Component fault diagnosis method and system based on data and mechanism fusion

Abstract

The application discloses a kind of based on data and mechanism fusion component fault diagnosis method and system, the component fault diagnosis method includes obtaining the vibration acceleration of predetermined key component of screw pair and the rotational speed of screw;The vibration acceleration of predetermined key component of screw pair is Fourier transformed, corresponding frequency domain data is obtained, and the frequency domain feature of predetermined key component of screw pair is extracted from frequency domain data;The fault main frequency of predetermined key component of screw pair is calculated using the physical parameters of screw pair and the rotational speed of screw;The fault diagnosis of predetermined key component of screw pair is carried out based on the frequency domain feature and fault main frequency of predetermined key component of screw pair.The application fuses data driving and mechanism analysis, and can improve fault diagnosis efficiency and reliability.

Description

Technical Field

[0001] This invention belongs to the field of CNC machine tool fault diagnosis technology, specifically relating to a component fault diagnosis method and system based on the fusion of data and mechanism. Background Technology

[0002] CNC machine tools are the core foundational equipment for achieving intelligent manufacturing in the equipment manufacturing industry. They can significantly improve production efficiency and complete high-precision, flexible machining of complex structural parts. As an integrated and composite upgrade of CNC machine tools, boring and milling machining centers combine multiple process capabilities such as boring, milling, and drilling. Their operational reliability is crucial to the stability of the production system.

[0003] Currently, the maintenance of feed systems in boring and milling machining centers generally adopts a combination of preventative maintenance based on fixed cycles and reactive repair after a failure. This model faces the following significant problems: the core components of the feed system (such as ball screw pairs) have complex structures, high precision requirements, long procurement cycles, and high unit value; when a failure occurs, the diagnostic process relies on manual experience, and disassembly and inspection are time-consuming, resulting in a significant increase in unplanned downtime, which seriously affects equipment utilization and production cycle time.

[0004] Current maintenance methods lack the ability to provide early warnings of failures in critical mechanical components such as lead screws. Therefore, how to conduct effective fault diagnosis to achieve predictive maintenance, accurately guide repair operations, and optimize spare parts inventory management has become an urgent technical problem to be solved.

[0005] Chinese invention patent application CN120831933A discloses a dynamic fault diagnosis method and system. This method generates a global time reference signal using a spindle encoder and clock synchronization protocol, collects vibration data of the spindle bearing, current data of the electrical cabinet, and process parameters at various unit times, and generates a preprocessed data sequence after transmission delay compensation and multi-rate upsampling. The vibration and current data are then input into a preset mechanical-electrical transfer function model to calculate time delay parameters. Based on these time delay parameters, the preprocessed data sequence is phase-aligned to generate an aligned data sequence. This aligned data sequence is then input into a time-series neural network to output a fused feature vector. Finally, the cross-correlation coefficient of the fused feature vector is calculated, and a fault diagnosis result is generated based on a preset cross-correlation threshold. The technical defects of this solution include: 1) The system is complex to implement and depends on specific hardware: The core of the solution depends on the spindle encoder Z-phase signal and high-precision clock synchronization protocol, which increases the system cost and deployment threshold, and is especially unsuitable for old machine tools without encoders or with unreliable encoder signals; 2) High dependence on data and knowledge: The training of the temporal neural network depends heavily on a large number of labeled fault samples, while early fault samples are often scarce in actual industry, affecting the model's generalization ability.

[0006] Chinese invention patent application CN121143204A discloses a machine learning-based fault diagnosis system for CNC machine tools. This system constructs a dynamically adaptive fault diagnosis model through modules such as multi-source sensor data acquisition, operational feature encoding, incremental learning analysis, genetic optimization, and integrated diagnostic decision-making. Combined with adaptive control parameters and early warning execution, it achieves efficient and accurate diagnosis of CNC machine tool faults, adapting to dynamic changes in equipment, improving diagnostic reliability and accuracy, reducing misjudgments, supporting real-time control parameter adjustment and early warning, and ensuring safe and efficient equipment operation. However, this solution has several technical drawbacks: 1) The system is overly complex and difficult to implement in engineering: The system integrates at least eight core modules and dozens of algorithms (genetic algorithms, gradient boosting, near-end strategy optimization, etc.), forming an extremely complex serial link. Deploying, debugging, and maintaining such a system in an industrial setting presents high costs and technical barriers; 2) The decision-making process lacks interpretability: The core of the system is still a data-driven machine learning model, and its decision-making process lacks interpretability. When misdiagnosis occurs, technicians find it difficult to trace the cause, hindering root cause analysis of the fault. Summary of the Invention

[0007] To address the aforementioned issues, this invention proposes a component fault diagnosis method and system based on data and mechanism fusion, which integrates data-driven and mechanism analysis to improve fault diagnosis efficiency and reliability.

[0008] To achieve the above-mentioned technical objectives and effects, the present invention is implemented through the following technical solution:

[0009] In a first aspect, the present invention provides a component fault diagnosis method based on data and mechanism fusion, comprising:

[0010] Obtain the vibration acceleration and rotational speed of the key components of the lead screw pair;

[0011] The vibration acceleration of the predetermined key component of the lead screw pair is subjected to Fourier transform to obtain the corresponding frequency domain data, and the frequency domain characteristics of the predetermined key component of the lead screw pair are extracted from the frequency domain data.

[0012] Using the physical parameters of the lead screw pair and the rotational speed of the lead screw, the failure frequency of the predetermined key components of the lead screw pair is calculated;

[0013] Fault diagnosis of key components of the lead screw pair is performed based on the frequency domain characteristics and fault frequency of the key components.

[0014] In conjunction with the first aspect, optionally, the frequency domain features include the frequency center, mean square frequency, and frequency variance;

[0015] The formula for calculating the frequency center is:

[0016] ,

[0017] In the formula, As the frequency center, This represents the number of spectral lines. For the first Frequency values ​​of the spectral lines for The corresponding amplitude;

[0018] The formula for calculating the mean square frequency is:

[0019] ,

[0020] In the formula, The mean square frequency;

[0021] The formula for calculating the frequency variance is:

[0022] ,

[0023] In the formula, Here is the frequency variance.

[0024] In conjunction with the first aspect, optionally, the predetermined key components include a lead screw nut, a lead screw, and / or balls.

[0025] In conjunction with the first aspect, optionally, the formula for calculating the fault frequency of the lead screw nut is:

[0026] ,

[0027] The formula for calculating the fault frequency of the leadscrew is:

[0028] ,

[0029] The formula for calculating the failure frequency of the ball bearing is:

[0030] ,

[0031] In the formula, This refers to the number of balls; The frequency of the lead screw rotation is calculated from the lead screw's rotational speed. The diameter of the ball bearing is... The diameter of the lead screw pitch circle. It represents the contact angle.

[0032] In conjunction with the first aspect, optionally, the fault diagnosis of the predetermined key component based on the frequency domain characteristics and the fault dominant frequency of the predetermined key component includes:

[0033] When the amplitude of a predetermined critical component at its fault frequency is abnormal and its frequency domain characteristics meet the preset early fault threshold conditions, the predetermined critical component is determined to be in an early fault state, and a diagnostic report containing the equipment information, warning time, abnormal location and level of the predetermined critical component is generated.

[0034] In conjunction with the first aspect, optionally, the fault diagnosis of the predetermined key component based on the frequency domain characteristics and the fault dominant frequency of the predetermined key component includes:

[0035] When the amplitude of the Nth harmonic of a predetermined critical component is abnormal and its frequency domain characteristics meet the preset mid-term fault threshold condition, the predetermined critical component is determined to be in a mid-term fault state, and a diagnostic report containing the equipment information, warning time, abnormal location and level of the predetermined critical component is generated, where N>1.

[0036] In conjunction with the first aspect, optionally, the fault diagnosis of the predetermined key component based on the frequency domain characteristics and the fault dominant frequency of the predetermined key component includes:

[0037] When the amplitude of the Mth harmonic of a predetermined critical component is abnormal and its frequency domain characteristics meet the preset late fault threshold condition, the predetermined critical component is determined to be in a late fault state, and a diagnostic report containing the equipment information, warning time, abnormal location and level of the predetermined critical component is generated, where M>N.

[0038] In conjunction with the first aspect, optionally, the component fault diagnosis method further includes: preprocessing the obtained vibration acceleration and rotational speed of the predetermined key components of the lead screw pair, specifically including:

[0039] The vibration acceleration of the key components of the lead screw and the rotational speed of the lead screw are obtained by wavelet transform to remove high-frequency noise and obtain the noise-reduced data.

[0040] The denoised data is processed using a high-pass filter to remove ultra-low frequency interference components, resulting in data after secondary denoising.

[0041] Linear interpolation is used to replace outliers in the data after secondary denoising, resulting in preprocessed data.

[0042] Secondly, the present invention provides a component fault diagnosis system based on data and mechanism fusion, comprising:

[0043] Vibration acceleration sensor, used to acquire the vibration acceleration of predetermined key components of the lead screw pair;

[0044] A speed sensor is used to obtain the speed of the leadscrew;

[0045] The controller, connected to the vibration acceleration sensor and the rotation speed sensor respectively, includes:

[0046] The data acquisition module is used to acquire the vibration acceleration of key components of the lead screw pair and the rotational speed of the lead screw;

[0047] The feature value preparation module is used to perform Fourier transform on the vibration acceleration of the predetermined key components of the lead screw pair to obtain the corresponding frequency domain data, and extract the preset frequency domain features from the frequency domain data; and calculate the fault main frequency of the predetermined key components of the lead screw pair using the physical parameters of the lead screw pair and the rotational speed of the lead screw.

[0048] The fault diagnosis algorithm module is used to diagnose faults in the predetermined key components of the lead screw pair based on the frequency domain characteristics and the main fault frequency.

[0049] Thirdly, the present invention provides a component fault diagnosis system based on data and mechanism fusion, including a storage medium and a processor;

[0050] The storage medium is used to store instructions;

[0051] The processor is configured to operate according to the instructions to perform the method according to any one of the first aspects.

[0052] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0053] This invention provides a component fault diagnosis method and system based on data and mechanism fusion, which integrates data-driven and mechanism analysis to improve fault diagnosis efficiency and reliability.

[0054] Specifically: First, based on the inherent parameters and other mechanistic knowledge of the lead screw pair, the theoretical fault dominant frequency of the key components of the lead screw pair is calculated. Simultaneously, a spectrum diagram is plotted using real-time vibration acceleration data of the predetermined key components, and key frequency domain features such as frequency center, frequency variance, and mean square frequency are extracted. By analyzing whether the amplitude of the fault dominant frequency and its harmonics of the predetermined key components of the lead screw pair in the spectrum diagram increases abnormally, and combining this with the quantitative changes in frequency domain features, a comprehensive judgment is made on the presence of a fault, the faulty component is located, and its severity is assessed. This enables accurate diagnosis and significantly improves the practical feasibility and application value of the technical solution of this invention. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly described below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort, wherein:

[0056] Figure 1 This is one of the flowcharts illustrating a component fault diagnosis method based on data and mechanism fusion according to an embodiment of the present invention;

[0057] Figure 2 This is a second schematic flowchart of a component fault diagnosis method based on data and mechanism fusion according to an embodiment of the present invention;

[0058] Figure 3 This is a schematic diagram of the data processing flow of a fault diagnosis module according to an embodiment of the present invention;

[0059] Figure 4 This is a schematic diagram of the installation of a vibration acceleration sensor according to an embodiment of the present invention. Detailed Implementation

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

[0061] Furthermore, if the embodiments of this invention involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. If the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0062] Example 1

[0063] This invention provides a component fault diagnosis method based on data and mechanism fusion, such as... Figure 1 As shown, it includes the following steps:

[0064] (1) Obtain the vibration acceleration of the key components of the lead screw pair and the rotational speed of the lead screw;

[0065] (2) Perform Fourier transform on the vibration acceleration of the predetermined key components of the lead screw pair to obtain the corresponding frequency domain data, and extract the preset frequency domain features from the frequency domain data;

[0066] (3) Calculate the main frequency of failure of the key components of the lead screw pair using the physical parameters of the lead screw pair and the rotational speed of the lead screw;

[0067] (4) Based on the frequency domain characteristics and fault frequency of the predetermined key components of the lead screw pair, perform fault diagnosis of the predetermined key components of the lead screw pair.

[0068] The above-described solution integrates data-driven approaches with mechanism analysis, which improves the efficiency and reliability of fault diagnosis. Furthermore, all data in this invention originates from real, complex, and high-noise industrial environments, rather than idealized test benches. This ensures that the component fault diagnosis method in this embodiment not only effectively overcomes data bottlenecks and achieves precise location diagnosis of key components such as lead screw pairs, but also that its conclusions possess extremely high engineering credibility and direct applicability.

[0069] In one specific embodiment of the present invention, the frequency domain features include frequency center, mean square frequency, and frequency variance;

[0070] The formula for calculating the frequency center is:

[0071] ,

[0072] In the formula, As the frequency center, This represents the number of spectral lines. For the first Frequency values ​​of the spectral lines for The corresponding amplitude, where the number of spectral lines , No. Frequency values ​​of spectral lines , Corresponding amplitude All of these are obtained from frequency domain data based on Fourier transform.

[0073] The formula for calculating the mean square frequency is:

[0074] ,

[0075] In the formula, The mean square frequency;

[0076] The formula for calculating the frequency variance is:

[0077] ,

[0078] In the formula, Here is the frequency variance.

[0079] In one specific embodiment of the present invention, the predetermined key components include a lead screw nut, a lead screw, and / or balls;

[0080] The formula for calculating the fault frequency of the lead screw nut is as follows:

[0081] ,

[0082] The formula for calculating the fault frequency of the leadscrew is:

[0083] ,

[0084] The formula for calculating the failure frequency of the ball bearing is:

[0085] ,

[0086] In the formula, This refers to the number of balls; The frequency of the lead screw rotation is calculated from the lead screw's rotational speed. The diameter of the ball bearing is... The diameter of the lead screw pitch circle. It represents the contact angle.

[0087] In one specific embodiment of the present invention, the fault diagnosis of the predetermined key component based on the frequency domain characteristics and fault dominant frequency of the predetermined key component includes:

[0088] When the amplitude of a predetermined critical component at its fault frequency is abnormal and its frequency domain characteristics meet the preset early fault threshold conditions, the predetermined critical component is determined to be in an early fault state, and a diagnostic report containing the equipment information, warning time, abnormal location and level of the predetermined critical component is generated.

[0089] In one specific embodiment of the present invention, the fault diagnosis of the predetermined key component based on the frequency domain characteristics and fault dominant frequency of the predetermined key component includes:

[0090] When the amplitude of the Nth harmonic of a predetermined critical component is abnormal and its frequency domain characteristics meet the preset mid-term fault threshold conditions, the predetermined critical component is determined to be in a mid-term fault state, and a diagnostic report containing the equipment information, warning time, abnormal location and level of the predetermined critical component is generated. In the specific implementation process, N = 2 / 3 / 4.

[0091] In one specific embodiment of the present invention, the fault diagnosis of the predetermined key component based on the frequency domain characteristics and fault dominant frequency of the predetermined key component includes:

[0092] When the amplitude of the Mth harmonic of a predetermined critical component is abnormal and its frequency domain characteristics meet the preset late fault threshold condition, the predetermined critical component is determined to be in a late fault state, and a diagnostic report containing the equipment information, warning time, abnormal location and level of the predetermined critical component is generated, M>N, and in the specific implementation process, M≥5.

[0093] In one specific embodiment of the present invention, the component fault diagnosis method further includes: preprocessing the obtained vibration acceleration and rotational speed of the predetermined key components of the lead screw pair, specifically including:

[0094] The vibration acceleration of the key components of the lead screw and the rotational speed of the lead screw are obtained by wavelet transform to remove high-frequency noise and obtain the noise-reduced data.

[0095] The denoised data is processed using a high-pass filter to remove ultra-low frequency interference components, resulting in data after secondary denoising.

[0096] Linear interpolation is used to replace outliers in the data after secondary denoising, resulting in preprocessed data.

[0097] Example 2

[0098] This invention provides a component fault diagnosis system based on data and mechanism fusion, comprising:

[0099] A vibration acceleration sensor is used to acquire the vibration acceleration of predetermined key components of the lead screw assembly. In specific implementations, the vibration acceleration sensor can be installed at the left and right bearings and the lead screw nut of the lead screw assembly to collect the vibration acceleration of the lead screw nut, lead screw, and / or balls in real time. Figure 4 As shown, the vibration acceleration sensor is installed at measuring points 1-3;

[0100] A speed sensor is used to obtain the speed of the lead screw; in a specific implementation, the speed sensor can be installed on the non-drive side of the lead screw pair.

[0101] The controller, connected to the vibration acceleration sensor and the rotation speed sensor respectively, includes:

[0102] The data acquisition module is used to acquire the vibration acceleration of key components of the lead screw pair and the rotational speed of the lead screw;

[0103] The feature value preparation module is used to perform Fourier transform on the vibration acceleration of the predetermined key components of the lead screw pair to obtain the corresponding frequency domain data, and extract the preset frequency domain features from the frequency domain data; and calculate the fault main frequency of the predetermined key components of the lead screw pair using the physical parameters of the lead screw pair and the rotational speed of the lead screw.

[0104] The fault diagnosis algorithm module is used to diagnose faults in the predetermined key components of the lead screw pair based on the frequency domain characteristics and the main fault frequency.

[0105] The above-described solution integrates data-driven approaches with mechanistic analysis, improving the efficiency and reliability of fault diagnosis. Furthermore, all data in this invention originates from real, complex, and high-noise industrial environments, rather than idealized test benches. This allows the component fault diagnosis method in this embodiment to not only effectively overcome data bottlenecks and achieve precise location diagnosis of key components such as lead screws, but also provides conclusions with extremely high engineering credibility and direct applicability. In practical implementation, the component fault diagnosis system can employ a hierarchical data storage strategy, storing the collected raw vibration acceleration data at the edge, while uploading frequency domain characteristics and fault dominant frequencies to the cloud data center. This supports long-term tracking and in-depth analysis of the component fault diagnosis system's status.

[0106] In one specific embodiment of the present invention, the frequency domain features include frequency center, mean square frequency, and frequency variance;

[0107] The formula for calculating the frequency center is:

[0108] ,

[0109] In the formula, As the frequency center, This represents the number of spectral lines. For the first Frequency values ​​of the spectral lines for The corresponding amplitude, where the number of spectral lines , No. Frequency values ​​of spectral lines , Corresponding amplitude All of these are obtained from frequency domain data based on Fourier transform.

[0110] The formula for calculating the mean square frequency is:

[0111] ,

[0112] In the formula, The mean square frequency;

[0113] The formula for calculating the frequency variance is:

[0114] ,

[0115] In the formula, Here is the frequency variance.

[0116] In one specific embodiment of the present invention, the predetermined key components include a lead screw nut, a lead screw, and / or balls;

[0117] The formula for calculating the fault frequency of the lead screw nut is as follows:

[0118] ,

[0119] The formula for calculating the fault frequency of the leadscrew is:

[0120] ,

[0121] The formula for calculating the failure frequency of the ball bearing is:

[0122] ,

[0123] In the formula, This refers to the number of balls; The frequency of the lead screw rotation is calculated from the lead screw's rotational speed. The diameter of the ball bearing is... The diameter of the lead screw pitch circle. It represents the contact angle.

[0124] In one specific embodiment of the present invention, the fault diagnosis of the predetermined key component based on the frequency domain characteristics and fault dominant frequency of the predetermined key component includes:

[0125] When the amplitude of a predetermined critical component at its fault frequency is abnormal and its frequency domain characteristics meet the preset early fault threshold conditions, the predetermined critical component is determined to be in an early fault state, and a diagnostic report containing the equipment information, warning time, abnormal location and level of the predetermined critical component is generated.

[0126] In one specific embodiment of the present invention, the fault diagnosis of the predetermined key component based on the frequency domain characteristics and fault dominant frequency of the predetermined key component includes:

[0127] When the amplitude of the Nth harmonic of a predetermined critical component is abnormal and its frequency domain characteristics meet the preset mid-term fault threshold conditions, the predetermined critical component is determined to be in a mid-term fault state, and a diagnostic report containing the equipment information, warning time, abnormal location and level of the predetermined critical component is generated. In the specific implementation process, N = 2 / 3 / 4.

[0128] In one specific embodiment of the present invention, the fault diagnosis of the predetermined key component based on the frequency domain characteristics and fault dominant frequency of the predetermined key component includes:

[0129] When the amplitude of the Mth harmonic of a predetermined critical component is abnormal and its frequency domain characteristics meet the preset late fault threshold condition, the predetermined critical component is determined to be in a late fault state, and a diagnostic report containing the equipment information, warning time, abnormal location and level of the predetermined critical component is generated, M>N, and in the specific implementation process, M≥5.

[0130] In one specific embodiment of the present invention, the component fault diagnosis method further includes: preprocessing the obtained vibration acceleration and rotational speed of the predetermined key components of the lead screw pair, specifically including:

[0131] The vibration acceleration of the key components of the lead screw and the rotational speed of the lead screw are obtained by wavelet transform to remove high-frequency noise and obtain the noise-reduced data.

[0132] The denoised data is processed using a high-pass filter to remove ultra-low frequency interference components, resulting in data after secondary denoising.

[0133] Linear interpolation is used to replace outliers in the data after secondary denoising, resulting in preprocessed data.

[0134] In one specific embodiment of the present invention, the component fault diagnosis system further includes a display module, which is connected to the controller and transforms complex fault diagnosis results and frequency domain characteristics into an operating status diagram that equipment managers can understand at a glance, supporting them to quickly locate anomalies and intervene in a timely manner.

[0135] The following is combined with Figures 1-3 The present invention provides a detailed description of a component fault diagnosis system according to a specific embodiment.

[0136] The component fault diagnosis system includes a data acquisition module, a data preprocessing module, a feature preparation module, a fault diagnosis algorithm, and a system early warning module. In this embodiment, the application is a boring and milling machining center. In other embodiments, the application can also be key components (such as spindles, feed systems, servo drives, etc.) of various CNC machine tools, such as turning centers, multi-functional machining centers, and electrical discharge machining equipment. The working process of the component fault diagnosis system is as follows: Figure 2 As shown.

[0137] The data acquisition module is used to construct the monitoring system and acquire raw monitoring data. Specifically, a three-axis vibration acceleration sensor is installed at the left and right bearings and the nut of the lead screw pair in the boring and milling machining center to simultaneously monitor abnormal vibrations of the lead screw nut, lead screw, and balls. Simultaneously, a Hall effect speed sensor is installed on the non-drive side of the lead screw pair to measure the lead screw's operating speed in real time. Subsequently, the multi-channel signals from each sensor are synchronously acquired, time-series aligned, and initially integrated by the data acquisition unit. Finally, the integrated multi-source time-series data is stored in an edge computing device, providing a complete and consistent raw data foundation for subsequent processing.

[0138] The data preprocessing module purifies and standardizes the raw monitoring data to improve data quality. First, wavelet transform is used to reduce noise in the raw monitoring data, effectively separating and suppressing high-frequency noise in the signal. Then, a high-pass filter is used to filter out ultra-low-frequency interference components introduced by environmental or equipment drift. For potential outliers in the data, the preprocessing module uses linear interpolation to perform reasonable replacements based on a set replacement ratio threshold, ensuring data continuity and reliability. The final preprocessed data output will serve as input to the feature value preparation module, providing a high-quality data foundation for subsequent analysis and diagnosis.

[0139] Feature value preparation module: responsible for extracting key frequency domain features (including frequency center, mean square frequency and frequency variance) of each key component (lead screw, lead screw nut and ball) from the data preprocessing module, and calculating the fault main frequency of the lead screw, the fault main frequency of the lead screw nut and the ball based on the physical parameters of the lead screw pair, so as to provide quantitative basis for subsequent condition monitoring and fault location.

[0140] The frequency center (FC) represents the weighted average frequency of the signal spectrum, obtained by weighting the amplitudes of each frequency component in the spectrum. It is an indicator describing the central location of the signal's frequency distribution.

[0141] ,

[0142] In the formula, As the frequency center, This represents the number of spectral lines. For the first Frequency values ​​of the spectral lines for The corresponding amplitude;

[0143] Mean Square Frequency (MSF) represents the weighted average of the squares of the spectral amplitudes, reflecting the degree of energy concentration in the frequency domain.

[0144] ,

[0145] Frequency variance (VF) represents the degree of dispersion of frequency distribution in the spectrum, and is calculated as the weighted average of the squares of the differences between frequencies and their centers.

[0146] ,

[0147] The main frequency of failure in the lead screw nut: ;

[0148] The main frequency of the lead screw failure: ,

[0149] Ball bearing failure frequency: ,

[0150] In the formula, This refers to the number of balls; The frequency of the lead screw rotation is calculated from the lead screw's rotational speed. The diameter of the ball bearing is... The diameter of the lead screw pitch circle. It represents the contact angle.

[0151] The fault diagnosis algorithm module analyzes the vibration acceleration data of different key components at the edge side and plots the frequency domain data (horizontal axis is frequency (Hz), vertical axis is amplitude (representing vibration amplitude)) to obtain a spectrum diagram. It then checks for abnormal increases in amplitude at the fault frequencies of the nut, lead screw, and ball bearings, as well as their harmonics. Combining this with frequency domain characteristic values ​​(frequency center, mean square frequency, frequency variance), it comprehensively judges the fault status and severity at the corresponding locations and automatically generates a diagnostic report containing equipment information of predetermined key components, warning time, abnormal location, and level. For example... Figure 3 As shown, specifically:

[0152] When the amplitude of a predetermined critical component at its fault frequency is abnormal and its frequency domain characteristics meet the preset early fault threshold conditions, the predetermined critical component is determined to be in an early fault state, and a diagnostic report containing the equipment information, warning time, abnormal location and level of the predetermined critical component is generated; the preset early fault threshold conditions can be set according to actual needs, and are not specifically limited in this invention.

[0153] When the amplitude of the Nth harmonic of a predetermined critical component is abnormal and its frequency domain characteristics meet the preset mid-term fault threshold condition, the predetermined critical component is determined to be in a mid-term fault state, and a diagnostic report containing the equipment information, warning time, abnormal location and level of the predetermined critical component is generated, where N>1; in specific implementation, N can be set to 2-4; the preset mid-term fault threshold condition can be set according to actual needs, and is not specifically limited in this invention;

[0154] When the amplitude of the Mth harmonic of a predetermined critical component is abnormal and its frequency domain characteristics meet the preset late fault threshold condition, the predetermined critical component is determined to be in a late fault state, and a diagnostic report containing the equipment information, warning time, abnormal location and level of the predetermined critical component is generated, M>N; in specific implementation, M can be set to be greater than or equal to 5; the preset late fault threshold condition can be set according to actual needs, and is not specifically limited in this invention;

[0155] System early warning module: On the edge side, the diagnostic report and frequency domain characteristic values ​​(including frequency center, mean square frequency and frequency variance, as well as the fault main frequency of the lead screw, the fault main frequency of the lead screw nut and the fault main frequency of the ball) output by the fault diagnosis algorithm module are transmitted to the cloud data center through the Kafka platform. The system early warning module obtains data from the cloud data center, draws the spectrum diagram of the lead screw, nut and ball, and displays information such as whether there is a fault at different locations, the severity of the fault, and frequency domain characteristics.

[0156] Example 3

[0157] This invention provides a component fault diagnosis system based on data and mechanism fusion, including a storage medium and a processor;

[0158] The storage medium is used to store instructions;

[0159] The processor is configured to operate according to the instructions to execute the method according to any one of Embodiment 1.

[0160] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0161] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0162] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0163] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0164] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

[0165] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A component fault diagnosis method based on data and mechanism fusion, characterized in that, include: Obtain the vibration acceleration and rotational speed of the key components of the lead screw pair; Fourier transform is performed on the vibration acceleration of the predetermined key components of the lead screw pair to obtain the corresponding frequency domain data, and the frequency domain characteristics of the predetermined key components of the lead screw pair are extracted from the frequency domain data. Using the physical parameters of the lead screw pair and the rotational speed of the lead screw, the failure frequency of the predetermined key components of the lead screw pair is calculated; Fault diagnosis of key components of the lead screw pair is performed based on the frequency domain characteristics and fault frequency of the key components.

2. The component fault diagnosis method based on data and mechanism fusion according to claim 1, characterized in that: The frequency domain features include the frequency center, mean square frequency, and frequency variance; The formula for calculating the frequency center is: ， In the formula, As the frequency center, This represents the number of spectral lines. For the first Frequency values ​​of the spectral lines for The corresponding amplitude; The formula for calculating the mean square frequency is: ， In the formula, The mean square frequency; The formula for calculating the frequency variance is: ， In the formula, Here is the frequency variance.

3. The component fault diagnosis method based on data and mechanism fusion according to claim 1, characterized in that: The predetermined key components include lead screw nuts, lead screws, and / or ball bearings.

4. The component fault diagnosis method based on data and mechanism fusion according to claim 3, characterized in that: The formula for calculating the fault frequency of the lead screw nut is: ， The formula for calculating the fault frequency of the leadscrew is: ， The formula for calculating the failure frequency of the ball bearing is: ， In the formula, This refers to the number of balls; The frequency of the lead screw rotation is calculated from the lead screw's rotational speed. The diameter of the ball bearing is... The diameter of the lead screw pitch circle. It represents the contact angle.

5. The component fault diagnosis method based on data and mechanism fusion according to claim 1, characterized in that: The fault diagnosis of predetermined key components based on the frequency domain characteristics and fault dominant frequency of predetermined key components includes: When the amplitude of a predetermined critical component at its fault frequency is abnormal and its frequency domain characteristics meet the preset early fault threshold conditions, the predetermined critical component is determined to be in an early fault state, and a diagnostic report containing the equipment information, warning time, abnormal location and level of the predetermined critical component is generated.

6. The component fault diagnosis method based on data and mechanism fusion according to claim 1, characterized in that: The fault diagnosis of predetermined key components based on the frequency domain characteristics and fault dominant frequency of predetermined key components includes: When the amplitude of the Nth harmonic of a predetermined critical component is abnormal and its frequency domain characteristics meet the preset mid-term fault threshold condition, the predetermined critical component is determined to be in a mid-term fault state, and a diagnostic report containing the equipment information, warning time, abnormal location and level of the predetermined critical component is generated, where N>1.

7. A component fault diagnosis method based on data and mechanism fusion according to claim 6, characterized in that: The fault diagnosis of predetermined key components based on the frequency domain characteristics and fault dominant frequency of predetermined key components includes: When the amplitude of the Mth harmonic of a predetermined critical component is abnormal and its frequency domain characteristics meet the preset late fault threshold condition, the predetermined critical component is determined to be in a late fault state, and a diagnostic report containing the equipment information, warning time, abnormal location and level of the predetermined critical component is generated, where M>N.

8. The component fault diagnosis method based on data and mechanism fusion according to claim 1, characterized in that: The component fault diagnosis method further includes: preprocessing the obtained vibration acceleration and rotational speed of the predetermined key components of the lead screw pair, specifically including: The vibration acceleration of the key components of the lead screw and the rotational speed of the lead screw are obtained by wavelet transform to remove high-frequency noise and obtain the noise-reduced data. The denoised data is processed using a high-pass filter to remove ultra-low frequency interference components, resulting in data after secondary denoising. Linear interpolation is used to replace outliers in the data after secondary denoising, resulting in preprocessed data.

9. A component fault diagnosis system based on data and mechanism fusion, characterized in that, include: Vibration acceleration sensor, used to acquire the vibration acceleration of predetermined key components of the lead screw pair; A speed sensor is used to obtain the speed of the leadscrew; The controller, connected to the vibration acceleration sensor and the rotation speed sensor respectively, includes: The data acquisition module is used to acquire the vibration acceleration of key components of the lead screw pair and the rotational speed of the lead screw; The feature value preparation module is used to perform Fourier transform on the vibration acceleration of the predetermined key components of the lead screw pair to obtain the corresponding frequency domain data, and extract the preset frequency domain features from the frequency domain data; and calculate the fault main frequency of the predetermined key components of the lead screw pair using the physical parameters of the lead screw pair and the rotational speed of the lead screw. The fault diagnosis algorithm module is used to diagnose faults in the predetermined key components of the lead screw pair based on the frequency domain characteristics and the main fault frequency.

10. A component fault diagnosis system based on data and mechanism fusion, characterized in that, Including storage media and processor; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the method according to any one of claims 1-8.

Images