Rubbing fault identification method based on harmonic amplitude

Abstract

The invention discloses a rub-impact fault identification method based on harmonic amplitude, and relates to the technical field of rub-impact fault diagnosis based on vibration, the method realizes the construction of a rub-impact fault identification index based on the processing of the harmonic amplitude, and the constructed rub-impact fault identification index has strong fault directivity. The rub-impact fault identification method based on the index has good practicability when being used for rub-impact fault identification. In order to achieve the purpose, according to the technical scheme, the method comprises the following steps that firstly, vibration monitoring data of the aero-engine in the running state are collected through a vibration sensor; secondly, converting the vibration monitoring data into a frequency domain through Fourier transform to obtain a spectrogram; thirdly, extracting a harmonic amplitude in the spectrogram, and constructing a rub-impact fault identification index Ca based on the harmonic amplitude; and finally, whether the equipment has a rub-impact fault or not is identified through Ca.

Description

Technical Field

[0001] The invention relates to the technical field of vibration-based rubbing fault diagnosis, and in particular to a rubbing fault identification method based on harmonic amplitude. Background Art

[0002] The rotor system is the core of high-end rotating machinery. Accurate status assessment and fault diagnosis are key to ensuring the normal operation of the equipment. In order to improve the performance and efficiency of the equipment, the rotor system with a large number of blades is placed in a casing with a very small gap with it, which makes it very easy for the blades and casing to come into contact, causing friction failures and affecting the normal operation of the equipment. In order to ensure the safety and stable production of high-end rotating machinery, it is extremely important to study the friction failure mechanism and then construct indicators to identify the system status.

[0003] The rubbing fault diagnosis indicators currently studied and actually used can be divided into two categories according to the input parameters: time domain indicators and frequency domain indicators.

[0004] In terms of time domain indicators, in addition to the vibration effective value ( RMS In addition to the energy method, it has also attracted more and more attention. By combining vibration signals with other parameters, characteristic indicators with physical significance are constructed, which provides an effective way for rubbing fault diagnosis.

[0005] The existing time domain indicators are mainly constructed based on the time domain waveform of the vibration signal, so they contain all the vibration information. Although this method can sensitively reflect the occurrence of rubbing faults, it will also be affected by other typical rotor faults (such as imbalance and misalignment), resulting in poor fault directivity of the time domain indicators and difficulty in achieving directional diagnosis of rubbing faults.

[0006] Among the existing frequency domain indicators: (1) Diagnosis of rubbing faults based on fault characteristic frequency. Since rubbing faults are typical nonlinear faults, when they occur, the system vibration will have a large number of frequency component changes in the frequency domain, including harmonic frequencies (1 X , 2 X , blade passing frequency, etc.), combined frequency (high and low pressure and differential frequency), modulation frequency (sideband around blade passing frequency) and natural frequency, etc. This results in that the diagnosis of rubbing faults by the change of a single frequency component has good directivity, but its sensitivity is extremely poor; (2) The nonlinear output frequency response function better integrates the characteristics of rubbing faults, and it will not be affected by other typical rotor faults. From a theoretical perspective, it is a better indicator for rubbing fault diagnosis. However, the nonlinear output frequency response function needs to be run twice with different inputs under the same operating conditions to be calculated, and a known input signal is required, but the input signal is difficult to obtain, which leads to a large limitation on the practicality of this method.

[0007] The existing solutions cannot solve the problem that the fault directivity of the rubbing fault identification index is not strong, so the identification of rubbing faults is not practical. Summary of the invention

[0008] In view of this, the present invention provides a rubbing fault identification method based on harmonic amplitude. The method realizes the construction of a rubbing fault identification index based on the processing of the harmonic amplitude. The constructed rubbing fault identification index has a strong fault directivity, and the use of the index for rubbing fault identification has good practicality.

[0009] To achieve the above-mentioned purpose, the technical solution of the present invention is as follows: first, collecting vibration monitoring data of an aircraft engine in operation through a vibration sensor; second, converting the vibration monitoring data into the frequency domain through Fourier transform to obtain a spectrum diagram; third, extracting the harmonic amplitude in the spectrum diagram, and constructing a rubbing fault identification index Ca based on the harmonic amplitude; finally, identifying whether a rubbing fault has occurred in the equipment through Ca.

[0010] The constructed rubbing fault identification index Ca It is: the sum of the accumulated value and the accumulated value of the multi-harmonic amplitude minus the maximum value of the harmonic amplitude, that is:

[0011] in For the i The harmonic amplitude of the order, N is the harmonic number of the indicator fusion.

[0012] Further, through Ca Identify whether the equipment has a friction fault, specifically: obtain the friction fault identification index at each moment Ca Value, if the current moment Ca If the value changes more than the set threshold value compared with the previous moment, it is determined that a friction fault has occurred at the current moment.

[0013] Furthermore, the specific process of constructing the rubbing fault identification index includes the following steps: S 1. Establish a rotor-support-casing model and obtain the dynamic equation of the model; S 2. Set up the blade-casing friction fault simulation method, through Lankarani - Nikravesh The friction force model calculates the friction force between the blade and the casing, and calculates the additional force on the rotor due to the friction failure through the blade deformation; S3. The dynamic equations of the model are numerically solved by the Runge-Kutta method, and the vibration characteristics of the casing obtained by the solution are analyzed to summarize the characteristic frequencies of the casing under the rubbing fault state; S 4. Based on the characteristic frequency of the casing under the rubbing fault state obtained by analysis, a fault identification index is constructed.

[0014] Further, S 2, a certain non-uniform gap is set between the blade and the casing, and its expression is: (1) in, represents the initial gap, Indicates the change gap, represents the gap angle, Indicates the gap range; subscript k Indicates the leaf number, Indicates k The angular position of the blades, Indicates k The non-uniform gap corresponding to the angular position of each blade.

[0015] When the relative displacement between the blade and the casing exceeds the gap, that is, contact occurs, the Lankarani - Nikravesh The radial friction force generated by the model calculation is expressed as: (2) in represents the friction stiffness, represents the coefficient of restitution, represents the penetration depth, represents the friction velocity, represents the initial rubbing velocity; Based on Coulomb's friction law and combined with the blade tip tangential velocity, the expression of the friction tangential force is obtained: (3) in They represent symbolic functions, is the blade tangential speed; At the same time, the disk is connected to the blade, and the blade is deformed by the friction force, which causes the disk to be subjected to an additional force. The additional force is transmitted by the blade deformation combined with the equivalent stiffness, and its expression is: (4) in represents the additional force acting on the disk, Indicates the number of disk blades, represents the equivalent stiffness of the disk blade in the chord direction, represents the equivalent axial stiffness of the disk blade, represents the normal equivalent stiffness of the disk blade; k The three coordinate transformation matrices generated by the gyroscopic effect of each blade are: , coordinate transformation matrix 2 And coordinate transformation matrix three , the gyroscopic effect causes the turntable to swing, and the force is adjusted back to the horizontal and vertical positions through three coordinate transformation matrices; , , They are the blade chord displacement, blade axial displacement and blade normal displacement respectively.

[0016] Further, S 3, the obtained vibration characteristics of the casing under the rubbing fault state are analyzed, and the characteristic frequencies of the casing under the rubbing fault state are summarized, specifically: Set different variable gaps to simulate different degrees of rubbing faults. The larger the variable gap, the more serious the rubbing fault. The variable gap is 0μ m That is the normal state; The obtained casing vibration signal is differentiated into an acceleration vibration signal and plotted into a spectrum diagram. The rotation frequency vibration is intercepted for comparison, and the first conclusion of the casing vibration characteristics is obtained: the rotation frequency amplitude in the casing vibration increases with the severity of the fault; Further, S 3, by comparing the harmonic frequency components, the second conclusion about the casing vibration characteristics is obtained: the harmonic components in the casing vibration are excited to different degrees due to the occurrence of friction faults, and as the degree of fault increases, the degree of change in the amplitude of each harmonic becomes more significant.

[0017] Beneficial effects: 1. The rubbing fault identification method based on harmonic amplitude provided by the present invention can comprehensively consider the various fault characteristic frequencies of rubbing faults in the frequency domain, isolate the influence of other typical rotor faults on the index values, and identify rubbing faults in a directional and quantitative manner. This method provides a technical basis and method guidance for the healthy operation of mechanical equipment. Compared with the existing methods, this method can monitor the system status in real time when the rotating equipment is in operation, realize real-time rubbing fault identification, and is more practical.

[0018] 2. The present invention realizes directional and quantitative identification of rubbing faults. The fault characteristic index constructed by the present invention can better isolate the interference of typical rotor faults represented by imbalance on the basis of realizing quantitative identification of rubbing faults.

[0019] 3. The present invention realizes the fusion of multiple fault features, and the indicator has strong sensitivity. The indicator aggregates the scattered fault features of rubbing faults by fusing multiple harmonic amplitudes. Compared with a single characteristic frequency, this indicator can more sensitively perceive the occurrence of rubbing faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A flow chart of a rubbing fault identification method based on harmonic amplitude provided by the present invention; Figure 2 The invention provides a construction index for identifying friction faults. Ca Flow chart of the method; Figure 3 In the embodiment of the present invention S Schematic diagram of the rotor-support-casing model established in 1; Figure 4 : is a diagram of the amplitude of the casing vibration intermediate frequency under different changing gaps in the embodiment of the present invention; Figure 5 a ) is the variable gap of 60 in the embodiment of the present invention μm Harmonic component diagram of the lower casing vibration spectrum; Figure 5 b ) is the variable gap of 80 in the embodiment of the present invention μm Harmonic component diagram of the lower casing vibration spectrum; Figure 6 This is a comparison chart of indicator values ​​under different fault types and degrees in an embodiment of the present invention; Figure 7 This is a comparison chart of indicator values ​​under different fault ranges in an embodiment of the present invention; Figure 8 It is a comparison chart of index values ​​under different rubbing frequencies in an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0022] The present invention provides a method for identifying rubbing faults based on harmonic amplitude, which can comprehensively consider the various fault characteristic frequencies of rubbing faults in the frequency domain, isolate the influence of other typical rotor faults on the index values, and identify rubbing faults in a directional and quantitative manner. The flow chart of the method is as follows: Figure 1 As shown in FIG, the method includes the following steps: first, the vibration monitoring data of the aircraft engine in operation is collected through a vibration sensor; second, the vibration monitoring data is converted into the frequency domain through Fourier transform to obtain a spectrum diagram; third, the harmonic amplitude in the spectrum diagram is extracted, and based on the harmonic amplitude, a friction fault identification index is constructed. Ca ; Finally, through Ca Identify whether the equipment has a friction fault; The constructed rubbing fault identification index Ca It is: the sum of the accumulated value and the accumulated value of the multi-harmonic amplitude minus the maximum value of the harmonic amplitude; that is:

[0023] in For the i The harmonic amplitude of the order, N is the harmonic number of the indicator fusion.

[0024] In the embodiment of the present invention, by Ca Identify whether the equipment has a friction fault, specifically: obtain the friction fault identification index at each moment Ca Value, if the current moment Ca If the value changes more than the previous moment and exceeds the set threshold, it is determined that a friction fault has occurred at the current moment. The threshold can be set based on experience.

[0025] This method provides a technical basis and methodological guidance for the healthy operation of mechanical equipment. Compared with existing methods, this method can monitor the system status in real time when the rotating equipment is in operation, and is more practical.

[0026] In the embodiment of the present invention, based on the harmonic amplitude, a friction fault identification index is constructed. Ca The main steps of construction are as follows Figure 2 As shown, the following steps are included: S 1. Establish a rotor-support-casing model and obtain the dynamic equation of the model; in the embodiment of the present invention, S The schematic diagram of the rotor-support-casing model established in 1 is as follows Figure 3 The model mainly includes disk, blades, bearings and casing. Considering the gyroscopic effect caused by the disk offset, the blades are constructed based on the plate hypothesis theory.

[0027] The structure corresponding to the rotor-support-casing model includes blades, a turntable, a rotor shaft, a support and a casing, wherein the casing is a cylindrical shell, and the blades, rotor and support are all arranged inside the casing. The casing is arranged horizontally, that is, its cylindrical center axis is parallel to the horizontal plane; a support is arranged on each side of the casing, and the support is arranged along the diameter of the two bottom circles of the cylinder, and the support is perpendicular to the center axis of the cylinder; the rotor shaft is arranged at the cylindrical center axis position of the casing, and the two ends of the rotor shaft are fixedly connected to the two supports; a turntable is arranged on the rotor shaft, and blades are evenly arranged on the edge of the turntable.

[0028] The rotor-support-casing model is constructed for the above structure as follows: Select the center of mass of one of the supports as the origin and the axial direction of the rotor as X The central axis of the rotor shaft coincides with the line connecting the centroids of the two supports; YOZ Plane perpendicular to X Axis, along the rotor radial direction, thus constructed OXYZ Coordinate system; the displacement vector of the turntable is , For the turntable Z Direction displacement, For the turntable Y Direction displacement, α For turntable edge Z The direction of the corner, β For turntable edge Y The rotation angle of the direction; the casing and the casing do not consider the rotation angle, and the support displacement vector is , For support n exist Z Direction displacement, For support n exist Y displacement in the Z direction and Y The rotation angles of the directions are all 0. n The values ​​are 1 and 2, respectively referring to the two supports; the displacement vector of the casing is , is the displacement of the casing in the Z direction, For the receiver Y Displacement in the direction of the casing Z Direction and Y The rotation angle of the direction is 0.

[0029] S 2. Set up the blade-casing friction fault simulation method, through Lankarani - Nikravesh The friction force model calculates the friction force and calculates the additional force on the rotor due to the friction fault through the blade deformation. In the embodiment of the present invention, S 2, a certain non-uniform gap is set between the blade and the casing, and its expression is: (1) In the formula, Respectively represent the initial gap, variable gap, gap angle, and gap range; subscript k Indicates the leaf number, Indicates k The angular position of the blades, Indicates k The non-uniform gap corresponding to the angular position of each blade.

[0030] When the relative displacement between the blade and the casing exceeds the gap, that is, contact occurs, the Lankarani - NikraveshThe radial friction force generated by the model calculation is expressed as: (2) In the formula represents the friction stiffness, represents the coefficient of restitution, represents the penetration depth, represents the friction velocity, Indicates the initial rubbing velocity.

[0031] Based on Coulomb's friction law and combined with the blade tip tangential velocity, the expression of the friction tangential force can be obtained: (3) in They represent symbolic functions, is the blade tangential velocity. At the same time, it is noted that the disk is connected to the blade, and the blade is deformed by the friction force, which in turn causes the disk to be subjected to an additional force. This force is transmitted by the blade deformation combined with the equivalent stiffness, and its expression is: (4) in represents the additional force acting on the disk, Indicates the number of disk blades, represents the equivalent stiffness of the disk blade in the chord direction, represents the equivalent axial stiffness of the disk blade, represents the normal equivalent stiffness of the disk blade; k The three coordinate transformation matrices generated by the gyroscopic effect of each blade are: , coordinate transformation matrix 2 And coordinate transformation matrix three , the gyroscopic effect causes the turntable to swing, and the force is adjusted back to the horizontal and vertical positions through three coordinate transformation matrices; , , They are the blade chord displacement, blade axial displacement and blade normal displacement respectively.

[0032] S 3. The dynamic equations of the model are numerically solved by the Runge-Kutta method, and the vibration characteristics of the casing obtained by the solution are analyzed to summarize the characteristic frequencies of the casing under the rubbing fault state; In the embodiment of the present invention, S The vibration characteristics of the casing under the rubbing fault state obtained in 3 are analyzed. Since the actual rotating equipment measurement can only measure the casing vibration, the simulation analysis mainly focuses on the casing response, such as Figure 3 Shown is the amplitude of the casing vibration intermediate frequency under different changing gaps.

[0033] Set different variable gaps to simulate different degrees of friction failure. The larger the variable gap, the more serious the friction failure. The variable gap is 0. μm The obtained casing vibration signal is differentiated into an acceleration vibration signal and plotted into a spectrum diagram. The rotation frequency vibration is intercepted for comparison, such as Figure 4 shown.

[0034] from Figure 4 The first conclusion that can be drawn about the vibration characteristics of the casing is that the frequency amplitude of the casing vibration increases with the severity of the fault.

[0035] Further comparison of harmonic frequency components is shown in Figure 5 ( a ) and Figure 5 ( b ), where Figure 5 ( a ) is the variable gap of 60 in the embodiment of the present invention μm Figure 5 (Figure 5) b ) is the variable gap of 80 in the embodiment of the present invention μm The harmonic component diagram of the vibration spectrum of the lower casing. From Figure 5 ( a ) and Figure 5 ( b ) The second conclusion about the casing vibration characteristics can be drawn. The harmonic components in the casing vibration will be excited to different degrees due to the occurrence of friction faults, and as the degree of fault increases, the degree of change in the amplitude of each harmonic will become more significant.

[0036] S 4. Based on the characteristic frequency of the casing under the friction fault state obtained by analysis, a fault identification index is constructed; In the embodiment of the present invention, S 4, based on S 3. The conclusion of the casing vibration characteristics is obtained, and the friction fault identification index is constructed Ca : (5) in For the i The harmonic amplitude of the order, N is the harmonic number of the indicator fusion.

[0037] The overall meaning of the indicator is the sum of the accumulated value and the multiplied value of the multiple harmonic amplitudes minus the maximum value of the harmonic amplitudes.

[0038] The construction of the index mainly considers the following three aspects: (1) Figure 4 It can be seen that the friction fault will cause the frequency amplitude of the casing to increase. If a single accumulation and subtraction of the maximum value (usually the frequency amplitude) is performed, the effect of the friction fault on the frequency amplitude will be ignored. Therefore, a multiplication term is added; (2) Combined Figure 4 Compared with Figure 5 ( a) and Figure 5 ( b ) and actual engineering conditions, the frequency amplitude is usually several times that of other harmonic components. If it is based solely on the cumulative term, the indicator will be easily affected by the maximum value (frequency) and the minimum value, resulting in unstable indicator values. Therefore, the cumulative term is added to stabilize the overall value; (3) Typical rotor faults usually mainly cause changes in the value of a single frequency component. For example, rotor imbalance faults will affect 1 X , misalignment failure will affect 2 X By subtracting the maximum value of the harmonic amplitude, the index amplitude is isolated from other faults, and the sensitivity of the index to rubbing faults that affect multiple harmonic components at the same time is amplified through the accumulation and multiplication terms.

[0039] S 5. Set up different forms of tests to verify the quantitative and directional identification capabilities of indicators for rubbing faults.

[0040] In the embodiment of the present invention, S 5, two aspects of the test content are carried out, which shows that the constructed indicators can quantitatively and directionally identify rubbing faults.

[0041] Preset harmonic number N =5, by changing the mass eccentricity in the model, the characteristic index calculation results under different degrees of unbalanced faults are calculated, and compared with the index values ​​under different degrees of rubbing faults. The comparison results are shown in Figure 6 shown. Figure 6 In the figure, the degree of unbalance 1 fault is greater than that of unbalance 2 (mass eccentricity is larger), and the degree of rubbing 2 fault is greater than that of rubbing 1 (uneven gap setting is larger). It can be clearly seen from the comparison in the figure that when a rubbing fault occurs, the index value will change significantly, while the index value under an unbalanced fault is not much different from the normal state. At the same time, the change in the index value also corresponds well to the change in the degree of rubbing fault, that is, the more serious the rubbing fault is, the larger the index value is.

[0042] By changing the friction range, the indicator values ​​under different friction fault ranges are compared. The comparison results are as follows: Figure 7 shown. Figure 7 In the example, the rubbing range is set to 5° to simulate single-point rubbing; the rubbing range is set to 20° to simulate local rubbing; the rubbing range is set to 360° to simulate full-circle rubbing.

[0043] By comparing the figures, it can be seen that as the range of the friction-impact fault increases, the index value also increases, which verifies that the index can reflect the changes in the degree of the friction-impact fault, that is, the larger the friction-impact range, the more severe the fault degree.

[0044] By setting uneven gaps in different quadrants, the index values ​​under different frequencies of rubbing faults are compared. The comparison results are as follows: Figure 8 As shown. Figure 8 It can be seen that with the increase of the frequency of friction, the value of the index also increases, corresponding to the change in the degree of friction failure, that is, the greater the frequency of friction, the more serious the failure.

[0045] Combined with the above verification tests, different fault types and fault degrees were set through the simulation model, which verified that the indicators proposed by this technology can provide effective quantitative feedback on the changes in fault degree caused by penetration depth, rubbing range, and rubbing frequency, and at the same time have a good isolation effect on typical rotor faults represented by imbalance.

[0046] In summary, the above are only preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for identifying rubbing faults based on harmonic amplitude, characterized in that: The method includes the following steps: first, collecting vibration monitoring data of the aircraft engine in operation through a vibration sensor; second, converting the vibration monitoring data into the frequency domain through Fourier transform to obtain a spectrum diagram; third, extracting the harmonic amplitude in the spectrum diagram, and constructing a friction fault identification index based on the harmonic amplitude. Ca ; Finally, through Ca Identify whether the equipment has a friction fault; The constructed rubbing fault identification index Ca It is: the sum of the accumulated value and the accumulated value of the multi-harmonic amplitude minus the maximum value of the harmonic amplitude; that is: in For the i The harmonic amplitude of the order, N is the harmonic number of the indicator fusion.

2. A method for identifying rubbing faults based on harmonic amplitude as claimed in claim 1, characterized in that: Said through Ca Identify whether the equipment has a friction fault, specifically: obtain the friction fault identification index at each moment Ca Value, if the current moment Ca If the value changes more than the set threshold value compared with the previous moment, it is determined that a friction fault has occurred at the current moment.

3. A method for identifying rubbing faults based on harmonic amplitude as claimed in claim 1, characterized in that: Based on the harmonic amplitude, the friction fault identification index is constructed Ca The specific process includes the following steps: S 1. Establish a rotor-support-casing model and obtain the dynamic equation of the model; S 2. Set up the blade-casing friction fault simulation method, through Lankarani - Nikravesh The friction force model calculates the friction force between the blade and the casing, and calculates the additional force on the rotor due to the friction failure through the blade deformation; S 3. The dynamic equations of the model are numerically solved by the Runge-Kutta method, and the vibration characteristics of the casing obtained by the solution are analyzed to summarize the characteristic frequencies of the casing under the rubbing fault state; S 4. Based on the characteristic frequency of the casing under the rubbing fault state obtained by analysis, a fault identification index is constructed.

4. The method for identifying rubbing faults based on harmonic amplitude according to claim 3, characterized in that: Said S 2, a certain non-uniform gap is set between the blade and the casing, and its expression is: （1） in, represents the initial gap, Indicates the change gap, represents the gap angle, Indicates the gap range; subscript k Indicates the leaf number, Indicates k The angular position of the blades, Indicates k The non-uniform gaps corresponding to the angular positions of the blades; When the relative displacement between the blade and the casing exceeds the gap, that is, contact occurs, the Lankarani - Nikravesh The radial friction force generated by the model calculation is expressed as: （2） in represents the friction stiffness, represents the coefficient of restitution, represents the penetration depth, represents the friction velocity, represents the initial rubbing velocity; Based on Coulomb's friction law and combined with the blade tip tangential velocity, the expression of the friction tangential force is obtained: （3） in They represent symbolic functions, is the blade tangential velocity; At the same time, the disk is connected to the blade, and the blade is deformed by the friction force, which causes the disk to be subjected to an additional force. The additional force is transmitted by the blade deformation combined with the equivalent stiffness, and its expression is: （4） in represents the additional force acting on the disk, Indicates the number of disk blades, represents the equivalent stiffness of the disk blade in the chord direction, represents the equivalent axial stiffness of the disk blade, represents the normal equivalent stiffness of the disk blade; k The three coordinate transformation matrices generated by the gyroscopic effect of each blade are: , coordinate transformation matrix 2 And coordinate transformation matrix three , the gyroscopic effect causes the turntable to swing, and the force is adjusted back to the horizontal and vertical positions through three coordinate transformation matrices; , , They are the blade chord displacement, blade axial displacement and blade normal displacement respectively.

5. A rubbing fault identification method based on harmonic amplitude as claimed in claim 3 or 4, characterized in that: Said S 3, the obtained vibration characteristics of the casing under the rubbing fault state are analyzed, and the characteristic frequencies of the casing under the rubbing fault state are summarized, specifically: Set different variable gaps to simulate different degrees of rubbing faults. The larger the variable gap, the more serious the rubbing fault. The variable gap is 0μ m That is the normal state; The obtained casing vibration signal is differentiated into an acceleration vibration signal and plotted into a spectrum diagram. The rotational frequency vibration is intercepted for comparison, and the first conclusion of the casing vibration characteristics is obtained: the rotational frequency amplitude in the casing vibration increases with the severity of the fault.

6. A method for identifying rubbing faults based on harmonic amplitude as claimed in claim 5, characterized in that: Said S 3, the harmonic frequency components are further compared, and the second conclusion of the casing vibration characteristics is obtained: the harmonic components in the casing vibration are excited to different degrees due to the occurrence of friction faults, and as the degree of fault increases, the degree of change of each harmonic amplitude becomes more significant.

Images