CN112197968A - Confidence rule base-based intelligent diagnosis method for dynamic and static rub-impact faults of steam turbine rotor - Google Patents
Confidence rule base-based intelligent diagnosis method for dynamic and static rub-impact faults of steam turbine rotor Download PDFInfo
- Publication number
- CN112197968A CN112197968A CN202011046200.2A CN202011046200A CN112197968A CN 112197968 A CN112197968 A CN 112197968A CN 202011046200 A CN202011046200 A CN 202011046200A CN 112197968 A CN112197968 A CN 112197968A
- Authority
- CN
- China
- Prior art keywords
- dynamic
- diagnosis
- confidence
- amplitude
- impact
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000003745 diagnosis Methods 0.000 title claims abstract description 99
- 230000003068 static effect Effects 0.000 title claims abstract description 79
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000006073 displacement reaction Methods 0.000 claims abstract description 8
- 230000008569 process Effects 0.000 claims abstract description 8
- 208000024891 symptom Diseases 0.000 claims description 42
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000014509 gene expression Effects 0.000 claims description 5
- 230000008439 repair process Effects 0.000 claims description 5
- 206010063385 Intellectualisation Diseases 0.000 abstract description 2
- 238000005070 sampling Methods 0.000 description 25
- 230000006870 function Effects 0.000 description 11
- 230000007246 mechanism Effects 0.000 description 6
- 238000000819 phase cycle Methods 0.000 description 4
- 230000000630 rising effect Effects 0.000 description 4
- 238000001228 spectrum Methods 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 230000001174 ascending effect Effects 0.000 description 2
- XEBWQGVWTUSTLN-UHFFFAOYSA-M phenylmercury acetate Chemical compound CC(=O)O[Hg]C1=CC=CC=C1 XEBWQGVWTUSTLN-UHFFFAOYSA-M 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M15/00—Testing of engines
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H11/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves by detecting changes in electric or magnetic properties
- G01H11/06—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves by detecting changes in electric or magnetic properties by electric means
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M15/00—Testing of engines
- G01M15/02—Details or accessories of testing apparatus
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/17—Mechanical parametric or variational design
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Geometry (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computational Mathematics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Control Of Turbines (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention relates to a steam turbine rotor dynamic and static rubbing fault intelligent diagnosis method based on a confidence rule base, which comprises the following steps: setting a fundamental frequency amplitude threshold value and a double frequency amplitude threshold value of the steam turbine rotor with dynamic and static rubbing faults; adopting X, Y two eddy current displacement sensors to obtain an absolute shaft vibration X-direction signal and an absolute shaft vibration Y-direction signal of the turbine rotor, and further obtaining fundamental frequency amplitude and double frequency amplitude characteristics of the turbine rotor in the X direction and the Y direction; and judging whether the fundamental frequency amplitude and the double frequency amplitude in the X direction and the Y direction exceed a set threshold value, if so, starting a corresponding dynamic and static rub-impact rule base, and diagnosing the fault type of the steam turbine rotor with dynamic and static rub-impact faults, otherwise, judging that the steam turbine rotor does not have the dynamic and static rub-impact faults. Compared with the prior art, the method improves the efficiency of diagnosing and repairing the dynamic and static rubbing faults of the high-pressure rotor of the steam turbine, and solves the problems of intellectualization, accuracy and timeliness in the fault diagnosis process.
Description
Technical Field
The invention relates to an intelligent diagnosis method for equipment faults, in particular to an intelligent diagnosis method for dynamic and static rub-impact faults of a steam turbine rotor based on a confidence rule base.
Background
At present, development and research have been carried out on a steam turbine rotor dynamic and static rub-impact fault diagnosis system, however, in the existing system, most evidences of a rule base of the system can only be determined manually and can not be semantically, so that intelligent diagnosis of dynamic and static rub-impact faults is difficult to realize. Meanwhile, the evidence of the rule base is based on the static characteristics of the symptoms, and the dynamic change characteristics of the fault cannot be reflected. In addition, in the fault intelligent diagnosis system based on the deep learning strategy of big data analysis, the accuracy of the diagnosis result is low because of less actual fault samples and poor model generalization capability.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provide an intelligent diagnosis method for the dynamic and static rub-impact faults of the steam turbine rotor based on a confidence rule base.
The purpose of the invention can be realized by the following technical scheme:
the intelligent diagnosis method for the dynamic and static rub-impact fault of the steam turbine rotor based on the confidence rule base comprises the following steps:
s1, setting a fundamental frequency amplitude threshold value and a double-frequency amplitude threshold value of the steam turbine rotor with dynamic and static rub-impact faults;
s2, acquiring an absolute shaft vibration X-direction signal and an absolute shaft vibration Y-direction signal of the turbine rotor by adopting X, Y two eddy current displacement sensors, and further acquiring fundamental frequency amplitude and double-frequency amplitude characteristics of the turbine rotor in the X direction and the Y direction;
and S3, judging whether the obtained fundamental frequency amplitude and the double frequency amplitude in the X direction and the Y direction exceed a set threshold value, if so, starting a corresponding dynamic and static rub-impact rule base, and diagnosing the fault type of the steam turbine rotor with dynamic and static rub-impact faults, otherwise, judging that the steam turbine rotor does not have the dynamic and static rub-impact faults. Further, after the corresponding dynamic and static rub-impact rule base is started, initial rub-impact diagnosis is conducted, if the diagnosis result is established, the diagnosis is turned to a diagnosis human-computer interface of the initial rub-impact fault, the diagnosis human-computer interface displays the type of the judged initial rub-impact fault, the confidence coefficient of the judged initial rub-impact fault, the reason causing the initial rub-impact fault and fault repairing measure information, and the whole diagnosis process is ended.
And when only the obtained fundamental frequency amplitudes in the X direction and the Y direction exceed the specified threshold, starting the dynamic and static rub-impact rule base for fault diagnosis when the fundamental frequency amplitudes exceed the set threshold.
And when the obtained fundamental frequency amplitude and the double-frequency amplitude in the X direction and the Y direction exceed the set threshold simultaneously, starting the dynamic-static rub-impact rule base for fault diagnosis when the fundamental frequency amplitude and the double-frequency amplitude exceed the set threshold simultaneously.
When only the obtained fundamental frequency amplitudes in the X direction and the Y direction exceed a specified threshold, sequentially judging the following evidences: judging whether the amplitude of the fundamental frequency of the X-direction axial vibration is not related to the load, if so, finishing the diagnosis, if not, judging whether the amplitude of the fundamental frequency of the Y-direction axial vibration is not related to the load, if so, finishing the diagnosis, if not, judging whether the amplitude of the fundamental frequency of the X-direction axial vibration is fluctuated, if not, finishing the diagnosis, otherwise, judging whether the amplitude of the fundamental frequency of the Y-direction axial vibration is fluctuated, if not, finishing the diagnosis, otherwise, judging whether the X-direction axial vibration generates subsynchronous harmonic vibration, if not, finishing the diagnosis, otherwise, judging whether the phase of the fundamental frequency of the X-direction eddy current displacement sensor is fluctuated, if not, finishing the diagnosis, if fluctuated, judging whether the phase of the fundamental frequency of the Y-direction eddy current displacement sensor is fluctuated, if the fluctuation does not occur, the diagnosis is finished, and if the fluctuation occurs, the initial dynamic and static rubbing faults of the turbine rotor are diagnosed.
When the obtained fundamental frequency amplitude and the double frequency amplitude in the X direction and the Y direction simultaneously exceed a set threshold, sequentially judging the following evidences: judging whether the amplitude of the X-direction axial vibration base frequency is related to fluctuation or not, if not, ending the diagnosis, if so, judging whether the amplitude of the basic frequency of the Y-direction axial vibration is related to fluctuation or not, if not, ending the diagnosis, if so, judging whether the amplitude of the double frequency of the X-direction axial vibration is continuously increased, if not, ending the diagnosis, if so, judging whether the amplitude of the double frequency of the Y-direction axial vibration is continuously increased, if not, ending the diagnosis, if so, judging whether the phase of the X-direction axial vibration fundamental frequency fluctuates or not, if not, finishing the diagnosis, otherwise, judging whether the phase of the Y-direction axial vibration fundamental frequency fluctuates or not, if no fluctuation occurs, the diagnosis is finished, otherwise, whether the average axial center position of the turbine rotor deviates or not is judged, if no deviation occurs, the diagnosis is finished, and if deviation occurs, the initial dynamic and static rubbing faults of the turbine rotor are diagnosed.
In the diagnosis process, diagnosis is carried out on each condition according to the fault symptom confidence coefficient matched with the rule. The concrete contents are as follows:
obtaining a matching degree theta according to a rule matching algorithm:
θ＝max{0,δ1-δ′1}+max{0,δ2-δ′2}+...+max{0,δN-δ′N}＜ε
in the formula, deltaiConfidence of the ith evidence in the rule base, deltai' is the calculated confidence of the ith symptom, and epsilon is a matching index;
if the calculation result on the left side in the above formula is not greater than epsilon, that is, the evidence confidence is smaller than the precondition confidence of the rule, the matching result is true, and then the conclusion confidence α is calculated, which has the following calculation formula:
α＝[1-max{0,δ1-δ′1}]×[1-max{0,δ2-δ′2}]×...×[1-max{0,δN-δ′N}]×β
where β is the confidence in the conclusion given in the rule base.
The confidence coefficient of the fault symptom matched with the rule is calculated through a confidence coefficient function, and the specific content is as follows:
the threshold value of the ith symptom confidence function is defined as xLiAnd xHiThe upper limit value and the lower limit value are xmaxiAnd xminiWhen the symptom xiFalls within the interval [ x ]Li,xHi]Time indicates normal symptom, expressed as N, exceeds xmaxiIndicates a symptom of abnormally high, denoted FH, less than xminiThe symptom is represented to be abnormally low, and is represented by LF, the symptom fact is converted into evidence with certain confidence through a fuzzy membership function, the symptom confidence is an expression of deviation of the symptom parameter from a limit state, and then the expression of the ith symptom confidence function is:
further, as a preferred scheme, the set threshold of the fundamental frequency amplitude is 125 μm, and the set threshold of the frequency doubling amplitude is 120 um.
Further, when the initial rub-impact fault is judged, the cause and fault repair measure information of the initial rub-impact fault are obtained from a turbine high-pressure rotor dynamic and static rub-impact fault knowledge base, and the turbine high-pressure rotor dynamic and static rub-impact fault knowledge base is obtained by summarizing fault cases and expert experience.
Compared with the prior art, the intelligent diagnosis method for the dynamic and static rub-impact faults of the steam turbine rotor based on the confidence rule base at least has the following beneficial effects:
firstly, analyzing and extracting fault characteristic parameters through the mechanism of dynamic and static rub-impact faults, extracting the fault characteristic parameters through the mechanism of the dynamic and static rub-impact faults and expert experience, forming evidences, carrying out logical AND on the evidences to obtain the conclusion of the dynamic and static rub-impact faults, and constructing a complete dynamic and static rub-impact fault diagnosis confidence rule base; according to the method, after the vibration signal of the high-pressure rotor of the steam turbine is abnormal, the dynamic and static rub-impact fault diagnosis is carried out on the steam turbine rotor, the occurrence rate of the dynamic and static rub-impact fault of the steam turbine rotor is reduced, the dynamic and static rub-impact fault diagnosis and repair efficiency of the high-pressure rotor of the steam turbine is improved, and the safe and stable operation of a steam turbine unit is guaranteed;
secondly, the intelligent diagnosis method for the dynamic and static rub-impact faults of the steam turbine rotor based on the confidence rule base is used for establishing a one-to-one mapping relation between fault symptoms and fault dynamic and static rub-impact fault types through the combination of a fault mechanism model and expert experience to obtain a fault diagnosis confidence rule base; all evidences of the rule base realize semantization, and reflect the dynamic characteristics of the rub-impact fault through the statistical characteristics of the evidences, so that the problems of intellectualization, accuracy and timeliness in the fault diagnosis process are solved.
Drawings
FIG. 1 is a schematic flow chart of an intelligent diagnosis method for dynamic and static rub-impact faults of a turbine rotor in an embodiment when a fundamental frequency amplitude exceeds a set threshold;
FIG. 2 is a schematic flow chart of an intelligent diagnosis method for dynamic and static rub-impact faults of a turbine rotor when the fundamental frequency amplitude and the double frequency amplitude exceed the set threshold value simultaneously in the embodiment;
FIG. 3 is a schematic representation of the orientation of the sensor X, Y in an example embodiment;
FIG. 4 is a diagram illustrating a fault symptom confidence function in an embodiment.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. It is to be understood that the embodiments described are only a few embodiments of the present invention, and not all embodiments. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, shall fall within the scope of protection of the present invention.
Examples
The embodiment provides an intelligent diagnosis method for dynamic and static rub-impact faults of a steam turbine rotor based on a confidence rule base. The system adopts two trigger mechanisms of 1X trigger when the frequency spectrum of the system has fundamental frequency and 1X2X trigger when the frequency spectrum has fundamental frequency and frequency doubling simultaneously according to the frequency characteristics of the dynamic and static rub-impact faults at different severity degrees to trigger the rule base diagnosis, and judges whether the dynamic and static rub-impact faults occur to the high-pressure rotor according to the fault rule.
The method firstly obtains two kinds of frequency spectrum characteristics of dynamic and static rub-impact faults, aiming at the two kinds of frequency spectrum characteristics of the dynamic and static rub-impact faults, respectively establishing a rule base when the fundamental frequency and the double frequency simultaneously exceed a certain value, and the starting conditions of the rule base are as follows:
IF AfR＞An
THEN startup rule base
Wherein A isfRIs an actual value, AnIn order to set the threshold of the start rule base, the present embodiment uses the threshold A of the fundamental frequency amplitude1Set to 125 μm, threshold A of the frequency doubling amplitude2Set to 120 um.
In this embodiment, a 1000MW steam turbine high-pressure rotor is taken as an example, and the intelligent diagnosis method for dynamic and static rub-impact faults of the steam turbine rotor based on the confidence rule base is specifically described.
Firstly, when only the fundamental frequency amplitude exceeds the set threshold, the starting flow chart of the dynamic and static rub-impact rule base is shown in fig. 1.
a1) When the amplitude of the fundamental frequency exceeds the threshold value, the started dynamic and static rub-impact rule base can be diagnosed as an initial rub-impact fault.
a2) Setting the threshold value of the vibration fundamental frequency amplitude of the high-pressure rotor of the 1000MW steam turbine to be 125 mu m, and starting the dynamic-static rub-impact rule base when the vibration fundamental frequency amplitude of the high-pressure rotor of the steam turbine is higher than 125 mu m.
a3) After the rule base is started, initial rub-impact diagnosis is carried out, if the diagnosis result is established, the method is switched to a diagnosis human-computer interface explanation of the initial rub-impact fault: the fault types are initial rub-on, confidence coefficient of the initial rub-on, cause of the initial rub-on fault and fault repairing measure information, and the whole diagnosis process is ended; and if the diagnosis result is not true, judging that no dynamic and static rubbing faults occur in the high-pressure rotor of the steam turbine.
The cause and fault repair measure information of the initial rub-impact fault are obtained from a knowledge base of the dynamic and static rub-impact fault of the high-pressure rotor of the steam turbine, the knowledge base of the dynamic and static rub-impact fault of the high-pressure rotor of the steam turbine is obtained by summarizing fault cases and expert experience, and relevant operation parameters, fault causes and fault solutions are shown in table 1.
TABLE 1 knowledge base of high-pressure rotor dynamic and static rubbing faults of steam turbine
The initial rub-impact fault rule base when the 1X amplitude exceeds the set threshold includes evidence, conclusions and confidence levels as follows:
(1) initial rub-impact fault diagnosis rule:
IF X directional axial vibration fundamental frequency amplitude is not related to load (0.90)
The amplitude of the fundamental frequency of the AND Y-direction axial vibration is not related to the load (0.90)
The amplitude of the fundamental frequency of the AND X-direction axial vibration fluctuates (0.95)
The amplitude of the fundamental frequency of the AND Y-direction axial vibration fluctuates (0.95)
AND X-directional axial vibration generating subsynchronous harmonic vibration (1/nX) (0.90)
AND Y-direction axial vibration generating subsynchronous harmonic vibration (1/nX) (0.90)
Fundamental frequency phase fluctuation of AND X sensor (0.85)
Fundamental frequency phase fluctuation of AND Y sensor (0.85)
THEN initial collision and friction (0.90)
(2) The semantic algorithm comprises the following steps:
x, Y two eddy current displacement sensors are used for acquiring X-direction signals and Y-direction signals of absolute axial vibration of the high-voltage rotor. X, Y the sensors are installed on the surface of the high-pressure rotor in the orthogonal direction, wherein the X sensor is at 45 degrees left, the Y sensor is at 45 degrees right, the equal interval sampling is carried out on the fundamental frequency amplitude and the frequency doubling amplitude (phase), the sampling time T is 10min, the sampling period T is 5h, and the total acquisition is carried out
A data point. The 30 points form a fundamental/frequency doubled amplitude (phase) curve.
As shown in fig. 3, the vibration signal obtained by the X-sensor in the k-th sampling is subjected to Fast Fourier Transform (FFT), so as to obtain the fundamental frequency amplitude a in the X-directionX(k) And phase of fundamental frequency
Performing Fast Fourier Transform (FFT) on the vibration signal obtained by the Y sensor in the k sampling, thereby obtaining the fundamental frequency amplitude a in the Y directionY(k) And phase of fundamental frequency
And recording the output power at the turbine end obtained in the k sampling as p (k).
Among the above evidence:
(ii) evidence 1: the amplitude of the fundamental frequency in the X direction is not related to the load:
degree of gray correlation rXThreshold value: 0.4, lower limit: 0, upper limit value: 1.
within one sampling period, obtaining an X-direction fundamental frequency amplitude sequence { aX(k) K is 1,2 …, N, and the mean of the sequence is:
wherein N is the sequence length.
Dividing the value of the X-direction fundamental frequency amplitude sequence by the mean value
Obtaining a preprocessed X-direction fundamental frequency amplitude sequence { a'X(k) K is 1,2, …, N }, wherein
In a sampling period, a turbine load sequence { p (k) }, k ═ 1,2, …, N } is obtained, and the average value of the sequence is:
dividing the values of the load sequence by the mean value MPA preprocessed load sequence { p' (k), k ═ 1,2, …, N } is obtained, where
Calculating an X-direction fundamental frequency amplitude sequence { a'X(k) Gray correlation coefficient of 1,2, …, N with the load sequence { p' (k), 1,2, …, N }:
wherein ρ is a correlation coefficient, and ρ is 0.5;
further, { a'X(k) The degree of association of k 1,2, …, N with { p' (k), k 1,2, …, N }:
if 0 < rXIf the frequency is less than 0.085, the confidence that the amplitude of the fundamental frequency in the X direction is irrelevant to the load is more than 0.90.
② evidence 2: the amplitude of the fundamental frequency in the Y direction is not related to the load:
degree of gray correlation rYThreshold value: 0.4, lower limit: 0, upper limit value: 1
Within one sampling period, obtaining a Y-direction fundamental frequency amplitude sequence { aY(k) K is 1,2 …, N, and the mean of the sequence is
Dividing the value of the Y-direction fundamental frequency amplitude sequence by the mean value
Obtaining a preprocessed Y-direction fundamental frequency amplitude sequence { a'Y(k) K is 1,2, …, N }, wherein
In a sampling period, a turbine load sequence { p (k) }, k ═ 1,2, …, N } is obtained, and a mean value M of the sequence is calculated according to the formula (2)PDividing the values of the load sequence by the mean value yields a preprocessed load sequence { p' (k), k ═ 1,2, …, N }, where
Calculating a Y-direction fundamental frequency amplitude sequence { a'Y(k) Gray correlation coefficient of 1,2, …, N with the load sequence { p' (k), 1,2, …, N }:
wherein ρ is a correlation coefficient, and ρ is 0.5;
further, { a'Y(k) The degree of association of k 1,2, …, N with { p' (k), k 1,2, …, N }:
if 0 < rXIf the frequency is less than 0.085, the confidence that the amplitude of the fundamental frequency in the Y direction is irrelevant to the load is more than 0.90.
③ evidence 3: the amplitude of the fundamental frequency in the X direction fluctuates:
in a sampling period, the amplitudes of any two adjacent points in the X-direction fundamental frequency amplitude sequence are respectively aX(k)，aX(k +1) obtaining a new judgment sequence according to the formula (8)
Definition of [ X]For rounding operations, i.e. [ X ]]Representing taking the largest integer not exceeding X. If the new judgment sequence { A }X(k) If k is 1,2, …, N-1, and equations (9) to (10) are satisfied, the confidence that the fundamental frequency amplitude fluctuates is greater than 0.95.
In the above formula, RX+The ratio of the data segment with rising trend between two adjacent points in the X-direction amplitude curve to the whole length of the curve is shown; rX-The ratio of the data segment with the descending trend between two adjacent points in the X-direction amplitude curve to the whole length of the curve.
(iv) evidence 4: the amplitude of the fundamental frequency in the Y direction fluctuates:
in a sampling period, the amplitudes of any two adjacent points in the Y-direction fundamental frequency amplitude sequence are respectively aY(k)，aY(k +1), obtaining a new judgment sequence according to the formula (11):
definition of [ X]For rounding operations, i.e. [ X ]]Representing taking the largest integer not exceeding X. If the new judgment sequence { A }Y(k) If k is 1,2, …, N-1, and equations (12) to (13) are satisfied, the confidence that the fundamental frequency amplitude fluctuates is greater than 0.95.
In the above formula, RY+The ratio of the data segment with rising trend between two adjacent points in the amplitude curve in the Y direction to the whole length of the curve is shown; rY-The ratio of the data segment with the descending trend between two adjacent points in the Y-direction amplitude curve to the whole length of the curve.
Evidence 5: synchronous harmonic vibration is generated in the X direction (1/nX, n is 2,3, 4):
1/nX amplitude threshold A1/nComprises the following steps: 40 μm, limit: 80 μm.
When a collision or a friction fault occurs, subsynchronous self-excited vibration having a large amplitude is generated, and the vibration frequency is a fractional harmonic frequency (1/2X, 1/3X, 1/4X, etc.). If the X-direction axial vibration satisfies equation (14), the confidence of the X-direction generated subsynchronous harmonic vibration is greater than 0.90.
Wherein the content of the first and second substances,
at a vibration frequency of X direction of
The actual amplitude of (c).
Sixthly, evidence 6: subsynchronous harmonic vibrations are generated in the Y direction (1/nX, n is 2,3, 4):
when a collision or a friction fault occurs, subsynchronous self-excited vibration having a large amplitude is generated, and the vibration frequency is a fractional harmonic frequency (1/2X, 1/3X, 1/4X, etc.). If the Y-direction axial vibration satisfies equation (15), the confidence of the Y-direction generated subsynchronous harmonic vibration is greater than 0.90.
Wherein the content of the first and second substances,
has a vibration frequency of Y direction of
The actual amplitude of (c).
Seventh, evidence 7: x-sensor fundamental frequency phase fluctuation:
the phase of any two adjacent points in the X-direction fundamental frequency phase sequence in one sampling period is
Obtaining a new judgment sequence according to equation (16)
Definition of [ X]For rounding operations, i.e. [ X ]]Representing taking the largest integer not exceeding X. If the new judgment sequence { phiX(k) K is 1,2, …, N-1, satisfies the formulas (17) to (18),the confidence that the fundamental frequency phase fluctuation fluctuates is greater than 0.85.
In the above formula, ΨX+The proportion of a data segment with an ascending trend between two adjacent points in the X-direction phase curve to the whole length of the curve is shown; ΨX-The ratio of the data segment with the descending trend between two adjacent points in the X-direction phase curve to the whole length of the curve.
The evidence 8: y sensor fundamental frequency phase fluctuation:
the phase of any two adjacent points in the Y-direction fundamental frequency phase sequence in one sampling period is
Obtaining a new judgment sequence according to equation (19)
Definition of [ X]For rounding operations, i.e. [ X ]]Representing taking the largest integer not exceeding X. If the new judgment sequence { phiY(k) And k is 1,2, …, N-1, which satisfies the formulas (20) to (21), the confidence that the fundamental frequency phase fluctuation is greater than 0.85.
In the above formula, ΨY+The proportion of a data segment with an ascending trend between two adjacent points in the phase curve in the Y direction to the whole length of the curve is shown; ΨY-The ratio of the data segment with the descending trend between two adjacent points in the Y-direction phase curve to the whole length of the curve.
Secondly, when the fundamental frequency amplitude and the double-frequency amplitude exceed the set threshold value at the same time, the starting flow chart of the dynamic and static rub-impact rule base is shown in fig. 2.
b1) When the fundamental frequency (1X) amplitude and the double frequency (2X) amplitude exceed the threshold value simultaneously, the activated dynamic and static rub-impact rule base can diagnose rub-impact faults.
b2) Setting the vibration fundamental frequency amplitude threshold value of a high-pressure rotor of a 1000MW steam turbine to be 125 mu m and the frequency doubling amplitude threshold value to be 120 mu m respectively, and starting the dynamic and static rub-impact fault diagnosis rule base when the vibration fundamental frequency amplitude and the frequency doubling vibration amplitude of the steam turbine rotor are both higher than the set threshold values.
b3) After the rule base is started, the rubbing fault is diagnosed, if the diagnosis result is established, the rubbing fault is turned to a diagnosis human-computer interface explanation: the fault types are rubbing, confidence coefficient of rubbing, cause of rubbing fault and fault repairing measure information, and the whole diagnosis process is finished; if the diagnosis result is not true, the high-pressure rotor of the steam turbine has no dynamic and static rubbing fault.
(1) The rub-impact fault rule base when the 1X amplitude and the 2X amplitude exceed the set threshold simultaneously includes evidence, conclusions and confidence levels as follows:
(1) rub-impact fault diagnosis rule:
IF X directional axial vibration 1X amplitude fluctuation (0.90)
Fluctuation of 1X amplitude of AND Y-direction axial vibration (0.90)
The amplitude of the AND X axial vibration 2X is continuously increased (0.95)
The amplitude of the AND Y-direction axial vibration 2X is continuously increased (0.95)
AND X-directional axial vibration 1X phase fluctuation (0.85)
AND Y-direction axial vibration 1X phase fluctuation (0.85)
AND average axial center position deviation (0.85)
THEN Bomo (0.90)
(2) The semantic algorithm comprises the following steps:
sampling the sensor at equal intervals, wherein the sampling interval T is 10min, and the sampling period T is 5h, and obtaining
A data point. Performing Fast Fourier Transform (FFT) on the vibration signal obtained by the X sensor in the k sampling, thereby obtaining a 1X amplitude a in the X directionX(k) And 1X phase
2X amplitude a2X(k) And 2X phase
Performing Fast Fourier Transform (FFT) on the vibration signal obtained by the Y sensor in the k sampling, thereby obtaining a 1X amplitude a in the Y directionY(k) 1X phase
2X amplitude a2Y(k) And 2X phase
Evidence 1: 1X amplitude fluctuation in X direction
In a sampling period, the amplitudes of any two adjacent points in the X-direction 1X amplitude sequence are respectively aX(k)，aX(k +1), a new judgment sequence { A ] is obtained from the formula (8)X(k) K is 1,2, …, N-1 }. If the new judgment sequence { A }X(k) If k is 1,2, …, N-1, and equations (9) to (10) are satisfied simultaneously, the confidence that the amplitude of 1X fluctuates in the X direction is greater than 0.95.
Evidence 2: 1X amplitude fluctuation in Y direction
In a sampling period, the amplitudes of any two adjacent points in the 1X amplitude sequence in the Y direction are respectively aY(k)，aY(k +1), a new judgment sequence { A ] is obtained from the formula (11)Y(k) K is 1,2, …, N-1 }. If the new judgment sequence { A }Y(k) K-1, 2, …, N-1, and further satisfy the formula (12) to E(13) Then the confidence that the amplitude of 1X in the Y direction fluctuates is greater than 0.95.
Evidence 3: 2X amplitude continuously rising in X direction
In a sampling period, the amplitudes of any two adjacent points in the X-direction 2X amplitude sequence are respectively a2X(k)，a2X(k +1) obtaining a new judgment sequence according to the formula (22)
Definition of [ X]For rounding operations, i.e. [ X ]]Representing taking the largest integer not exceeding X. If the new judgment sequence { A }2X(k) If k is 1,2, …, N-1, satisfying equation (23), the confidence that the X-direction 2X amplitude is increasing is greater than 0.95.
Evidence 4: 2X amplitude continuously rising in Y direction
In a sampling period, the amplitudes of any two adjacent points in the 2X amplitude sequence in the Y direction are respectively a2Y(k)，a2Y(k +1), obtaining a new judgment sequence according to the formula (24):
definition of [ X]For rounding operations, i.e. [ X ]]Representing taking the largest integer not exceeding X. If the new judgment sequence { A }2Y(k) If k is 1,2, …, N-1, satisfying equation (25), the confidence that the Y-direction 2X amplitude is increasing is greater than 0.95.
Evidence 5: the phase of the fundamental frequency in the X direction fluctuates
Within one sampling period, the phase of any two adjacent points in the X-direction 1X phase sequence is
Obtaining a new judgment sequence [ phi ] according to the formula (15)X(k) K is 1,2, …, N-1 }. If the new judgment sequence { phiX(k) When k is 1,2, …, N-1, and satisfies equations (17) to (18), the confidence that the 1X phase fluctuation in the X direction fluctuates is greater than 0.85.
Evidence 6: the phase of the fundamental frequency in the Y direction fluctuates
Within one sampling period, the phase of any two adjacent points in the Y-direction 1X phase sequence is
Obtaining a new judgment sequence [ phi ] according to the formula (18)Y(k) K is 1,2, …, N-1 }. If the new judgment sequence { phiY(k) When k is 1,2, …, N-1, and satisfies equations (20) to (21), the confidence that the 1X phase fluctuation in the Y direction fluctuates is greater than 0.85.
Evidence 7: deviation of average axis locus
When the steam turbine normally operates, the position angle theta of the bearing rotor is usually between 30 degrees and 60 degrees, the range of the eccentricity epsilon is 0.6 to 0.8, and if the formula (26) or the formula (27) is satisfied, the confidence coefficient of the deviation of the average axial position is more than 0.85.
0°≤θ＜8.75°OR 89.75°＜θ≤360° (27)
And thirdly, in the rule base, the confidence coefficient of the diagnosis rule is determined according to the field expert experience, but in the fault reasoning process, the fault symptom confidence coefficient matched with the rule needs to be calculated through a confidence coefficient function.
The normal value boundary (threshold) defining the ith symptom confidence function is xLiAnd xHiThe upper limit and the lower limit are xmaxiAnd xminiI.e. symptom xiFalls within the interval [ x ]Li,xHi]Time indicates normal (N) signs, exceeding xmaxiIndicating a symptom of abnormally high (FH), less than xminiIndicating an abnormally low symptom (LF). The symptom facts are converted into evidences with certain confidence degrees through a fuzzy membership function, the symptom confidence degrees are expressions of symptom parameters deviating from the boundary state and reflect the abnormal degree of the parameters, and the value of the symptom confidence degrees is generally [0,1 ]]In the meantime. The ith symptom confidence function may be expressed as:
the confidence function for the ith symptom of failure can be represented by fig. 4.
The setting of the threshold and the limit of the confidence function of the dynamic and static rubbing fault symptom is shown in table 2:
TABLE 2 Fault symptom threshold, Limit initialization value settings
| Serial number | Fault diagnosis rules | Unit of | xmini | xLi | xHi | xmaxi |
| 1 | The amplitude of the fundamental frequency of the X/Y direction axial vibration is not related to the load | -- | 0 | 0.4 | -- | -- |
| 2 | The amplitude of the fundamental frequency of the axial vibration in the X/Y direction fluctuates | -- | 0.4 | 0.45 | 0.55 | 0.6 |
| 3 | Subsynchronous harmonic vibration generated by X/Y direction axial vibration | um | -- | -- | 40 | 80 |
| 4 | Fundamental frequency phase fluctuation of X/Y sensor | -- | 0.4 | 0.45 | 0.55 | 0.6 |
| 5 | The 2X amplitude of the X/Y direction axial vibration is continuously increased | -- | -- | -- | 0.55 | 0.9 |
| 6 | Shift of average axial position | -- | 0.1 | 0.6 | 0.8 | 0.9 |
Fourthly, calculating a rule matching algorithm and a conclusion confidence coefficient:
the rule matching algorithm is as follows:
θ＝max{0,δ1-δ′1}+max{0,δ2-δ′2}+...+max{0,δN-δ′N}＜ε (31)
in the above formula, θ represents a matching degree; deltaiConfidence of the ith evidence in the rule base, deltai' is the calculated confidence of the ith symptom, and epsilon represents the matching index. If the calculation result on the left side of the formula is less than or equal to epsilon, namely the evidence confidence is less than the precondition confidence of the rule, the matching result is true, and then the conclusion confidence alpha is calculated, wherein the calculation formula is as follows:
α＝[1-max{0,δ1-δ′1}]×[1-max{0,δ2-δ′2}]×...×[1-max{0,δN-δ′N}]×β (32)
where β is the confidence in the conclusion given in the rule base.
The method comprises the steps of analyzing and extracting fault characteristic parameters through the mechanism of the dynamic and static rub-impact faults, extracting the fault characteristic parameters through the mechanism of the dynamic and static rub-impact faults and expert experience, forming evidences, and constructing a complete dynamic and static rub-impact fault diagnosis confidence rule base through comparing logical phases of the evidences with the obtained conclusion of the dynamic and static rub-impact faults. According to the method, after the vibration signal of the high-pressure rotor of the steam turbine is abnormal, the dynamic and static rub-impact fault diagnosis is carried out on the steam turbine rotor, the occurrence rate of the dynamic and static rub-impact fault of the steam turbine rotor is reduced, the dynamic and static rub-impact fault diagnosis and repair efficiency of the high-pressure rotor of the steam turbine is improved, and the safe and stable operation of a steam turbine unit is guaranteed.
While the invention has been described with reference to specific embodiments, the invention is not limited thereto, and those skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope of the invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. The intelligent diagnosis method for the dynamic and static rub-impact fault of the steam turbine rotor based on the confidence rule base is characterized by comprising the following steps of:
setting a fundamental frequency amplitude threshold value and a double frequency amplitude threshold value of the steam turbine rotor with dynamic and static rubbing faults;
adopting X, Y two eddy current displacement sensors to obtain an absolute shaft vibration X-direction signal and an absolute shaft vibration Y-direction signal of the turbine rotor, and further obtaining fundamental frequency amplitude and double frequency amplitude characteristics of the turbine rotor in the X direction and the Y direction;
and judging whether the obtained fundamental frequency amplitude and the double frequency amplitude in the X direction and the Y direction exceed a set threshold value, if so, starting a corresponding dynamic and static rub-impact rule base, and diagnosing the fault type of the steam turbine rotor with dynamic and static rub-impact faults, otherwise, judging that the steam turbine rotor does not have the dynamic and static rub-impact faults.
2. The intelligent diagnosis method for steam turbine rotor dynamic and static rub-impact faults based on the confidence rule base as claimed in claim 1, wherein when only the obtained fundamental frequency amplitudes in the X direction and the Y direction exceed a specified threshold, the dynamic and static rub-impact rule base is started to perform fault diagnosis when the fundamental frequency amplitudes exceed a set threshold.
3. The intelligent diagnosis method for steam turbine rotor dynamic and static rub-impact faults based on the confidence rule base as claimed in claim 1, wherein when the obtained fundamental frequency amplitude and the double frequency amplitude in the X direction and the Y direction exceed the set threshold simultaneously, the dynamic and static rub-impact rule base is started to perform fault diagnosis when the fundamental frequency amplitude and the double frequency amplitude exceed the set threshold simultaneously.
4. The intelligent diagnosis method for the dynamic and static rub-impact fault of the steam turbine rotor based on the confidence rule base as claimed in claim 2, wherein when only the obtained fundamental frequency amplitudes in the X direction and the Y direction exceed a specified threshold, the following evidences are sequentially judged: judging whether the amplitude of the fundamental frequency of the X-direction axial vibration is not related to the load, if so, finishing the diagnosis, if not, judging whether the amplitude of the fundamental frequency of the Y-direction axial vibration is not related to the load, if so, finishing the diagnosis, if not, judging whether the amplitude of the fundamental frequency of the X-direction axial vibration is fluctuated, if not, finishing the diagnosis, otherwise, judging whether the amplitude of the fundamental frequency of the Y-direction axial vibration is fluctuated, if not, finishing the diagnosis, otherwise, judging whether the X-direction axial vibration generates subsynchronous harmonic vibration, if not, finishing the diagnosis, otherwise, judging whether the phase of the fundamental frequency of the X-direction eddy current displacement sensor is fluctuated, if not, finishing the diagnosis, if fluctuated, judging whether the phase of the fundamental frequency of the Y-direction eddy current displacement sensor is fluctuated, if the fluctuation does not occur, the diagnosis is finished, and if the fluctuation occurs, the initial dynamic and static rubbing faults of the turbine rotor are diagnosed.
5. The intelligent diagnosis method for steam turbine rotor dynamic and static rub-impact faults based on the confidence rule base according to claim 3, wherein when the acquired fundamental frequency amplitude and the double frequency amplitude in the X direction and the Y direction exceed the set threshold simultaneously, the following evidences are sequentially judged: judging whether the amplitude of the X-direction axial vibration base frequency is related to fluctuation or not, if not, ending the diagnosis, if so, judging whether the amplitude of the basic frequency of the Y-direction axial vibration is related to fluctuation or not, if not, ending the diagnosis, if so, judging whether the amplitude of the double frequency of the X-direction axial vibration is continuously increased, if not, ending the diagnosis, if so, judging whether the amplitude of the double frequency of the Y-direction axial vibration is continuously increased, if not, ending the diagnosis, if so, judging whether the phase of the X-direction axial vibration fundamental frequency fluctuates or not, if not, finishing the diagnosis, otherwise, judging whether the phase of the Y-direction axial vibration fundamental frequency fluctuates or not, if no fluctuation occurs, the diagnosis is finished, otherwise, whether the average axial center position of the turbine rotor deviates or not is judged, if no deviation occurs, the diagnosis is finished, and if deviation occurs, the initial dynamic and static rubbing faults of the turbine rotor are diagnosed.
6. The intelligent diagnosis method for the dynamic and static rub-impact faults of the turbine rotor based on the confidence rule base according to claim 4 or 5, wherein the diagnosis is carried out on each condition according to the fault symptom confidence matched with the rules.
7. The intelligent diagnosis method for the dynamic and static rub-impact faults of the steam turbine rotor based on the confidence rule base as claimed in claim 6 is characterized in that the specific contents of diagnosis according to the fault symptom confidence degree matched with the rules are as follows:
obtaining a matching degree theta according to a rule matching algorithm:
θ＝max{0,δ1-δ′1}+max{0,δ2-δ′2}+...+max{0,δN-δ′N}＜ε
in the formula, deltaiConfidence of the ith evidence in the rule base, deltai' is the calculated confidence of the ith symptom, and epsilon is a matching index;
if the calculation result on the left side in the above formula is not greater than epsilon, that is, the evidence confidence is smaller than the precondition confidence of the rule, the matching result is true, and then the conclusion confidence α is calculated, which has the following calculation formula:
α＝[1-max{0,δ1-δ′1}]×[1-max{0,δ2-δ′2}]×...×[1-max{0,δN-δ′N}]×β
where β is the confidence in the conclusion given in the rule base.
8. The intelligent diagnosis method for dynamic and static rub-impact faults of the steam turbine rotor based on the confidence rule base as claimed in claim 7, wherein the confidence of the fault symptoms matched with the rules is calculated by a confidence function, and the specific contents are as follows:
the threshold value of the ith symptom confidence function is defined as xLiAnd xHiThe upper limit value and the lower limit value are xmaxiAnd xminiWhen the symptom xiFalls within the interval [ x ]Li,xHi]Time indicates normal symptom, expressed as N, exceeds xmaxiIndicates a symptom of abnormally high, denoted FH, less than xminiThe symptom is represented to be abnormally low, and is represented by LF, the symptom fact is converted into evidence with certain confidence through a fuzzy membership function, the symptom confidence is an expression of deviation of the symptom parameter from a limit state, and then the expression of the ith symptom confidence function is:
9. the intelligent diagnosis method for steam turbine rotor dynamic and static rub-impact faults based on the confidence rule base according to claim 1, wherein the set threshold value of the fundamental frequency amplitude is 125 μm, and the set threshold value of the frequency doubling amplitude is 120 μm.
10. The intelligent diagnosis method for steam turbine rotor dynamic and static rub-and-impact faults based on the confidence rule base as claimed in claim 1, wherein after the corresponding dynamic and static rub-and-impact rule base is started, initial rub-and-impact diagnosis is performed, if the diagnosis result is established, the diagnosis man-machine interface is turned to the diagnosis man-machine interface of the initial rub-and-impact faults, the diagnosis man-machine interface displays the type of the initial rub-and-impact faults judged, the confidence level of the initial rub-and-impact judged, the reasons causing the initial rub-and-impact faults and fault repair measure information, and the whole diagnosis process is ended.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011046200.2A CN112197968B (en) | 2020-09-29 | 2020-09-29 | Turbine rotor dynamic and static collision friction fault intelligent diagnosis method based on confidence rule base |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011046200.2A CN112197968B (en) | 2020-09-29 | 2020-09-29 | Turbine rotor dynamic and static collision friction fault intelligent diagnosis method based on confidence rule base |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112197968A true CN112197968A (en) | 2021-01-08 |
| CN112197968B CN112197968B (en) | 2023-04-28 |
Family
ID=74007025
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011046200.2A Active CN112197968B (en) | 2020-09-29 | 2020-09-29 | Turbine rotor dynamic and static collision friction fault intelligent diagnosis method based on confidence rule base |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112197968B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113514144A (en) * | 2021-07-28 | 2021-10-19 | 郑州轻工业大学 | Unbalance-rubbing coupling fault detection method based on eddy current displacement sensor |
| CN115683687A (en) * | 2023-01-03 | 2023-02-03 | 成都大汇物联科技有限公司 | Dynamic and static rub-impact fault diagnosis method for hydroelectric mechanical equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040199573A1 (en) * | 2002-10-31 | 2004-10-07 | Predictive Systems Engineering, Ltd. | System and method for remote diagnosis of distributed objects |
| CN101532911A (en) * | 2009-04-24 | 2009-09-16 | 华北电力大学 | Large steam turbine-generator set rotor crack fault real-time diagnosis method |
| CN107066759A (en) * | 2017-05-12 | 2017-08-18 | 华北电力大学（保定） | A kind of Vibration Fault Diagnosis of Turbine Rotor method and device |
| CN110057581A (en) * | 2019-04-29 | 2019-07-26 | 杭州电子科技大学 | Rotary machinery fault diagnosis method based on interval type reliability rule-based reasoning |
| CN110469371A (en) * | 2019-07-16 | 2019-11-19 | 河北冀研能源科学技术研究院有限公司 | A method of solving rubbing failure in large capacity turbine start-up course |
| CN111312415A (en) * | 2020-03-02 | 2020-06-19 | 上海交通大学 | Fault diagnosis method and device |
-
2020
- 2020-09-29 CN CN202011046200.2A patent/CN112197968B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040199573A1 (en) * | 2002-10-31 | 2004-10-07 | Predictive Systems Engineering, Ltd. | System and method for remote diagnosis of distributed objects |
| CN101532911A (en) * | 2009-04-24 | 2009-09-16 | 华北电力大学 | Large steam turbine-generator set rotor crack fault real-time diagnosis method |
| CN107066759A (en) * | 2017-05-12 | 2017-08-18 | 华北电力大学（保定） | A kind of Vibration Fault Diagnosis of Turbine Rotor method and device |
| CN110057581A (en) * | 2019-04-29 | 2019-07-26 | 杭州电子科技大学 | Rotary machinery fault diagnosis method based on interval type reliability rule-based reasoning |
| CN110469371A (en) * | 2019-07-16 | 2019-11-19 | 河北冀研能源科学技术研究院有限公司 | A method of solving rubbing failure in large capacity turbine start-up course |
| CN111312415A (en) * | 2020-03-02 | 2020-06-19 | 上海交通大学 | Fault diagnosis method and device |
Non-Patent Citations (4)
| Title |
|---|
| 李江峰等: "某1000MW机组动静碰摩故障诊断与特征分析", 《广东电力》 * |
| 祝世兴等: "航空发动机高压转子振动故障分析", 《航空维修与工程》 * |
| 钱虹等: "基于故障样本的滚动轴承故障诊断置信规则库研究", 《电力科学与技术学报》 * |
| 钱虹等: "报警触发式蒸汽发生器传热管破裂事故诊断专家系统的研究", 《核动力工程》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113514144A (en) * | 2021-07-28 | 2021-10-19 | 郑州轻工业大学 | Unbalance-rubbing coupling fault detection method based on eddy current displacement sensor |
| CN113514144B (en) * | 2021-07-28 | 2022-07-26 | 郑州轻工业大学 | Unbalance-rubbing coupling fault detection method based on eddy current displacement sensor |
| CN115683687A (en) * | 2023-01-03 | 2023-02-03 | 成都大汇物联科技有限公司 | Dynamic and static rub-impact fault diagnosis method for hydroelectric mechanical equipment |
| CN115683687B (en) * | 2023-01-03 | 2023-04-18 | 成都大汇物联科技有限公司 | Dynamic and static bump-grinding fault diagnosis method for hydroelectric mechanical equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112197968B (en) | 2023-04-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Meng et al. | Fault analysis of wind power rolling bearing based on EMD feature extraction | |
| Li et al. | Bearing fault feature selection method based on weighted multidimensional feature fusion | |
| CN112197968A (en) | Confidence rule base-based intelligent diagnosis method for dynamic and static rub-impact faults of steam turbine rotor | |
| Pan et al. | Incipient fault detection of wind turbine large-size slewing bearing based on circular domain | |
| CN111797567A (en) | Deep learning network-based bearing fault classification method and system | |
| CN112633098B (en) | Rotary machine fault diagnosis method, system and storage medium | |
| Shi et al. | Generalized variable-step multiscale lempel-ziv complexity: A feature extraction tool for bearing fault diagnosis | |
| CN114894468B (en) | Chaos detection-based early weak fault diagnosis method for planetary gear box | |
| Cao et al. | Intelligent fault diagnosis of wind turbine gearbox based on Long short-term memory networks | |
| Huo et al. | A class-level matching unsupervised transfer learning network for rolling bearing fault diagnosis under various working conditions | |
| AU2021102131A4 (en) | Fault diagnosis method of rolling bearing based on generalized composite multi-scale weighted permutation entropy and supervised isometric mapping | |
| Lee et al. | A quantitative diagnostic method of feature coordination for machine learning model with massive data from rotary machine | |
| Mi et al. | A nonparametric cumulative sum-based fault detection method for rolling bearings using high-level extended isolated forest | |
| CN114046993A (en) | Slewing bearing state evaluation method based on multi-feature parameter fusion and FCM-HMM | |
| Li et al. | A novel adaptive STFT-SFA based fault detection method for nonstationary processes | |
| CN112504678B (en) | Motor bearing fault diagnosis method based on knowledge distillation | |
| Hu et al. | Fault diagnosis based on multi-scale redefined dimensionless indicators and density peak clustering with geodesic distances | |
| Meng et al. | Health condition identification of rolling element bearing based on gradient of features matrix and MDDCs-MRSVD | |
| Wang et al. | Multiscale sparsity measure fusion for bearing performance degradation assessment | |
| Jablonski et al. | Unsupervised detection of rotary machine imbalance based on statistical signal properties | |
| Qin et al. | Application of sensitive dimensionless parameters and PSO–SVM for fault classification in rotating machinery | |
| CN112541155B (en) | Steam turbine rotor unbalance detection method based on confidence rule base | |
| Yuan et al. | Fault detection of rolling bearing based on principal component analysis and empirical mode decomposition | |
| CN114742093A (en) | Rolling bearing fault diagnosis method and device based on time-frequency curve extraction and classification | |
| CN117851873B (en) | Bearing running state evaluation method and system based on dynamic contact angle |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |