pair for the target feed system, includes: analyzing the mechanical characteristics of the balls in the raceway and constructing the dynamic equations of the target feed system; determining the coupling relationship between the rotation and linear motion of the ball screw based on the dynamic equations of the target feed system; constructing differential equations of the dynamic behavior of the ball screw pair in the target feed system by combining the coupling relationship between the rotation and linear motion of the ball screw, and generating a dynamic model of the ball screw pair.

[0064] In this embodiment, the mechanical properties of the balls within the raceways are analyzed, and the dynamic equations of the target feed system are constructed. This includes: analyzing the contact forces between the balls and the raceways, as well as between the two raceways, and constructing the Lagrange equation; based on the displacement of the nut on the screw, the Lagrange equation is calculated to obtain the linear motion equation of the nut, and the linear kinetic energy of the nut is determined; based on the rotation angle of the ball screw, the Lagrange equation is calculated to obtain the rotational motion equation of the ball screw, and the rotational kinetic energy of the ball screw is determined.

[0065] For example, as shown in Figure 2, which is a schematic diagram of the dynamic model of the ball screw pair in an embodiment of the present invention, the dynamic model of the ball screw pair comprehensively considers the conversion of rotational motion to linear motion, torsional vibration, and the translation and rotation of the nut in various directions. Based on Newton's laws of motion and the principles of contact mechanics, a differential equation describing the dynamic behavior of the ball screw pair is constructed.

[0066] Assuming the servo motor output torque is T, and the rotation angle is... The moment of inertia, rotational damping coefficient, and rotational stiffness of the shaft are respectively , , The moment of inertia, rotational damping coefficient, and rotational stiffness of the ball screw are respectively... , , The mass, damping coefficient, and stiffness of the worktable are respectively... , , The table displacement and the lead of the leadscrew are respectively... , The relationship between the table displacement and the servo motor rotation angle is as follows:

[0067]

[0068] Where t represents time, and J0 represents the equivalent moment of inertia. B0 represents the equivalent damping coefficient. K0 represents the equivalent stiffness. .

[0069] Due to the existence of the ball screw lead angle, the worktable and screw will be affected by axial force and generate axial vibration. Therefore, when establishing the dynamic model of the ball screw pair, its axial movement degree of freedom must be considered. At the same time, to reduce computational complexity, the ball screw and ball nut are assumed to be rigid bodies, and their elastic deformation is ignored. In the accurate dynamic modeling, the mechanical characteristics of the balls in the raceway are first analyzed. When the ball screw is driven by the rotation of the servo motor, under the action of the lead angle, the balls generate an axial thrust on the screw nut, causing contact forces between the balls and the screw raceway and the nut raceway. The contact forces can be decomposed into components along the x, y, and z directions. Among them, the torque transmitted by the servo motor is T, the contact angle between the balls and the raceway is α, the lead angle of the screw is γ, and the contact force between the balls and the two raceways and between the raceways is P. The specific expressions are as follows:

[0070]

[0071]

[0072]

[0073] Among them, P x P represents the contact force in the x-direction. y P represents the contact force in the y-direction. z This represents the contact force in the z-direction. The basic form of the Lagrange equation is:

[0074]

[0075] in These are Lagrange quantities, where T represents kinetic energy and V represents potential energy. Represents generalized coordinates. This represents the corresponding generalized force. In a ball screw system, the generalized coordinates are... and Generalized forces are respectively and F represents the external force received by the system, and T represents the external torque experienced by the system.

[0076] right The linear motion equation of the nut can be obtained from the Lagrange equation:

[0077]

[0078]

[0079] in, Indicates displacement Find the first derivative, i.e., the velocity. Indicates displacement Find the second derivative, i.e., the acceleration.

[0080] right The Lagrange equations can be used to calculate the rotational motion equation of the lead screw:

[0081]

[0082]

[0083] in, Indicates the angle Find the first derivative, i.e., the angular velocity; Indicates the angle Find the second derivative, i.e., the angular acceleration. In the above expression, the displacement of the nut on the leadscrew is... The ball screw rotation angle is The external load on the system (axial load acting on the nut) is .

[0084] By constructing the system's kinetic and potential energies, the dynamic equations are obtained. The linear kinetic energy of the nut and the rotational kinetic energy of the ball screw are respectively:

[0085]

[0086]

[0087] Among them, T nut This represents the linear kinetic energy of the nut. For the mass of the nut, T represents the linear velocity of the nut. screw This represents the rotational kinetic energy of the ball screw. Let be the moment of inertia of the ball screw. Let be the angular velocity of the ball screw. Combining this with the above formula, we can obtain... and The main total kinetic energy equation:

[0088]

[0089] Where T represents the total kinetic energy of the system. Assume the linear stiffness of the ball screw is... The potential energy of the system (which can be represented by a spring model of the lead screw structure) is:

[0090]

[0091] Where V represents the system potential energy. The rotation and linear motion of the ball screw are coupled, determined by the screw pitch. By determining its derivative, we can obtain the coupling relationship between velocity and acceleration, namely:

[0092]

[0093]

[0094]

[0095] in, Indicates displacement. Indicates displacement Find the first derivative, i.e., the velocity; Indicates displacement Find the second derivative, i.e., the acceleration; Indicates a corner. Indicates the angle Find the first derivative, i.e., the angular velocity; Indicates the angle Find the second derivative, i.e., the angular acceleration.

[0096] Considering horizontal displacement when modeling a ball screw pair , Vertical displacement , axial displacement , This is simplified to a six-degree-of-freedom system, and its dynamic simulation model is constructed. Simultaneously considering the centrifugal force caused by manufacturing or installation errors, the dynamic differential equations of the ball screw pair are established:

[0097]

[0098] Each parameter represents a different physical meaning and system characteristic. This represents the displacement of the leadscrew in the x-direction. This indicates the velocity of the leadscrew in the x-direction. This represents the acceleration of the leadscrew in the x-direction. This represents the displacement of the leadscrew in the y-direction. This indicates the velocity of the leadscrew in the y-direction. This represents the acceleration of the leadscrew in the y-direction. This represents the displacement of the leadscrew in the z-direction. This indicates the velocity of the leadscrew in the z-direction. This represents the acceleration of the leadscrew in the z-direction. This represents the displacement of the nut in the x-direction. This indicates the velocity of the nut in the x-direction. This represents the acceleration of the nut in the x-direction. This indicates the displacement of the nut in the y-direction. This indicates the velocity of the nut in the y-direction. This represents the acceleration of the nut in the y-direction. This represents the displacement of the nut in the z-direction. This indicates the velocity of the nut in the z-direction. This represents the acceleration of the nut in the z-direction;

[0099] The total mass of the lead screw nut and the worktable, and the mass of the lead screw itself are respectively , Its magnitude is directly related to the dynamic performance of the system during startup, shutdown, and speed change; the connection damping between the lead screw and its fixed point, and between the slider and the guide rail, are respectively , Appropriate damping settings can effectively reduce the resonance risk of the system and improve its stability; the equivalent connection stiffness between the lead screw and its fixed part, and between the slider and the guide rail, are respectively , The components of the contact force in the x, y, and z directions are respectively , , These contact force components are not only crucial for driving the system's motion but also reflect the stress distribution within the system, making them significant for analyzing ball screw wear, fatigue life, and failure mechanisms. Furthermore, the frictional resistance between the slider and the guide rail is... The lead angle of the lead screw is The motor torque is The radius of the lead screw is eccentricity is The angular frequency of the lead screw rotation is The acceleration due to gravity is g, and the time is t.

[0100] In this embodiment, the screw drive system uses a fault evolution and coupling response mechanism model based on Lagrange multibody dynamics equations and Hertz contact theory to lay a solid physical foundation for the virtual entity modeling of the servo feed system, and successfully establishes virtual entities of various failure faults of the servo feed system.

[0101] Correspondingly, a five-degree-of-freedom rolling bearing vibration model for the target feed system is constructed, including: determining the total deformation of the corresponding rolling element based on the angular position of each rolling element and the angular velocity of the cage; determining the contact force between each rolling element and the raceway by combining Hertzian contact theory, and constructing a five-degree-of-freedom rolling bearing vibration model.

[0102] The five-DOF rolling bearing vibration model in this embodiment fully considers the radial, axial, and tangential vibration characteristics of the inner ring, outer ring, and rolling elements, as well as the nonlinearity of the contact between the balls and raceways. Based on the requirements of both research accuracy and practical needs, the dynamic simulation model of the supporting bearing is reasonably simplified to accurately describe the vibration response of the bearing under different operating conditions, providing crucial support for the digital twin model of the servo feed system. Specifically, the model assumes that the motion is in the same plane, focusing on the horizontal and vertical vibrations of the inner and outer rings. The rolling elements are considered rigid bodies, ignoring the stiffness, damping, and rotational inertia of the components due to elastic fluid lubrication. The lumped mass method is used to estimate the component mass, and the elastic contact between the raceway and rolling elements follows Hertz contact theory. Thus, a system with four degrees of freedom (horizontal and vertical) for the inner and outer rings and one unit resonator degree of freedom is constructed. The unit resonator is used to simulate the high-frequency natural vibrations of the excited bearing and other components, effectively capturing high-frequency vibration characteristics. Based on the above assumptions, the established dynamic simulation model of the supporting bearing is shown in Figure 3. Figure 3 is a schematic diagram of the five-DOF rolling bearing vibration model in this embodiment of the invention, which comprehensively integrates the key dynamic characteristics of the bearing.

[0103] In Figure 3, k s k p k r These represent the stiffness of the inner ring, outer ring, bearing housing, and unit resonator, respectively; c s c p c r The damping corresponding to the inner ring, outer ring, bearing housing, and unit resonator, respectively; m s m p m r These represent the masses of the inner ring, outer ring, bearing housing, and unit resonator, respectively; x s x p y s y p y b These represent the horizontal and vertical displacements of the inner and outer rings, respectively, as well as the numerical displacement of the unit resonator. When a bearing is subjected to a pure radial load, a load zone and a non-load zone appear on the bearing raceway. In the load zone, the rolling elements undergo elastic deformation due to compression, which in turn causes relative displacement between the rings, resulting in flexible vibration.

[0104] From a theoretical model perspective, the relevant expression for the total deformation of the j-th rolling element is:

[0105]

[0106]

[0107]

[0108] in, This represents the total deformation of the rolling elements, where c is the clearance. Represents the angular position of the j-th rolling element, n b w is the number of rolling elements. c Indicates the angular velocity of the cage. w represents the initial angular position of the rolling element. s Where is the inner angular velocity, a0 is the contact angle, and D is the inner angular velocity. b D is the diameter of the rolling element. p This is the diameter of the bearing pitch circle.

[0109] According to Hertz's contact theory, the contact force (i.e., elastic restoring force) between the i-th ball and the raceway is given by the following formula:

[0110]

[0111] in, δ represents the contact force between the i-th ball and the raceway. z Let z be the contact deformation between the rolling element and the inner and outer rings, z be the load deformation coefficient (typically 1.5 for ball bearings), and k be the equivalent contact stiffness. The expression for k is:

[0112]

[0113] Where, k i k is the contact stiffness between the inner ring and the rolling element. o This represents the contact stiffness between the outer ring and the rolling elements. In actual calculations, when the rolling elements pass through the load zone, the contact force between a single rolling element and the raceway can be accurately calculated using the above formula. Then, by summing these forces, the components f of the total nonlinear contact force on the raceway in the x and y directions can be obtained. x with f y .

[0114] In some embodiments, step S102, collecting vibration signals of different faults under different working conditions and constructing a multi-working-condition fault database, includes: driving the ball screw pair sample to move through a pre-built experimental test platform and collecting the vibration signal of the ball screw pair sample; simulating working conditions under normal conditions and different fault conditions, and organizing and preprocessing the vibration signals collected under different working conditions to construct a multi-working-condition fault database.

[0115] For example, Figure 4 is a schematic diagram of the simulation test bench in an embodiment of the present invention. As shown in Figure 4, a self-developed ball screw working condition simulation test bench is used. The test platform includes a platform body, a drive control system, and a signal acquisition system. The platform body is equipped with components such as the ball screw pair to be tested, BKBF bearing housing, screw nut housing, coupling, slider, guide rail, and stage. The drive control system is equipped with servo motors, drivers, servo controllers, and photoelectric gate sensors. The signal acquisition system mainly uses the 3255A2 model piezoelectric accelerometer and data acquisition instrument produced by DYTRAN Corporation of the United States. The sensor is installed on the screw nut housing to be close to the fault point and the stress point, reducing signal attenuation and interference, and ensuring that a comprehensive vibration signal reflecting the performance state of the ball screw pair is acquired.

[0116] To address various fault types, this embodiment conducted multiple sets of ball screw pair vibration signal acquisition experiments. These experimental data will serve as the basis for constructing the diagnostic model. To ensure the accuracy and reliability of the experimental data acquisition, the experiment will be conducted according to the following steps:

[0117] (1) Experimental preparation stage: Determine the ball screw pair sample required for the experiment, select a suitable accelerometer, prepare the sensor mounting fixture and matching signal transmission cable, and ensure the stability and accuracy of signal transmission;

[0118] (2) Set up an experimental test platform, install the ball screw and bearing housing, and install the 3255A2 and 3143M16 acceleration sensors in appropriate positions on the ball screw nut and bearing housing respectively. Ensure that the sensor and nut are in close contact and are firmly installed to avoid loosening or displacement during operation, which would affect the accuracy of signal acquisition.

[0119] (3) Connect the experimental platform to the drive control system and signal acquisition system, and set the parameters of the data acquisition equipment, including adjusting parameters such as sampling frequency, sampling time, and gain;

[0120] (4) Based on the working characteristics of the ball screw and the expected fault characteristic frequency range, the sampling frequency should be set reasonably to ensure that the key information in the vibration signal can be accurately captured, while avoiding excessively high sampling frequency that leads to excessive data volume and waste of computing resources.

[0121] (5) Start the experimental test platform, set the motor speed to 300 rpm, the effective stroke of uniform speed to 600 mm, and the sampling frequency fs to 6400 Hz according to the experimental requirements;

[0122] (6) During the operation of the ball screw, the data acquisition device is started synchronously to collect the vibration signal output by the acceleration sensor in real time and store it in the local storage medium of the data acquisition device. Each time, 10 round trips of data are collected, and a total of 4 cycles are performed to obtain a total of 40 sets of data.

[0123] (7) During the data acquisition process, pay close attention to the operating status of the data acquisition equipment and the quality of the acquired signals to ensure the continuity and stability of data acquisition. If abnormal signals or data acquisition interruptions are found, stop the experiment in time, investigate the cause of the fault, repair it, and then restart the data acquisition.

[0124] (8) Conduct working condition simulation experiments in the normal state and other 5 fault states in sequence, so that the ball screw pair can run under different working conditions, including different speed, load, running direction and other combination working conditions, and repeat the experimental process 2-7.

[0125] After data acquisition, the stored vibration data is transferred to a computer for further preprocessing and analysis. This experiment used a sliding window overlapping sampling method with M=2048 and L=1000. 240 samples were collected under each condition, with each sample containing 2048 data points. The data was then divided into datasets D, E, and F according to rotational speeds of 300, 600, and 1000 rpm, as detailed in Table 1.

[0126] Table 1. Detailed information on the ball screw pair operating condition simulation dataset.

[0127]

[0128] In Table 1, WQF represents a leadscrew bending fault, DSF represents a leadscrew pitting fault, FHF represents a combined internal fault in the leadscrew, NF represents a fault in the inner ring of the support bearing, and WF represents a fault in the outer ring of the support bearing. For each fault, experiments were conducted at three speeds: 300 r, 600 r, and 1000 r.

[0129] In some embodiments, step S103, by comparing vibration signals, optimizes the high-fidelity dynamics model, including: dynamically matching vibration signals and simulation signals based on Kalman filtering and Bayesian estimation; quantifying the similarity between simulation signals and vibration signals through probability density distribution analysis and frequency domain feature comparison, and establishing a transferability evaluation index; and using a domain adversarial network to optimize fault feature transfer, with the goal of minimizing the distribution difference between simulation signals and vibration signals, and adjusting the model parameters of the high-fidelity dynamics model.

[0130] Based on this, the high-fidelity dynamic model is validated, including: feature extraction and envelope spectrum analysis of the simulation signal and vibration signal to obtain key features; comparison of key features with theoretical feature values, and validation of the effectiveness of the high-fidelity dynamic model based on the comparison results.

[0131] For example, based on Kalman filtering and Bayesian estimation, experimental vibration signals (such as acceleration signals) and simulated signals are dynamically matched, and model parameters (such as wear amount and stiffness degradation coefficient) are adjusted. Specifically, through probability density distribution analysis and frequency domain feature comparison, the similarity between simulated signals and vibration signals (actual signals) is quantified, and a transferability evaluation index is established. As shown in Figures 5 and 6, Figure 5 is a schematic diagram of envelope spectrum analysis under lead screw pitting fault in an embodiment of the present invention, and Figure 6 is a schematic diagram of envelope spectrum analysis under bearing outer ring fault in an embodiment of the present invention. From the time domain waveforms in Figures 5 and 6, it can be observed that the impact waveforms of simulated signals and actual signals are very similar. The main frequency components of the simulated and actual signals in the lead screw pitting envelope spectrum are 33.3315Hz and 23.8322Hz, respectively, with relative deviations from the theoretical characteristic frequencies of 0.11% and 28.58%. Clearly, the lead screw experiences unstable vibrations in the initial stage of actual operation. Therefore, based on the difference between the latter two characteristic frequencies, the characteristic difference frequencies between the simulated and actual signals are calculated to be 33.3315Hz and 33.6993Hz, with relative deviations of 0.11% and 0.99%, respectively. In the bearing outer ring fault envelope spectrum, the main frequency components of the simulated and actual signals are 13.8889Hz and 13.8881Hz, respectively, with relative deviations from the theoretical characteristic frequencies of 0.46% and 0.47%, respectively.

[0132] Because the probability distributions of simulated signals and actual signals are inconsistent, the transfer diagnostic results may be unsatisfactory. Therefore, before applying simulated signals, it is necessary to conduct signal transferability analysis from the perspective of probability density distribution, and to estimate the probability density distributions of the two signals for pitting faults in the ball screw raceway in detail.

[0133] Figure 7 is a probability density distribution diagram of lead screw pitting fault in an embodiment of the present invention. As shown in Figure 7, the data distribution is relatively concentrated because the influencing factors considered in the simulation signal are relatively singular. In contrast, the actual signal contains more random influences, resulting in a relatively lower peak probability density. Although there are some differences between the simulation signal and the actual signal, the probability density distributions of the two signals are quite similar. Obviously, both signals contain important fault characteristic information, and the simulation signal of lead screw pitting fault has transferability for the diagnosis of actual lead screw pitting faults.

[0134] Figure 8 is a probability density distribution diagram of the outer ring fault signal of the support bearing in an embodiment of the present invention. As shown in Figure 8, the probability density distribution of the simulated signal is relatively concentrated, while the actual signal has a lower probability density peak and a smoother probability distribution due to the influence of various random factors of other components. Although there are certain differences between the simulated signal and the actual signal, the probability density distributions of the two signals under the same fault condition are quite similar. The difference in fault feature distribution between the two signals under the same fault type can be reduced by a fault feature migration strategy based on a domain adversarial mechanism, thereby realizing the migration from simulated signal to actual fault diagnosis.

[0135] Based on the above embodiments, step S104 involves collecting actual vibration data of the target feed system and comparing the actual vibration data with twin data to determine the fault type of the target feed system. This includes: collecting data from the target feed system using multiple pre-deployed sensors and fusing the collected results based on a support vector machine or deep learning model to generate actual vibration data; and comparing the actual vibration data with twin data to determine the fault type of the target feed system.

[0136] For example, for pitting failures, the transient impact frequency (theoretical characteristic frequency ±5% deviation) and envelope spectrum energy distribution are extracted as key features; based on support vector machine (SVM) or deep learning model, multi-sensor data (acceleration, stress) are fused to achieve failure classification (normal, pitting, wear, bending, etc.).

[0137] This invention also provides a servo feed system fault diagnosis device. The servo feed system fault diagnosis device provided by this invention is described below. The servo feed system fault diagnosis device described below can be referred to in correspondence with the servo feed system fault diagnosis method described above. Figure 9 is a structural block diagram of the servo feed system fault diagnosis device provided by this invention. As shown in Figure 9, the device includes:

[0138] Module 901 is used to construct a high-fidelity dynamic model for the target feed system; the high-fidelity dynamic model includes a ball screw pair dynamic model based on a six-degree-of-freedom differential equation system and a five-degree-of-freedom rolling bearing vibration model;

[0139] The acquisition module 902 is used to acquire vibration signals of different faults under different working conditions and to build a multi-working-condition fault database.

[0140] The optimization module 903 is used to conduct fault simulation experiments using a high-fidelity dynamic model, obtain simulation signals, and compare them with vibration signals to optimize and verify the high-fidelity dynamic model.

[0141] The diagnostic module 904 is used to collect actual vibration data of the target feed system and compare the actual vibration data with twin data to determine the fault type of the target feed system; the twin data consists of the collected vibration signal and the simulation signal of the optimized high-fidelity dynamic model.

[0142] In use, this device first constructs a high-fidelity dynamic model using module 901. This model serves as a substitute for the target feed system during fault diagnosis, simulating its operation under different parameters or conditions. Then, module 902 acquires vibration signals under various conditions and faults through experiments. Optimization module 903 then acquires simulation signals from the high-fidelity dynamic model during faults and dynamically compares these signals with experimentally acquired signals to further optimize the model and improve its accuracy. Finally, diagnosis module 904 detects the actual operation of the target feed system and compares the vibration signal data detected by multiple sensors with the digital twin data in real time to diagnose faults in the target feed system. In this process, a high-fidelity dynamic model of the ball screw feed system is constructed based on digital twin data. Combined with multi-sensor data fusion, high-precision modeling and data-driven approaches are integrated, achieving real-time mapping and intelligent optimization of the physical system. This enables accurate fault diagnosis and prediction of the ball screw feed system, solving the problems of poor accuracy and real-time performance in existing related technologies.

[0143] Figure 10 illustrates a schematic diagram of the physical structure of an electronic device. As shown in Figure 10, the electronic device may include: a processor 1001, a communication interface 1002, a memory 1003, and a communication bus 1004. The processor 1001, communication interface 1002, and memory 1003 communicate with each other via the communication bus 1004. The processor 1001 can call logical instructions in the memory 1003 to execute a servo feed system fault diagnosis method, which includes:

[0144] A high-fidelity dynamic model for the target feed system is constructed; the high-fidelity dynamic model includes a ball screw pair dynamic model based on a six-degree-of-freedom differential equation system and a five-degree-of-freedom rolling bearing vibration model;

[0145] Vibration signals of different faults under different working conditions were collected, and a multi-working-condition fault database was constructed.

[0146] Fault simulation experiments were conducted using a high-fidelity dynamic model to obtain simulation signals, which were then compared with vibration signals to optimize and verify the high-fidelity dynamic model.

[0147] The actual vibration data of the target feed system is collected and compared with the twin data to determine the fault type of the target feed system. The twin data consists of the collected vibration signal and the simulation signal of the optimized high-fidelity dynamic model.

[0148] Furthermore, the logical instructions in the aforementioned memory 1003 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0149] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to perform the servo feed system fault diagnosis method provided by the above methods, the method including:

[0150] A high-fidelity dynamic model for the target feed system is constructed; the high-fidelity dynamic model includes a ball screw pair dynamic model based on a six-degree-of-freedom differential equation system and a five-degree-of-freedom rolling bearing vibration model;

[0151] Vibration signals of different faults under different working conditions were collected, and a multi-working-condition fault database was constructed.

[0152] Fault simulation experiments were conducted using a high-fidelity dynamic model to obtain simulation signals, which were then compared with vibration signals to optimize and verify the high-fidelity dynamic model.

[0153] The actual vibration data of the target feed system is collected and compared with the twin data to determine the fault type of the target feed system. The twin data consists of the collected vibration signal and the simulation signal of the optimized high-fidelity dynamic model.

[0154] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the servo feed system fault diagnosis method provided by the above methods, the method comprising:

[0155] A high-fidelity dynamic model for the target feed system is constructed; the high-fidelity dynamic model includes a ball screw pair dynamic model based on a six-degree-of-freedom differential equation system and a five-degree-of-freedom rolling bearing vibration model;

[0156] Vibration signals of different faults under different working conditions were collected, and a multi-working-condition fault database was constructed.

[0157] Fault simulation experiments were conducted using a high-fidelity dynamic model to obtain simulation signals, which were then compared with vibration signals to optimize and verify the high-fidelity dynamic model.

[0158] The actual vibration data of the target feed system is collected and compared with the twin data to determine the fault type of the target feed system. The twin data consists of the collected vibration signal and the simulation signal of the optimized high-fidelity dynamic model.

[0159] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0160] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0161] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A fault diagnosis method for a servo feed system, characterized in that, include: A high-fidelity dynamic model for the target feed system is constructed. This model includes a ball screw pair dynamic model based on a six-degree-of-freedom differential equation system and a five-degree-of-freedom rolling bearing vibration model. Vibration signals from different faults under various operating conditions are collected, and a multi-condition fault database is constructed. Fault simulation experiments are conducted using the high-fidelity dynamic model to obtain simulation signals, which are then compared with the vibration signals to optimize and verify the high-fidelity dynamic model. Actual vibration data of the target feed system is collected, and the actual vibration data is compared with twin data to determine the fault type of the target feed system. The twin data consists of the collected vibration signals and the simulation signals from the optimized high-fidelity dynamic model. The construction of the ball screw pair dynamic model for the target feed system includes analyzing the mechanics of the balls within the raceway. The system's dynamic equations are constructed based on the characteristics of the target feed system. Based on these equations, the coupling relationship between the ball screw's rotation and linear motion is determined. The differential equations of the ball screw pair's dynamic behavior in the target feed system are constructed by combining these coupling relationships, and a dynamic model of the ball screw pair is generated. The mechanical characteristics of the balls within the raceways are analyzed, and the dynamic equations of the target feed system are constructed, including: analyzing the contact forces between the balls and the raceways, and between the two raceways, and constructing the Lagrange equations; calculating the Lagrange equations based on the nut's displacement on the screw to obtain the linear motion equations of the nut, and determining the linear kinetic energy of the nut; and calculating the Lagrange equations based on the ball screw's rotation angle to obtain the rotational motion equations of the ball screw, and determining the rotational kinetic energy of the ball screw.

2. The servo feed system fault diagnosis method according to claim 1, characterized in that, A five-degree-of-freedom rolling bearing vibration model for the target feed system is constructed, including: determining the total deformation of the rolling element based on the angular position of each rolling element and the angular velocity of the cage; determining the contact force between each rolling element and the raceway using Hertzian contact theory; and constructing the five-degree-of-freedom rolling bearing vibration model.

3. The servo feed system fault diagnosis method according to claim 1, characterized in that, Vibration signals of different faults under different working conditions are collected, and a multi-working-condition fault database is constructed. This includes: driving the ball screw pair sample to move through a pre-built experimental test platform and collecting the vibration signal of the ball screw pair sample; simulating normal conditions and different fault conditions, and sorting and preprocessing the vibration signals collected under different working conditions to construct a multi-working-condition fault database.

4. The servo feed system fault diagnosis method according to claim 1, characterized in that, The high-fidelity dynamic model is optimized by comparing the vibration signals, including: dynamically matching the vibration signals and the simulated signals based on Kalman filtering and Bayesian estimation; quantifying the similarity between the simulated signals and the vibration signals through probability density distribution analysis and frequency domain feature comparison, and establishing a transferability evaluation index; and using a domain adversarial network to optimize fault feature transfer, with the goal of minimizing the distribution difference between the simulated signals and the vibration signals, and adjusting the model parameters of the high-fidelity dynamic model.

5. The servo feed system fault diagnosis method according to claim 1, characterized in that, The high-fidelity dynamic model is validated by: extracting features and performing envelope spectrum analysis on the simulated signal and the vibration signal to obtain key features; comparing the key features with the theoretical values ​​of the features; and validating the effectiveness of the high-fidelity dynamic model based on the comparison results.

6. The servo feed system fault diagnosis method according to claim 1, characterized in that, The process of collecting actual vibration data of the target feed system and comparing the actual vibration data with twin data to determine the fault type of the target feed system includes: collecting data from the target feed system using multiple pre-deployed sensors, fusing the collected results based on a support vector machine or deep learning model to generate actual vibration data; and comparing the actual vibration data with the twin data to determine the fault type of the target feed system.

7. A servo feed system fault diagnosis device, used to implement the servo feed system fault diagnosis method according to any one of claims 1-6, characterized in that, include: The system includes a construction module for building a high-fidelity dynamic model of the target feed system; the high-fidelity dynamic model includes a ball screw pair dynamic model based on a six-degree-of-freedom differential equation system and a five-degree-of-freedom rolling bearing vibration model; an acquisition module for acquiring vibration signals of different faults under different working conditions and building a multi-working-condition fault database; and an optimization module for conducting fault simulation experiments using the high-fidelity dynamic model, acquiring simulation signals, and comparing them with the vibration signals to optimize and verify the high-fidelity dynamic model. The diagnostic module is used to collect the actual vibration data of the target feed system and compare the actual vibration data with twin data to determine the fault type of the target feed system; the twin data consists of the collected vibration signal and the simulation signal of the optimized high-fidelity dynamic model.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the servo feed system fault diagnosis method as described in any one of claims 1 to 6.

Images