Real-time monitoring and damage evaluation system and method for rotor surface magnetic field characteristics

Abstract

The invention discloses a rotor surface magnetic field characteristic real-time monitoring and damage assessment system and method, and relates to the technical field of damage assessment. According to the method, a rotor surface magnetic field is monitored in real time under different rotating speed conditions, detailed magnetic field change data information is collected, the monitoring means can dynamically capture the magnetic field fluctuation condition, and the damage assessment accuracy is improved. According to the method, the damage prediction model is constructed by using the deep learning technology, the relevance between the damage prediction model and the rotating speed of the rotor is analyzed, early detection and classified evaluation of the unbalance condition are ensured, accurate data support is provided for further damage analysis, and accurate evaluation of the surface damage of the rotor is realized by using the deep learning technology to construct the damage prediction model and combining the unbalance type and the magnetic field change data. Meanwhile, the damage evaluation index is subjected to linear normalization processing, the influence between different data dimensions can be eliminated, and the evaluation result is more stable and reliable. And whether the damage state of the rotor surface is abnormal or not is judged in time by comparing and analyzing the damage evaluation index and a threshold value.

Description

Technical Field

[0001] The present invention relates to the field of damage assessment technology, and in particular to a rotor surface magnetic field characteristic real-time monitoring and damage assessment system and method. Background Art

[0002] In large rotating machinery (such as generators, steam turbines and motors), the rotor is a key component, and the stability of its operating state directly affects the overall performance and safety of the equipment. Therefore, the surface magnetic field characteristics monitoring and damage assessment of the rotor are crucial to ensure the stable operation of the equipment. During the operation of the rotor, due to long-term operation and load changes, problems such as imbalance, vibration and structural damage may occur. Therefore, real-time monitoring of the magnetic field characteristics of the rotor surface and obtaining damage data are of great significance for achieving efficient maintenance of the rotor and extending its service life.

[0003] In the current rotor damage detection technology, although there are many methods applied to rotor imbalance detection and vibration monitoring, the existing monitoring methods often only focus on static imbalance or dynamic performance under a single frequency, and it is difficult to accurately capture the complex changes of multiple states, especially the identification of minor rotor damage, potential dynamic imbalance and other phenomena is not accurate enough, resulting in limited early warning capabilities. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a system and method for real-time monitoring and damage assessment of magnetic field characteristics on the rotor surface, which solves the problems in the above-mentioned background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: A method for real-time monitoring and damage assessment of magnetic field characteristics on the rotor surface, comprising the following steps: S1. Use laser displacement measurement method to check the rotor imbalance type in advance. If the rotor is in static imbalance, perform correction work; S2. After the calibration operation, the magnetic field characteristics of the rotor surface at different speeds are monitored in real time to obtain relevant magnetic field change data information. Based on the relevant magnetic field change data information, the correlation between the magnetic field fluctuation and the rotor speed is analyzed. If there is a correlation, whether the magnetic field fluctuation is caused by the imbalance of the rotor is further analyzed to obtain the imbalance coefficient Shxs. Based on the value of the imbalance coefficient Shxs, the imbalance type of the rotor is determined; S3, using deep learning technology to build a damage prediction model, and based on the rotor imbalance type and the related magnetic field change data information obtained in S2, evaluate the damage of the rotor surface, and after linear normalization processing, construct a damage assessment index Spzs; S4. Preset an assessment threshold value Y and compare and analyze it with the damage assessment index Spzs to estimate the damage condition of the current rotor surface, and take corresponding adjustment measures based on the estimation result.

[0006] Preferably, the specific steps of S1 include: S11, pre-fixing a laser displacement sensor on the side of the rotor, aligning the laser displacement sensor with a point on the rotor surface, and marking it as a mark point I, taking the mark point I as a reference point, rotating the rotor, and collecting a time domain signal through the laser displacement sensor during the rotation process, and analyzing the change of the position of the time domain signal relative to the mark point I over time, wherein the time domain signal includes time information, displacement data, waveform characteristics and signal offset; S12, performing Fourier transform on the time domain signal collected by the laser displacement sensor to generate a spectrum diagram, and searching for a frequency peak with the same frequency as the rotation speed on the spectrum diagram. If there is a main frequency component with the same frequency as the rotation speed on the spectrum diagram, it is preliminarily determined that the current rotor is in a static unbalanced state; S13. When the rotor is in a static unbalanced state, the rotor will be corrected by adding a counterweight.

[0007] Preferably, the specific steps of S2 include: S21. After the calibration operation, real-time monitoring of the magnetic field characteristics of the rotor surface at different speeds of the rotor is performed to obtain relevant magnetic field change data information, wherein the relevant magnetic field change data information includes the magnetic field intensity at various positions on the rotor surface at different speeds of the rotor. ,frequency The corresponding frequency domain amplitude , the maximum amplitude in the frequency domain , the phase of magnetic field fluctuations , the phase of the speed signal , vibration amplitude And leakage magnetic field strength .

[0008] Preferably, the specific step S2 also includes: S22. Based on the relevant magnetic field change data information, analyze the correlation between magnetic field fluctuations and rotor speed to obtain the influencing factors , the impact factor Obtained by the following formula: ; In the formula, Expressed as Speed ​​value, Expressed as The No. 1 magnetic field fluctuation factor under speed conditions, Expressed as the standard deviation of the speed values, Expressed as the standard deviation of the No. 1 magnetic field fluctuation factor, Expressed as and The covariance of Indicated as different speed values Serial number; S23, No. No. 1 magnetic field fluctuation factor under speed conditions Obtained by the following formula: ; In the formula, Indicates The average magnetic field strength under the speed condition, Expressed as the number of different speed conditions, , It represents the average magnetic field strength under all speed conditions; If the impact factor If it falls within the preset threshold, it will be judged that there is no correlation between the current magnetic field fluctuation and the rotor speed; If the impact factor If it does not fall within the preset threshold, it will be determined that there is a correlation between the current magnetic field fluctuation and the rotor speed, and a dynamic imbalance type analysis instruction will be issued.

[0009] Preferably, the specific step S2 also includes: S24, after receiving the dynamic unbalance type analysis instruction obtained in S23, analyzing whether the magnetic field fluctuation is caused by the imbalance of the rotor to obtain the unbalance coefficient Shxs, wherein the unbalance coefficient Shxs is obtained by the following formula: ; In the formula, Expressed as the impact factor, Expressed as frequency The corresponding frequency domain amplitude is, is shown as the maximum amplitude in the frequency domain, Expressed as the absolute value of the phase difference, , and are weight values, among which , and The specific value is set by the user according to the situation.

[0010] Preferably, the specific step S2 also includes: S25, presetting the imbalance threshold K, and comparing the imbalance coefficient Shxs with the imbalance threshold K to determine whether the current magnetic field fluctuation is caused by the imbalance of the rotor, the specific contents are as follows: If the imbalance coefficient Shxs exceeds the imbalance threshold K, it will be determined that the current magnetic field fluctuation is caused by the imbalance of the rotor, and the current rotor will be determined to be a dynamic imbalance type again; If the imbalance coefficient Shxs does not exceed the imbalance threshold K, it is determined that the current magnetic field fluctuation is not caused by the imbalance of the rotor.

[0011] Preferably, the specific steps of S3 include: S31. When the rotor is of a dynamically unbalanced type, analyze the characteristics of the magnetic field on the rotor surface, analyze the damage on the rotor surface, and use deep learning technology to build an initial model. Input the relevant magnetic field change data information into the initial model and perform training and testing. Then use the trained initial model as a state recognition model, obtain feature information in the state recognition model, and train and test the state recognition model with the acquired feature information. Combined with the fact that the rotor is currently in a dynamically unbalanced type, use the trained state recognition model as a damage prediction model. According to the damage prediction model, fit and output the damage assessment index Spzs under the corresponding speed conditions.

[0012] Preferably, the specific step S3 also includes: S32, the damage assessment index Spzs under the corresponding speed condition is obtained by the following formula: ; In the formula, Expressed as the second magnetic field fluctuation factor, Expressed as the vibration amplitude, Expressed as leakage magnetic field intensity, and are weight values, Expressed as a correction constant, and The specific value is set by the user according to the situation, and ln is the symbol of natural logarithm.

[0013] Preferably, the specific steps of S4 include: S41, by comparing and analyzing the damage assessment index Spz with the assessment threshold value Y, to estimate the damage condition of the current rotor surface, the specific contents are as follows: If the damage assessment index Spz exceeds the assessment threshold value Y, it is estimated that the damage condition of the current rotor surface is in an abnormal state, and it is necessary to quickly accelerate to above the resonance frequency when the device is started; If the damage assessment index Spz does not exceed the assessment threshold value Y, it is estimated that the damage condition of the current rotor surface is not in an abnormal state, and the real-time monitoring operation continues.

[0014] A rotor surface magnetic field characteristics real-time monitoring and damage assessment system, comprising an initial type judgment module, a secondary type judgment module, a damage assessment module and a feedback module; The initial type judgment module is used to pre-check the rotor imbalance type by using a laser displacement measurement method, and if the rotor is in a static imbalance state, a correction operation is performed; The secondary type judgment module is used to monitor the magnetic field characteristics of the rotor surface under different rotation speeds in real time after the calibration operation to obtain relevant magnetic field change data information, and analyze the correlation between the magnetic field fluctuation and the rotor speed based on the relevant magnetic field change data information. If there is a correlation, continue to analyze whether the magnetic field fluctuation is caused by the imbalance of the rotor to obtain the imbalance coefficient Shxs, and judge the imbalance type of the rotor based on the value of the imbalance coefficient Shxs; The damage assessment module is used to construct a damage prediction model using deep learning technology, and to evaluate the damage condition of the rotor surface based on the acquired rotor imbalance type and the related magnetic field change data information, so as to construct a damage assessment index Spzs; The feedback module is used to pre-set the evaluation threshold value Y, and compare and analyze it with the damage evaluation index Spzs to estimate the damage condition of the current rotor surface, and take corresponding adjustment measures based on the estimation result.

[0015] The present invention provides a rotor surface magnetic field characteristics real-time monitoring and damage assessment system and method, which has the following beneficial effects: (1) First, the rotor imbalance type is checked by laser displacement measurement, which can accurately detect the static imbalance of the rotor and immediately perform correction work after confirming the existence of static imbalance. This method ensures accurate correction in the initial stage and provides a stable basis for subsequent damage monitoring and evaluation. After completing the correction work, the method collects detailed magnetic field change data information by real-time monitoring of the magnetic field on the rotor surface under different speed conditions. This monitoring method can dynamically capture the magnetic field fluctuation and analyze its correlation with the rotor speed, ensuring the early discovery and classification evaluation of the imbalance condition, and providing accurate data support for further damage analysis. The present invention uses deep learning technology to construct a damage prediction model, combined with the imbalance type and magnetic field change data, to achieve accurate evaluation of rotor surface damage. At the same time, the damage assessment index is linearly normalized to eliminate the influence between different data dimensions, making the evaluation result more stable and reliable. By comparing and analyzing the damage assessment index with the pre-set evaluation threshold, it is possible to timely determine whether the damage state of the rotor surface is abnormal. When the damage index exceeds the threshold, the system can quickly take corresponding adjustment measures to effectively avoid damage expansion and ensure the safety and operation stability of the equipment. In summary, this method realizes intelligent management of rotor damage based on multi-level monitoring and evaluation, thereby improving the accuracy and real-time performance of damage detection, further improving the safe operation level of the equipment, and extending the service life of the rotor.

[0016] (2) This method analyzes the correlation between magnetic field fluctuation and rotor speed by calculating the value of the influencing factor. The calculation formula of the influencing factor can quantify the degree of influence between magnetic field fluctuation and rotor through statistical methods such as covariance and standard deviation of speed and magnetic field fluctuation factors. This correlation analysis method not only improves the accuracy of dynamic imbalance identification, but also reduces the possibility of misjudgment.

[0017] (3) After the rotor detects the dynamic imbalance state, the method constructs an initial model through deep learning technology, and trains and tests the model in combination with the magnetic field change data to ensure that the model can accurately identify the changes in the magnetic field characteristics of the rotor surface. After multiple trainings, the initial model is upgraded to a state recognition model, and key feature information is further extracted and reinforced training is performed. Finally, the model is upgraded to a damage prediction model, and the damage assessment index under different speed conditions is fitted. This modeling process based on deep learning can capture complex nonlinear characteristics, thereby effectively improving the accuracy of damage assessment and enabling the system to adapt to variable rotor working conditions. After the damage prediction model generates the damage assessment index, it can automatically determine the state of the rotor surface damage by comparing it with the preset assessment threshold. When it exceeds the threshold, the system automatically identifies it as an abnormal state and triggers the preset response strategy. During the startup or shutdown process, the strategy of quickly passing through the resonance zone is adopted to ensure that the equipment quickly passes through the dangerous area and avoids staying near the resonance frequency for too long, thereby reducing the risk of damage caused by vibration. This intelligent identification and response mechanism ensures that the equipment can quickly take corresponding measures in an abnormal state, effectively improving the ability to prevent faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A schematic flow chart of a method for real-time monitoring and damage assessment of rotor surface magnetic field characteristics according to the present invention; Figure 2 This is a block diagram of a rotor surface magnetic field characteristic real-time monitoring and damage assessment system of the present invention. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] Example 1 See also Figure 1 The present invention provides a method for real-time monitoring and damage assessment of rotor surface magnetic field characteristics, comprising the following steps: S1. Use laser displacement measurement method to check the rotor imbalance type in advance. If the rotor is in static imbalance, perform correction work; S2. After the calibration operation, the magnetic field characteristics of the rotor surface at different speeds are monitored in real time to obtain relevant magnetic field change data information. Based on the relevant magnetic field change data information, the correlation between the magnetic field fluctuation and the rotor speed is analyzed. If there is a correlation, whether the magnetic field fluctuation is caused by the imbalance of the rotor is further analyzed to obtain the imbalance coefficient Shxs. Based on the value of the imbalance coefficient Shxs, the imbalance type of the rotor is determined; S3, using deep learning technology to build a damage prediction model, and based on the rotor imbalance type and the related magnetic field change data information obtained in S2, evaluate the damage of the rotor surface, and after linear normalization processing, construct a damage assessment index Spzs; S4. Preset an assessment threshold value Y and compare and analyze it with the damage assessment index Spzs to estimate the damage condition of the current rotor surface, and take corresponding adjustment measures based on the estimation result.

[0021] In this embodiment, in step S1, the rotor is checked for imbalance types by laser displacement measurement, and correction operations are performed when static imbalance is found. This process can effectively reduce the vibration and mechanical stress caused by imbalance, ensure the stability of rotor operation, and extend the service life of the equipment. At the same time, through accurate displacement measurement, signs of imbalance can be found at an early stage to prevent further expansion. In step S2, by real-time monitoring of the magnetic field characteristics of the rotor surface at different speeds, detailed magnetic field change data information is obtained, and the correlation between magnetic field fluctuations and rotor speed is analyzed. If the magnetic field fluctuation is related to the speed, it is further determined whether it is caused by different types of rotor imbalance (such as static imbalance or dynamic imbalance). This method can accurately identify the source and impact of imbalance, making the rotor imbalance diagnosis more comprehensive and targeted, and avoiding the limitations of single analysis of vibration signals. In step S3, a damage prediction model is constructed by deep learning technology, and the damage situation is evaluated based on the imbalance type of the rotor and related magnetic field change data information. The deep learning model can extract potential features from complex magnetic field and imbalance data to perform accurate damage prediction. The damage assessment index is generated through linear normalization processing. The index can reflect the damage degree of the rotor, improve the accuracy and reliability of diagnosis, and provide technical support for the intelligent monitoring of the equipment operation status. In step S4, the assessment threshold is set and compared with the damage assessment index to achieve automated damage assessment. When the damage assessment index exceeds the threshold, the system automatically prompts the abnormal state and reduces the risk of damage expansion through corresponding adjustment measures (such as the strategy of quickly passing through the resonance zone). This method not only improves the real-time performance of damage detection, but also can actively warn before an abnormality occurs, ensuring the safety and stability of the equipment.

[0022] Example 2 Please refer to Figure 1 , specifically: S1 specific steps include: S11, pre-fixing a laser displacement sensor on the side of the rotor, aligning the laser displacement sensor with a point on the rotor surface, and marking it as a mark point I, taking the mark point I as a reference point, rotating the rotor, and collecting a time domain signal through the laser displacement sensor during the rotation process, and analyzing the change of the position of the time domain signal relative to the mark point I over time, wherein the time domain signal includes time information, displacement data, waveform characteristics and signal offset; Among them, time information includes: time axis: the horizontal axis of the time domain signal is time, which indicates the specific time point in the acquisition process. The time information can be absolute time (such as seconds, milliseconds, etc.) or relative time (such as the time point relative to the start time of the measurement); sampling frequency: the sampling frequency (or sampling rate) refers to the number of signal samples collected per second, usually expressed in Hertz (Hz). For example, a sampling frequency of 1kHz means that 1000 data points are collected per second. The sampling frequency affects the resolution and accuracy of the signal.

[0023] Displacement data includes: displacement: the core data measured by the laser displacement sensor is the displacement of the object at a certain moment, that is, the offset of the object relative to the reference position. The displacement is usually measured in millimeters, microns or nanometers; instantaneous displacement value: the vertical axis of the time domain signal displays the displacement value at each time point, that is, the instantaneous displacement value. Each instantaneous displacement value represents the relative position of the object at a specific moment.

[0024] The waveform characteristics include: signal waveform: the curve formed by the displacement data recorded by the laser displacement sensor over time is the time domain waveform, through which the displacement trend over time can be observed; waveform amplitude: the amplitude of the waveform represents the maximum and minimum values ​​of the displacement signal, representing the offset range of the object relative to the initial position. Waveform period (if any): if the displacement of the object shows periodic changes (such as rotor imbalance), repeated periodic fluctuations can be observed in the waveform.

[0025] Signal offset means that in some cases, the displacement signal may contain an offset value, which represents the initial position or reference position of the object. The offset of the signal can be removed through data processing to more clearly analyze the actual displacement fluctuations.

[0026] S12, by performing Fourier transform (FFT) on the time domain signal collected by the laser displacement sensor to generate a spectrum diagram, And look for the frequency peak with the same frequency as the rotation speed on the spectrum graph. If there is a main frequency component with the same frequency as the rotation speed (i.e. 1 times the frequency) on the spectrum graph, it is preliminarily judged that the current rotor is in a static unbalanced state, indicating that the rotor gravity center offset has caused periodic displacement fluctuations, indicating the existence of static unbalance; S13. When the rotor is in a static unbalanced state, the rotor will be corrected by adding a counterweight.

[0027] In this embodiment, the laser displacement sensor is used to accurately detect the type of rotor imbalance, which can effectively identify the static imbalance state in the early stage. With the help of the marking point as the reference point, the system can accurately capture the displacement fluctuation information relative to the reference during the rotation of the rotor. In this way, the static imbalance state can be timely judged and corrected when the imbalance has not significantly affected the operation of the equipment, effectively reducing the mechanical stress and vibration amplitude of the equipment and extending the service life of the equipment. By performing Fourier transform (FFT) on the time domain signal collected by the laser displacement sensor to generate a spectrum diagram, the frequency characteristics of the rotor imbalance can be efficiently identified, especially the detection of 1 times the frequency (i.e., the main frequency component with the same frequency as the rotation speed), which helps to confirm whether the rotor is statically unbalanced. The spectrum analysis method can clearly show the vibration frequency distribution of the rotor, making the diagnosis more intuitive and accurate, and ensuring the rapid judgment of the imbalance state. After confirming the static imbalance state, the counterweight method is automatically triggered to perform the correction operation. By adding a counterweight to the light side of the rotor to balance the center of gravity, the offset and vibration caused by the imbalance are reduced. This method does not require manual intervention and can automatically complete the correction after the system detects the imbalance, which greatly improves the operating stability of the equipment and reduces the operating risks caused by the imbalance. In summary, the present invention realizes efficient identification and correction of static imbalance of the rotor through steps such as laser displacement measurement, Fourier spectrum analysis and automatic counterweight correction, thereby greatly improving the safety and operating stability of the equipment and significantly reducing the loss and maintenance cost caused by imbalance.

[0028] Example 3 Please refer to Figure 1 , specifically: S2 specific steps include: S21. After the calibration operation, real-time monitoring of the magnetic field characteristics of the rotor surface at different speeds of the rotor is performed to obtain relevant magnetic field change data information, wherein the relevant magnetic field change data information includes the magnetic field intensity at various positions on the rotor surface at different speeds of the rotor. ,frequency The corresponding frequency domain amplitude , the maximum amplitude in the frequency domain , the phase of magnetic field fluctuations , the phase of the speed signal , vibration amplitude And leakage magnetic field strength .

[0029] The specific steps of S2 also include: S22. Based on the relevant magnetic field change data information, analyze the correlation between magnetic field fluctuations and rotor speed to obtain the influencing factors , the impact factor Obtained by the following formula: ; In the formula, Expressed as Speed ​​value, Expressed as The No. 1 magnetic field fluctuation factor under speed conditions, Expressed as the standard deviation of the speed values, Expressed as the standard deviation of the No. 1 magnetic field fluctuation factor, Expressed as and The covariance of Indicated as different speed values Serial number; S23, No. No. 1 magnetic field fluctuation factor under speed conditions Obtained by the following formula: ; In the formula, Indicates The average magnetic field strength under the speed condition, Expressed as the number of different speed conditions, , It represents the average magnetic field strength under all speed conditions; The magnetic field strength It can be monitored and acquired through fluxgate sensor; If the impact factor If it falls within the preset threshold, it will be judged that there is no correlation between the current magnetic field fluctuation and the rotor speed; If the impact factor If it does not fall within the preset threshold, it will be determined that there is a correlation between the current magnetic field fluctuation and the rotor speed, and a dynamic imbalance type analysis instruction will be issued.

[0030] In this embodiment, after the calibration operation, by real-time monitoring of the magnetic field characteristics of the rotor under different speed conditions, the magnetic field strength at each position on the rotor surface, the frequency domain amplitude corresponding to the frequency, the maximum frequency domain amplitude, the phase of the magnetic field fluctuation, the phase of the speed signal, the vibration amplitude and the leakage magnetic field strength and other multi-dimensional data information can be accurately obtained. These data information provide rich basic data for the subsequent magnetic field fluctuation analysis, making the rotor state evaluation more comprehensive and effectively improving the accuracy of monitoring. Through the analysis of relevant magnetic field change data, the influence factor is calculated and detected whether it falls within the preset threshold to determine whether the magnetic field fluctuation is related to the rotor speed. This data-based correlation analysis can effectively distinguish between normal fluctuations and unbalanced fluctuations of the rotor, avoid misjudgment, and further improve the accuracy of unbalanced state identification. In addition, this process adaptively judges the relationship between fluctuations and speed, so that the system has a higher level of intelligence. When it is detected that the influence factor exceeds the threshold, indicating that the magnetic field fluctuation is significantly correlated with the speed, the system can automatically issue a dynamic unbalance type analysis instruction. The automatic identification and feedback mechanism enables the equipment to quickly respond to possible dynamic unbalance situations without manual intervention, realizing timely warning and processing of potential faults, and effectively improving the fault response efficiency. In summary, the present invention realizes efficient and accurate monitoring of the rotor magnetic field characteristics and imbalance state through a combination of multi-parameter monitoring, correlation analysis and automatic identification feedback, significantly improves the operating reliability and safety of the equipment, and provides strong support for equipment maintenance and fault prevention.

[0031] Example 4 Please refer to Figure 1 Specifically: S2 includes the following specific steps: S24, after receiving the dynamic unbalance type analysis instruction obtained in S23, analyzing whether the magnetic field fluctuation is caused by the imbalance of the rotor to obtain the unbalance coefficient Shxs, wherein the unbalance coefficient Shxs is obtained by the following formula: ; In the formula, Expressed as the impact factor, Expressed as frequency The corresponding frequency domain amplitude is, is shown as the maximum amplitude in the frequency domain, Expressed as the absolute value of the phase difference, , and are weight values, among which , and The specific value is set by the user according to the situation; Reflects the frequency The ratio of the vibration amplitude at the point to the maximum vibration amplitude. The closer the ratio is to 1, the closer the frequency component is to the maximum vibration component; the smaller the ratio is, the weaker the vibration component of this frequency is. This ratio is used to analyze the relative contribution of the vibration intensity of a specific frequency in the overall vibration, and to help determine the significance of a certain frequency component.

[0032] The above frequency The corresponding frequency domain amplitude and the maximum amplitude in the frequency domain It can be monitored and obtained through an acceleration sensor or a laser vibrometer; Absolute value of phase difference It can be obtained by the following formula: ; The specific steps of S2 also include: S25, presetting the imbalance threshold K, and comparing the imbalance coefficient Shxs with the imbalance threshold K to determine whether the current magnetic field fluctuation is caused by the imbalance of the rotor, the specific contents are as follows: If the imbalance coefficient Shxs exceeds the imbalance threshold K, it will be determined that the current magnetic field fluctuation is caused by the imbalance of the rotor, and the current rotor will be determined to be a dynamic imbalance type again; If the imbalance coefficient Shxs does not exceed the imbalance threshold K, it is determined that the current magnetic field fluctuation is not caused by the imbalance of the rotor.

[0033] In this embodiment, the method can quantify the cause of imbalance by analyzing the imbalance coefficient using magnetic field fluctuation data after receiving the dynamic imbalance type analysis instruction. The calculation formula of the imbalance coefficient is based on the influencing factor, frequency domain amplitude, phase difference and its weight parameter, and comprehensively considers the vibration characteristics of the rotor at different frequencies, especially the ratio of the frequency corresponding amplitude to the maximum amplitude, which helps to identify the contribution of a specific frequency to the overall vibration, thereby accurately judging whether the source of the magnetic field fluctuation is an imbalance problem. This method further improves the accuracy of identifying dynamic imbalance and avoids misjudging fluctuations caused by other factors. By presetting the imbalance threshold and comparing it with the imbalance coefficient, the method of the present invention can determine in real time whether the magnetic field fluctuation is caused by rotor imbalance. When the imbalance coefficient exceeds the threshold, the system automatically determines that the magnetic field fluctuation is caused by dynamic imbalance and quickly prompts the dynamic imbalance type. This threshold comparison mechanism reduces the reliance on complex data analysis, realizes automated fault identification, and can respond quickly in the early stage of the fault, effectively preventing the aggravation of the imbalance problem.

[0034] Example 5 Please refer to Figure 1 , specifically: S3 specific steps include: S31. When the rotor is of a dynamically unbalanced type, analyze the characteristics of the magnetic field on the rotor surface, analyze the damage on the rotor surface, and use deep learning technology to build an initial model. Input the relevant magnetic field change data information into the initial model and perform training and testing. Then use the trained initial model as a state recognition model, obtain feature information in the state recognition model, and train and test the state recognition model with the acquired feature information. Combined with the fact that the rotor is currently in a dynamically unbalanced type, use the trained state recognition model as a damage prediction model. According to the damage prediction model, fit and output the damage assessment index Spzs under the corresponding speed conditions.

[0035] The specific steps of S3 also include: S32, the damage assessment index Spzs under the corresponding speed condition is obtained by the following formula: ; In the formula, Expressed as the second magnetic field fluctuation factor, Expressed as the vibration amplitude, Expressed as leakage magnetic field intensity, and are weight values, Expressed as a correction constant, and The specific value is set by the user according to the situation, and ln is the symbol of natural logarithm.

[0036] Magnetic field fluctuation factor No. 2 It can be obtained by the following formula: ; in, It represents the average magnetic field strength under the corresponding speed conditions, Indicates The magnetic field strength at the monitoring point, Expressed as the number of monitoring points on the rotor surface, ; Magnetic field fluctuation factor No. 2 Indicates the magnetic field fluctuation between various monitoring points of the rotor under corresponding speed conditions; Vibration Amplitude It can be monitored and obtained through acceleration sensors; Leakage magnetic field strength It can be monitored and acquired through Hall effect sensors or magnetoresistive sensors.

[0037] The specific steps of S4 include: S41, by comparing and analyzing the damage assessment index Spz with the assessment threshold value Y, to estimate the damage condition of the current rotor surface, the specific contents are as follows: If the damage assessment index Spz exceeds the assessment threshold value Y, it is estimated that the damage condition of the current rotor surface is in an abnormal state. When the device is started, it is necessary to quickly accelerate to above the resonance frequency to reduce the time in the resonance area, and quickly decelerate when shutting down to avoid staying in the resonance area for a long time. If the damage assessment index Spz does not exceed the assessment threshold value Y, it is estimated that the damage condition of the current rotor surface is not in an abnormal state, and the real-time monitoring operation continues.

[0038] During the startup or shutdown process, the rotor will inevitably pass through the resonant frequency range. In order to avoid staying at the resonant frequency for a long time, a strategy of quickly passing through the resonance zone can be adopted.

[0039] In this embodiment, the method constructs an initial model through deep learning technology, inputs a variety of rotor magnetic field change data into the model for training and testing, gradually optimizes to obtain a state recognition model, and further forms a final damage prediction model. The deep learning model can extract potential feature information from a large amount of complex data, and then accurately identify the damage to the rotor surface. This damage assessment method based on intelligent algorithms can significantly improve the accuracy and reliability of identification, and avoid the uncertainty of human judgment in traditional methods. Through the damage prediction model, the damage assessment index of the rotor under specific speed conditions is fitted and output, and combined with key parameters such as the second magnetic field fluctuation factor, vibration amplitude, and leakage magnetic field strength, a quantitative description of the damage state is achieved. This quantitative assessment method facilitates real-time monitoring of the health status of the rotor, ensuring that abnormal signals can be captured in the early stages of damage. After obtaining the damage assessment index, by comparing and analyzing it with the preset assessment threshold, the severity of the rotor damage can be accurately judged. When it exceeds the threshold, the system will immediately determine that the rotor damage is in an abnormal state and take corresponding countermeasures, such as quickly accelerating through the resonance zone when starting and quickly decelerating when shutting down, to avoid staying in the resonance zone for too long and reduce the risk of further damage. This intelligent early warning and rapid response mechanism can effectively prevent the vibration amplification problem caused by resonance and ensure the safe operation of the equipment.

[0040] Example 6 Please refer to Figure 2 ,Specifically: A rotor surface magnetic field characteristics real-time monitoring and damage assessment system, including an initial type judgment module, a secondary type judgment module, a damage assessment module and a feedback module; The initial type judgment module is used to pre-check the rotor imbalance type by using a laser displacement measurement method, and if the rotor is in a static imbalance state, a correction operation is performed; The secondary type judgment module is used to monitor the magnetic field characteristics of the rotor surface under different rotation speeds in real time after the calibration operation to obtain relevant magnetic field change data information, and analyze the correlation between the magnetic field fluctuation and the rotor speed based on the relevant magnetic field change data information. If there is a correlation, continue to analyze whether the magnetic field fluctuation is caused by the imbalance of the rotor to obtain the imbalance coefficient Shxs, and judge the imbalance type of the rotor based on the value of the imbalance coefficient Shxs; The damage assessment module is used to construct a damage prediction model using deep learning technology, and to evaluate the damage condition of the rotor surface based on the acquired rotor imbalance type and the related magnetic field change data information, so as to construct a damage assessment index Spzs; The feedback module is used to pre-set the evaluation threshold value Y, and compare and analyze it with the damage evaluation index Spzs to estimate the damage condition of the current rotor surface, and take corresponding adjustment measures based on the estimation result.

[0041] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for real-time monitoring and damage assessment of rotor surface magnetic field characteristics, characterized in that: The following steps are included: S1. Use laser displacement measurement method to check the rotor imbalance type in advance. If the rotor is in static imbalance, perform correction work; S2. After the calibration operation, the magnetic field characteristics of the rotor surface at different speeds are monitored in real time to obtain relevant magnetic field change data information. Based on the relevant magnetic field change data information, the correlation between the magnetic field fluctuation and the rotor speed is analyzed. If there is a correlation, whether the magnetic field fluctuation is caused by the imbalance of the rotor is further analyzed to obtain the imbalance coefficient Shxs. Based on the value of the imbalance coefficient Shxs, the imbalance type of the rotor is determined; S3, using deep learning technology to build a damage prediction model, and based on the rotor imbalance type and the related magnetic field change data information obtained in S2, evaluate the damage of the rotor surface, and after linear normalization processing, construct a damage assessment index Spzs; S4. Preset an assessment threshold value Y and compare and analyze it with the damage assessment index Spzs to estimate the damage condition of the current rotor surface, and take corresponding adjustment measures based on the estimation result.

2. A rotor surface magnetic field characteristics real-time monitoring and damage assessment method according to claim 1, characterized in that: The specific steps of S1 include: S11, pre-fixing a laser displacement sensor on the side of the rotor, aligning the laser displacement sensor with a point on the rotor surface, and marking it as a mark point I, taking the mark point I as a reference point, rotating the rotor, and collecting a time domain signal through the laser displacement sensor during the rotation process, and analyzing the change of the position of the time domain signal relative to the mark point I over time, wherein the time domain signal includes time information, displacement data, waveform characteristics and signal offset; S12, performing Fourier transform on the time domain signal collected by the laser displacement sensor to generate a spectrum diagram, and searching for a frequency peak with the same frequency as the rotation speed on the spectrum diagram. If there is a main frequency component with the same frequency as the rotation speed on the spectrum diagram, it is preliminarily determined that the current rotor is in a static unbalanced state; S13. When the rotor is in a static unbalanced state, the rotor will be corrected by adding a counterweight.

3. A rotor surface magnetic field characteristics real-time monitoring and damage assessment method according to claim 2, characterized in that: The specific steps of S2 include: S21. After the calibration operation, real-time monitoring of the magnetic field characteristics of the rotor surface at different speeds of the rotor is performed to obtain relevant magnetic field change data information, wherein the relevant magnetic field change data information includes the magnetic field intensity at various positions on the rotor surface at different speeds of the rotor. ,frequency The corresponding frequency domain amplitude , the maximum amplitude in the frequency domain , the phase of magnetic field fluctuations , the phase of the speed signal , vibration amplitude And leakage magnetic field strength .

4. A rotor surface magnetic field characteristics real-time monitoring and damage assessment method according to claim 3, characterized in that: The specific steps of S2 also include: S22. Based on the relevant magnetic field change data information, analyze the correlation between magnetic field fluctuations and rotor speed to obtain the influencing factors , the impact factor Obtained by the following formula: ； In the formula, Expressed as Speed ​​value, Expressed as The No. 1 magnetic field fluctuation factor under speed conditions, Expressed as the standard deviation of the speed values, Expressed as the standard deviation of the No. 1 magnetic field fluctuation factor, Expressed as and The covariance of Indicated as different speed values Serial number; S23, No. No. 1 magnetic field fluctuation factor under speed conditions Obtained by the following formula: ； In the formula, Indicates The average magnetic field strength under the speed condition, Expressed as the number of different speed conditions, , It represents the average magnetic field strength under all speed conditions; If the influencing factor Yxyz falls within the preset threshold, it will be judged that there is no correlation between the current magnetic field fluctuation and the rotor speed; If the influencing factor Yxyz does not fall within the preset threshold, it will be determined that there is a correlation between the current magnetic field fluctuation and the rotor speed, and a dynamic unbalance type analysis instruction will be issued.

5. A rotor surface magnetic field characteristics real-time monitoring and damage assessment method according to claim 4, characterized in that: The specific steps of S2 also include: S24, after receiving the dynamic unbalance type analysis instruction obtained in S23, analyzing whether the magnetic field fluctuation is caused by the imbalance of the rotor to obtain the unbalance coefficient Shxs, wherein the unbalance coefficient Shxs is obtained by the following formula: ； In the formula, Expressed as the impact factor, Expressed as frequency The corresponding frequency domain amplitude is, is shown as the maximum amplitude in the frequency domain, Expressed as the absolute value of the phase difference, , and are weight values, among which , and The specific value is set by the user according to the situation.

6. A rotor surface magnetic field characteristics real-time monitoring and damage assessment method according to claim 5, characterized in that: The specific steps of S2 also include: S25, presetting the imbalance threshold K, and comparing the imbalance coefficient Shxs with the imbalance threshold K to determine whether the current magnetic field fluctuation is caused by the imbalance of the rotor, the specific contents are as follows: If the imbalance coefficient Shxs exceeds the imbalance threshold K, it will be determined that the current magnetic field fluctuation is caused by the imbalance of the rotor, and the current rotor will be determined to be a dynamic imbalance type again; If the imbalance coefficient Shxs does not exceed the imbalance threshold K, it is determined that the current magnetic field fluctuation is not caused by the imbalance of the rotor.

7. A rotor surface magnetic field characteristics real-time monitoring and damage assessment method according to claim 6, characterized in that: The specific steps of S3 include: S31. When the rotor is of a dynamically unbalanced type, analyze the characteristics of the magnetic field on the rotor surface, analyze the damage on the rotor surface, and use deep learning technology to build an initial model. Input the relevant magnetic field change data information into the initial model and perform training and testing. Then use the trained initial model as a state recognition model, obtain feature information in the state recognition model, and train and test the state recognition model with the acquired feature information. Combined with the fact that the rotor is currently in a dynamically unbalanced type, use the trained state recognition model as a damage prediction model. According to the damage prediction model, fit and output the damage assessment index Spzs under the corresponding speed conditions.

8. A rotor surface magnetic field characteristics real-time monitoring and damage assessment method according to claim 7, characterized in that: The specific steps of S3 also include: S32, the damage assessment index Spzs under the corresponding speed condition is obtained by the following formula: ； In the formula, Expressed as the second magnetic field fluctuation factor, Expressed as the vibration amplitude, Expressed as leakage magnetic field intensity, and are weight values, Expressed as a correction constant, and The specific value is set by the user according to the situation, and ln is the symbol of natural logarithm.

9. A rotor surface magnetic field characteristics real-time monitoring and damage assessment method according to claim 1, characterized in that: The specific steps of S4 include: S41, by comparing and analyzing the damage assessment index Spz with the assessment threshold value Y, to estimate the damage condition of the current rotor surface, the specific contents are as follows: If the damage assessment index Spz exceeds the assessment threshold value Y, it is estimated that the damage condition of the current rotor surface is in an abnormal state, and it is necessary to quickly accelerate to above the resonance frequency when the device is started; If the damage assessment index Spz does not exceed the assessment threshold value Y, it is estimated that the damage condition of the current rotor surface is not in an abnormal state, and the real-time monitoring operation continues.

10. A rotor surface magnetic field characteristic real-time monitoring and damage assessment system, used to implement a rotor surface magnetic field characteristic real-time monitoring and damage assessment method as claimed in any one of claims 1 to 9, characterized in that: It includes an initial type judgment module, a secondary type judgment module, a damage assessment module and a feedback module; The initial type judgment module is used to pre-check the rotor imbalance type by using a laser displacement measurement method, and if the rotor is in a static imbalance state, a correction operation is performed; The secondary type judgment module is used to monitor the magnetic field characteristics of the rotor surface under different rotation speeds in real time after the calibration operation to obtain relevant magnetic field change data information, and analyze the correlation between the magnetic field fluctuation and the rotor speed based on the relevant magnetic field change data information. If there is a correlation, continue to analyze whether the magnetic field fluctuation is caused by the imbalance of the rotor to obtain the imbalance coefficient Shxs, and judge the imbalance type of the rotor based on the value of the imbalance coefficient Shxs; The damage assessment module is used to construct a damage prediction model using deep learning technology, and to evaluate the damage condition of the rotor surface based on the acquired rotor imbalance type and the related magnetic field change data information, so as to construct a damage assessment index Spzs; The feedback module is used to pre-set the evaluation threshold value Y, and compare and analyze it with the damage evaluation index Spzs to estimate the damage condition of the current rotor surface, and take corresponding adjustment measures based on the estimation result.

Images