CN109242148B - State monitoring method for fault growth trend - Google Patents

State monitoring method for fault growth trend Download PDF

Info

Publication number
CN109242148B
CN109242148B CN201810909122.0A CN201810909122A CN109242148B CN 109242148 B CN109242148 B CN 109242148B CN 201810909122 A CN201810909122 A CN 201810909122A CN 109242148 B CN109242148 B CN 109242148B
Authority
CN
China
Prior art keywords
fault
monitoring
index
state
formula
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.)
Active
Application number
CN201810909122.0A
Other languages
Chinese (zh)
Other versions
CN109242148A (en
Inventor
谭晓栋
黄娟
程旭
胡志鹏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Individual
Original Assignee
Individual
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Individual filed Critical Individual
Priority to CN201810909122.0A priority Critical patent/CN109242148B/en
Publication of CN109242148A publication Critical patent/CN109242148A/en
Application granted granted Critical
Publication of CN109242148B publication Critical patent/CN109242148B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/04—Manufacturing
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Landscapes

  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Engineering & Computer Science (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Development Economics (AREA)
  • Theoretical Computer Science (AREA)
  • Marketing (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Physics & Mathematics (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • Game Theory and Decision Science (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Educational Administration (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Manufacturing & Machinery (AREA)
  • Primary Health Care (AREA)
  • Testing And Monitoring For Control Systems (AREA)

Abstract

The invention discloses a state monitoring method for a fault growth trend, which comprises the following steps: s1, acquiring and preprocessing the fault growth trend information; s2, constructing a fault growth trend prediction model, analyzing the influence relation of different fault growth trends on a fault prediction result, and calculating the square sum of different monitoring point data on a fault prediction residual error; and S3, analyzing the significance of the fault prediction result, and feeding back to guide the optimal selection of the state monitoring point. The invention can avoid the influence of the trend information collected by the invalid monitoring points on the fault prediction result, improve the efficiency of state monitoring, and reduce the high cost and low reliability caused by too many monitoring points.

Description

State monitoring method for fault growth trend
Technical Field
The invention belongs to the field of fault analysis, and particularly relates to a fault growth trend state monitoring method, which is a novel fault growth trend monitoring method mainly provided for typical mechanical parts in an electromechanical system, such as bearings, gears, shafts and the like.
Background
The condition monitoring is the key for ensuring the healthy and stable operation of the equipment, and the requirements on the condition monitoring are higher and higher due to the sharp increase of the complexity and the integration level of the equipment. The method has the advantages that the equipment fault is required to be detected and isolated, the fault growth process is required to be tracked in real time, the failure time is required to be predicted, the predicted maintenance is conveniently realized, the maintenance decision is triggered before the complete failure of the equipment function caused by the fault, the major loss caused by the sudden failure is reduced, and the operation efficiency of the equipment is improved to the maximum extent. The state monitoring technology belongs to a complex system engineering, and relates to links such as monitoring point deployment, data acquisition and processing, feature extraction, fault tracking and monitoring, fault prediction and the like, which are dependent on each other, complement each other and influence each other, so that the state monitoring cost is high and the efficiency is low. Therefore, the problem that how to clear up the influence relationship of each link related to the state monitoring technology is urgently needed to be solved at present, so that each link is optimized to operate, and the state monitoring effect is finally improved.
Currently, the known methods have the following problems:
firstly, the existing state monitoring technology carries out isolated analysis on front-end technologies such as monitoring point deployment, data processing and feature extraction and rear-end fault diagnosis and prediction technologies, so that the overall effect of state monitoring has certain limitation;
secondly, the existing state monitoring technology sequentially performs monitoring point deployment, data processing, feature extraction, fault tracking monitoring and fault prediction, and does not consider optimization of links such as analysis feedback guidance front-end monitoring point deployment, data processing and feature extraction according to a rear-end monitoring conclusion, so that the state monitoring cost is high and the efficiency is low.
Disclosure of Invention
The invention aims to overcome the defects that the prior state monitoring method carries out isolated analysis and sequential operation on related technical links, the influence relationship and the feedback effect of each link are not analyzed, so that the state monitoring points are more, the efficiency and the reliability are low, and the like. According to the method, the fault growth trend information is obtained, a fault growth trend prediction model is constructed, a fault prediction conclusion is analyzed, and monitoring point selection at the front end is guided in a feedback mode. The invention finally provides an effective strategy for state monitoring, and ensures that the state monitoring is more efficient and accurate.
The technical scheme adopted by the invention is as follows:
a method for monitoring the state of a fault growth trend comprises the following steps:
s1, acquiring and preprocessing the fault growth trend information;
s2, constructing a fault growth trend prediction model, analyzing the influence relation of different fault growth trends on a fault prediction result, and calculating the square sum of different monitoring point data on a fault prediction residual error;
and S3, analyzing the significance of the fault prediction result, and feeding back to guide the optimal selection of the state monitoring point.
Specifically, the specific implementation method of S1 is as follows:
s1.1, building a fault growth test experiment platform, and setting fault states with different severity degrees;
s1.2, setting different monitoring points to collect original fault growth data;
s1.3, preprocessing the original data.
Further, the specific implementation method of S2 is as follows:
s2.1, extracting the state indexes of the preprocessed fault growth original data;
s2.2, calculating a fault growth trend prediction model parameter;
s2.3, acquiring a fault growth trend predicted value;
and S2.4, calculating the sum of squares of prediction residuals of the fault growth trend.
Further, the specific implementation method of S3 is as follows:
s3.1, calculating the statistic of the variable of the monitoring point;
s3.2, checking the significance of the variables of the monitoring points and selecting the monitoring points.
Preferably, the specific implementation method of S1.2 is: deploying monitoring points on a test experiment platform and acquiring data of fault states with different severity, wherein the method comprises the following specific steps:
s1.2.1, initially deploying monitoring points TI＝{t1,t2,...,tMM is the total number of the measuring points;
s1.2.2, collecting the data O ═ O of fault states of different severity of system1,O2,...,ON]TWhere N is the number of fault conditions of different severity, and Y is defined as (Y) the fault condition of different severity1 y2...yN)；
The specific implementation method of S1.3 is as follows:
preprocessing the data O for different severity fault conditions using equation (1):
Figure BDA0001761315310000031
in the formula, Oi＝[oi1,oi2,...,oiM]Data, mu, collected for M monitoring points in the ith fault stateiAnd σiAre each OiThe mean value and the standard deviation of (a),
Figure BDA0001761315310000032
the data is preprocessed.
Preferably, the specific implementation method of S2.1 is:
calculating the state index CI of the data O collected by M monitoring points under N fault states by using an equation (2):
Figure BDA0001761315310000033
wherein f (-) is a state index calculation function.
Preferably, the specific implementation method of S2.2 is:
and (4) calculating the parameters of the fault growth trend prediction model according to the formula (3):
B＝(XTX)-1XTY (3)
in the formula (I), the compound is shown in the specification,
Figure BDA0001761315310000041
y is a true fault growth vector and,
Figure BDA0001761315310000042
b is a fault growth trend prediction model parameter,
Figure BDA0001761315310000043
preferably, the specific implementation method of S2.3 is:
Figure BDA0001761315310000044
wherein X is a state index of a known fault state,
Figure BDA0001761315310000045
Figure BDA0001761315310000046
in order for the failure to increase the predicted value,
Figure BDA0001761315310000047
b is a parameter of the failure growth trend prediction model,
Figure BDA0001761315310000048
preferably, the specific implementation method of S2.4 is:
the sum of squared fault growth trend prediction residuals Q is calculated using equation (5):
Figure BDA0001761315310000049
in the formula, yiIs the true value in the i-th fault severity state,
Figure BDA0001761315310000051
for y calculated using equation (4)iThe predicted value of (2).
Preferably, the specific implementation method of S3.1 is:
the statistics F for M monitoring points are calculated using equation (6):
Figure BDA0001761315310000052
in the formula, biFor the ith parameter calculated in equation (3),
Figure BDA0001761315310000053
n is the number of fault states, xtiThe state index value of the ith monitoring point data in the t fault state is obtained by calculation according to the formula (2),
Figure BDA0001761315310000054
is the ithThe average value of state index values of the data of the monitoring points in N fault states;
q is the sum of squares of prediction residuals of the calculated fault growth trend of the formula (5), N is the number of fault states, FiF distribution with a numerator degree of freedom of 1 and a denominator degree of freedom of N-2;
the specific implementation method of S3.2 is as follows:
the monitoring point T is selected using equation (7):
Figure BDA0001761315310000055
in the formula, tiIs the ith monitor point, F in the set of monitor points TiIs the statistic of the ith monitoring point calculated by the formula (6), M is the total number of the monitoring points, FαF distribution with a numerator degree of freedom of 1 and a denominator degree of freedom of N-2 at a significance level of alpha 0.05, i.e. F0.05(1, N-2) obtained by referring to the F distribution table.
The invention has the beneficial effects that:
according to the method, the influence relation of different monitoring point data on the fault prediction result can be quantized by constructing a fault growth trend prediction model and calculating the square sum of the fault prediction residual errors; by analyzing the statistical characteristics of different monitoring points on the fault growth trend prediction, the significance of the monitoring point data on the fault prediction can be checked, the monitoring point data which has no significant effect on the fault growth trend prediction result is removed, the monitoring points are optimally selected through the feedback of the prediction result, the influence of invalid monitoring point trend information on the fault prediction result is avoided, the state monitoring efficiency is improved, and the high cost and the low reliability caused by too many monitoring points are reduced.
Drawings
FIG. 1 is a schematic flow diagram of an embodiment of the present invention.
FIG. 2 is a raw waveform collected from three monitoring points of a 0.021 inch severity ring fault in accordance with an embodiment of the present invention.
Detailed Description
The invention is further described with reference to the following figures and specific embodiments.
As shown in fig. 1, the method for monitoring a state of a failure growth trend of the present embodiment includes the steps of:
firstly, acquiring and preprocessing the fault growth trend information.
The method comprises the following steps:
s11, building a fault growth test experiment platform, and setting fault states with different severity degrees;
s12, setting different monitoring points to collect the original data of the fault growth;
s13, preprocessing the original data;
and secondly, constructing a fault growth trend prediction model, analyzing the influence relation of different fault growth trends on a fault prediction result, and calculating the sum of squares of different monitoring point data on fault prediction residuals.
The method comprises the following steps:
s21, extracting the state indexes of the preprocessed fault growth original data;
s22, calculating the failure growth trend prediction model parameters;
s23, acquiring a failure growth trend predicted value;
and S24, calculating the sum of squares of the fault prediction residuals of different monitoring point data.
And thirdly, analyzing the significance of the fault prediction result, and feeding back to guide the optimal selection of the state monitoring point.
The method comprises the following steps:
s31, calculating the statistic of the monitoring point variable;
and S32, checking the significance of the monitoring point variable and selecting the monitoring point.
The process of the invention is described in detail below:
1. and acquiring and preprocessing the fault growth trend information.
S1.1, building a fault growth test experiment platform and setting fault states with different severity degrees.
S1.2, setting different monitoring points to collect original fault growth data, deploying the monitoring points on a test experiment platform, and collecting data of fault states with different severity degrees, wherein the method comprises the following specific steps:
firstly, according to the system function structure, fault propagation characteristics and state monitoring requirements, a monitoring point T is initially deployedI＝{t1,t2,...,tMM is the total number of the measuring points;
then, collecting data O ═ O of fault states of different severity of the system1,O2,...,ON]TWhere N is the number of fault conditions of different severity, Oi＝[oi1,oi2,...,oiM]Defining fault states Y with different severity degrees as (Y) for data collected by M monitoring points in ith fault state1y2...yN)。
S1.3, preprocessing original data: preprocessing the data O for different severity fault conditions using equation (1):
Figure BDA0001761315310000071
in the formula, Oi＝[oi1,oi2,...,oiM]Data, mu, collected for M monitoring points in the ith fault stateiAnd σiAre each OiThe mean value and the standard deviation of (a),
Figure BDA0001761315310000081
the data is preprocessed.
2. And constructing a fault growth trend prediction model, analyzing the influence relationship of different fault growth trends on a fault prediction result, and calculating the sum of squares of different monitoring point data on a fault prediction residual error.
S2.1, calculating the state index CI of the data O collected by M monitoring points in N fault states by using the formula (2):
Figure BDA0001761315310000082
wherein f (-) is a state index calculation function.
The state index calculation function is a time domain index, a frequency domain index or a time-frequency domain index;
the time domain index is a mean value, a root mean square value, a square root amplitude value, an absolute mean value, a skewness, a kurtosis, a variance, a peak value, a standard deviation, a peak-peak value, an average power, a waveform index, a peak index, a pulse index, a margin index, a skewness index or a kurtosis index;
the frequency domain index is a power spectrum or high-order statistic;
the time-frequency domain index is wavelet entropy or energy entropy.
S2.2, calculating the failure growth trend prediction model parameters according to the formula (3):
B＝(XTX)-1XTY (3)
in the formula (I), the compound is shown in the specification,
Figure BDA0001761315310000083
y is a true fault growth vector and,
Figure BDA0001761315310000084
b is a fault growth trend prediction model parameter,
Figure BDA0001761315310000091
s2.3, calculating a failure growth trend predicted value by using an equation (4)
Figure BDA0001761315310000092
Figure BDA0001761315310000093
Wherein X is a state index of a known fault state,
Figure BDA0001761315310000094
Figure BDA0001761315310000095
in order for the fault to grow the predicted value,
Figure BDA0001761315310000096
b is a fault growth trend prediction model parameter,
Figure BDA0001761315310000097
s2.4, calculating the square sum Q of the prediction residual errors of the fault growth trend by using an equation (5):
Figure BDA0001761315310000098
in the formula, yiIs the true value in the i-th fault severity condition,
Figure BDA0001761315310000099
for y calculated using equation (4)iThe predicted value of (2).
3. And analyzing the significance of the fault prediction result, and feeding back to guide the optimal selection of the state monitoring point.
S3.1, calculating statistics F of M monitoring points by using an equation (6):
Figure BDA00017613153100000910
in the formula, biIs the i-th parameter calculated in equation (3).
Figure BDA00017613153100000911
N is the number of fault states, xtiThe state index value of the ith monitoring point data in the t fault state is obtained by calculation according to the formula (2),
Figure BDA0001761315310000101
the average value of state index values of the ith monitoring point data in N fault states;
q is the sum of squares of the prediction residuals of the increasing trend of the faults calculated by the formula (5), N is the number of the fault states, and FiThe F distribution is such that the numerator degree of freedom is 1 and the denominator degree of freedom is N-2.
S3.2, selecting a monitoring point T using equation (7):
Figure BDA0001761315310000102
in the formula, tiIs the ith monitor point, F in the set of monitor points TiIs the statistic of the ith monitoring point calculated by the formula (6), M is the total number of the monitoring points, FαF distribution with a numerator degree of freedom of 1 and a denominator degree of freedom of N-2 at a significance level of 0.05, namely F0.05(1, N-2) obtained by referring to the F distribution table.
The invention provides a novel fault growth trend state monitoring method, which comprises the following key points: the method comprises the steps of establishing a fault growth trend described by monitoring points, analyzing the influence relation of different fault growth trends on a fault prediction result, calculating the square sum of different monitoring point data on a fault prediction residual error, analyzing the significance of the monitoring point data on the fault prediction result, and finally optimizing and selecting state monitoring points.
The points to be protected by the invention are as follows: a fault prediction model based on the fault growth trend described by the monitoring points; the monitoring point performs significance test on the fault prediction result; and feeding back a fault prediction result to guide a state monitoring point selection scheme.
The main idea of the invention is elaborated by taking a rolling bearing fault simulation test experiment table as an example:
1. and acquiring the fault growth trend information.
S1.1, establishing a fault growth test experimental platform.
The rolling bearing fault simulation test experiment platform comprises a 2hp motor (1 hp-746 w), a torque sensor, an indicator and an electric control device. The experimental bearing is a 6205-2RS JEMSKF type deep groove ball bearing, the power of the motor is 746W, and the rotating speed of the input shaft is 1772 r/min. The bearing uses the spark machining to process four kinds of inner ring faults with the fault severity being divided into 0, 0.007 inches, 0.014 inches and 0.021 inches.
S1.2, setting different monitoring points to collect fault growth data.
On-test experimental platformThe motor driving end, the motor fan end and the upper part of the bearing seat on the frame of the base are respectively provided with a monitoring point, namely a monitoring point set TI＝{t1,t2,t3Acquiring vibration data corresponding to four severity faults, acquiring two groups of data in each fault state, and totaling 8 groups of data, namely O ═ 31,O2,O3,O4,O5,O6,O7,O8]TFig. 2 shows the original waveforms collected at three monitoring points of a 0.021 inch severity ring fault. The true fault growth vector Y is (000.0070.0070.0140.0140.0210.021) corresponding to 0, 0.007 inches, 0.014 inches, 0.021 inches and 0.021 inches inner ring fault states, i.e. the number of fault states N is 8, Oi＝[oi1,oi2,oi3]And collecting data sets for 3 monitoring points in the ith fault state, wherein the total number M of the monitoring points is 3.
S1.3, preprocessing data: the data O for different severity fault conditions are preprocessed using equation (1).
2. And constructing a fault growth trend prediction model, analyzing the influence relationship of different fault growth trends on a fault prediction result, and calculating the sum of squares of different monitoring point data on fault prediction residuals.
S2.1, setting a state index function in the formula (2) as a root mean square value, and calculating the root mean square values of 3 monitoring point data under 8 inner ring fault states.
Figure BDA0001761315310000111
S2.2, calculating a fault growth trend prediction model parameter B according to the formula (3):
B＝(0.0193 0.1654 -0.2112 -0.2683)′
s2.3, calculating a failure growth trend predicted value by using an equation (4)
Figure BDA0001761315310000121
Figure BDA0001761315310000122
S2.4, calculating the square sum Q of the prediction residuals of the fault growth trend by using the formula (5):
Q＝3.3574×10-4
3. and analyzing the significance of the fault prediction result, and feeding back to guide the optimal selection of the state monitoring point.
S3.1, calculating statistic F of 3 monitoring points by using formula (6):
Fi＝(84.6847 0.4010 0.5100)
s3.2, inquiring the F distribution table to obtain F0.05(1,6) ═ 5.99, and the monitoring point T is selected according to equation (7)*：
T*＝{t1}
Therefore, aiming at the increasing trend of the faults of the inner ring of the rolling bearing fault simulation test experiment platform, the monitoring point t is adopted1(i.e., monitoring points disposed at the drive end of the motor) track and monitor the inner ring failure growth process of the rolling bearing.
The invention is not limited to the above alternative embodiments, and any other various forms of products can be obtained by anyone in the light of the present invention, but any changes in shape or structure thereof, which fall within the scope of the present invention as defined in the claims, fall within the scope of the present invention.

Claims (5)

1. A method for monitoring the state of a fault growth trend is characterized in that: the method comprises the following steps:
s1, acquiring and preprocessing the fault growth trend information;
s2, constructing a fault growth trend prediction model, analyzing the influence relation of different fault growth trends on a fault prediction result, and calculating the square sum of different monitoring point data on a fault prediction residual error;
s3, analyzing the significance of the monitoring point data to the fault prediction result, and feeding back to guide the state monitoring point to be optimized and selected;
the specific implementation method of the S2 comprises the following steps:
s2.1, calculating the state index CI of the data O collected by M monitoring points in N fault states by using the formula (1):
Figure FDA0003587307550000011
wherein f (·) is a state index calculation function, the state index is a time domain index, a frequency domain index or a time-frequency domain index, the time domain index is a mean value, a root mean square value, a square root amplitude, an absolute mean value, a skewness, a kurtosis, a variance, a peak value, a standard deviation, a peak-peak value, an average power, a waveform index, a peak value index, a pulse index, a margin index, a skewness index or a kurtosis index, the frequency domain index is a power spectrum or a high-order statistic, and the time-frequency domain index is a wavelet entropy or an energy entropy;
s2.2, calculating the parameters of the fault growth trend prediction model according to the formula (2):
B＝(XTX)-1XTY (2)
in the formula (I), the compound is shown in the specification,
Figure FDA0003587307550000012
y is the true fault growth vector and,
Figure FDA0003587307550000013
b is a fault growth trend prediction model parameter,
Figure FDA0003587307550000021
s2.3, calculating a failure growth trend predicted value by using the formula (3)
Figure FDA0003587307550000022
Figure FDA0003587307550000023
Wherein X is a state index of a known fault state,
Figure FDA0003587307550000024
Figure FDA0003587307550000025
in order for the failure to increase the predicted value,
Figure FDA0003587307550000026
b is a fault growth trend prediction model parameter,
Figure FDA0003587307550000027
s2.4, calculating the square sum Q of the prediction residuals of the fault growth trend by using the formula (4):
Figure FDA0003587307550000028
in the formula, yiIs the true value in the i-th fault severity state,
Figure FDA0003587307550000029
for y calculated using equation (3)iThe predicted value of (2).
2. A method for monitoring the status of a trend of increasing faults as defined in claim 1, wherein: the specific implementation method of the S1 comprises the following steps:
s1.1, building a fault growth test experiment platform, and setting fault states with different severity degrees;
s1.2, setting different monitoring points to collect original fault growth data;
s1.3, preprocessing the original data.
3. A method for monitoring the status of a trend of increasing faults as set forth in claim 2, wherein: the specific implementation method of the S3 comprises the following steps:
s3.1, calculating the statistic of the variable of the monitoring point;
and S3.2, checking the significance of the variables of the monitoring points and selecting the monitoring points.
4. A method for monitoring the status of a growing trend of faults according to claim 3, characterized in that: the specific implementation method of S1.2 is as follows: deploying monitoring points on a test experiment platform and acquiring data of fault states with different severity, wherein the method comprises the following specific steps:
s1.2.1, initially deploying monitoring points TI＝{t1,t2,...,tMM is the total number of the measuring points;
s1.2.2, collecting the data O ═ O of fault states of different severity of system1,O2,...,ON]TWhere N is the number of fault conditions of different severity, and Y is defined as (Y) the fault condition of different severity1 y2...yN)；
The specific implementation method of S1.3 is as follows:
preprocessing the data O for different severity fault conditions using equation (5):
Figure FDA0003587307550000031
in the formula, Oi＝[oi1,oi2,...,oiM]Data, mu, collected for M monitoring points in the ith fault stateiAnd σiAre each OiThe mean value and the standard deviation of (a),
Figure FDA0003587307550000032
the data is preprocessed.
5. The method for monitoring the state of the fault growth trend according to claim 4, wherein the method comprises the following steps: the specific implementation method of S3.1 is as follows:
the statistics F for M monitoring points are calculated using equation (6):
Figure FDA0003587307550000033
in the formula, biFor the ith parameter calculated in equation (2),
Figure FDA0003587307550000034
n is the number of fault states, xtiThe state index value of the ith monitoring point data in the t fault state is obtained by calculation according to the formula (1),
Figure FDA0003587307550000042
the average value of state index values of the ith monitoring point data in N fault states;
q is the sum of squares of the prediction residuals of the fault growth trend calculated by the formula (4), N is the number of fault states, FiF distribution with a numerator degree of freedom of 1 and a denominator degree of freedom of N-2;
the specific implementation method of S3.2 is as follows:
the monitoring point T is selected using equation (7):
Figure FDA0003587307550000041
in the formula, tiIs the ith monitor point, F in the set of monitor points TiIs the statistic of the ith monitoring point calculated by the formula (6), M is the total number of the monitoring points, FαF distribution with a numerator degree of freedom of 1 and a denominator degree of freedom of N-2 at a significance level of alpha 0.05, i.e. F0.05(1, N-2) obtained by referring to the F distribution table.
CN201810909122.0A 2018-08-10 2018-08-10 State monitoring method for fault growth trend Active CN109242148B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810909122.0A CN109242148B (en) 2018-08-10 2018-08-10 State monitoring method for fault growth trend

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810909122.0A CN109242148B (en) 2018-08-10 2018-08-10 State monitoring method for fault growth trend

Publications (2)

Publication Number Publication Date
CN109242148A CN109242148A (en) 2019-01-18
CN109242148B true CN109242148B (en) 2022-06-17

Family

ID=65070096

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810909122.0A Active CN109242148B (en) 2018-08-10 2018-08-10 State monitoring method for fault growth trend

Country Status (1)

Country Link
CN (1) CN109242148B (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2008098301A1 (en) * 2007-02-14 2008-08-21 The Commonwealth Of Australia Monitoring the structural health of components
CN107229272A (en) * 2017-06-22 2017-10-03 谭晓栋 A kind of sensor optimization dispositions method based on failure growth trend Controlling UEP
CN107782551A (en) * 2017-10-30 2018-03-09 电子科技大学 Method for evaluating damage degree of mechanical part

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2008098301A1 (en) * 2007-02-14 2008-08-21 The Commonwealth Of Australia Monitoring the structural health of components
CN107229272A (en) * 2017-06-22 2017-10-03 谭晓栋 A kind of sensor optimization dispositions method based on failure growth trend Controlling UEP
CN107782551A (en) * 2017-10-30 2018-03-09 电子科技大学 Method for evaluating damage degree of mechanical part

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
基于状态监测的风电机组主轴承早期故障预测方法;张小田 等;《广东电力》;20121125;第25卷(第11期);第6-9页 *
基于轴承温度模型的风电机组故障预测研究;丁佳煜 等;《可再生能源》;20180220;第36卷(第2期);第276-282页 *

Also Published As

Publication number Publication date
CN109242148A (en) 2019-01-18

Similar Documents

Publication Publication Date Title
CN106682814B (en) Wind turbine generator fault intelligent diagnosis method based on fault knowledge base
CN110907066B (en) Wind turbine generator gearbox bearing temperature state monitoring method based on deep learning model
CN112906473B (en) Fault diagnosis method for rotary equipment
CN108171263A (en) Based on the Fault Diagnosis of Roller Bearings for improving variation mode decomposition and extreme learning machine
CN111475921A (en) Tool residual life prediction method based on edge calculation and L STM network
CN111581597A (en) Wind turbine generator gearbox bearing temperature state monitoring method based on self-organizing kernel regression model
CN103758696A (en) SCADA (supervisory control and data acquisition) temperature parameter based wind turbine set security evaluation method
CN104200396A (en) Wind driven generator part fault early warning method
CN109556863B (en) MSPAO-VMD-based large turntable bearing weak vibration signal acquisition and processing method
Butler et al. A feasibility study into prognostics for the main bearing of a wind turbine
Zhan et al. Health assessment methods for wind turbines based on power prediction and mahalanobis distance
CN114861827A (en) Coal mining machine prediction diagnosis and health management method based on multi-source data fusion
Yang et al. Gearbox fault diagnosis based on artificial neural network and genetic algorithms
CN113221453A (en) Fault monitoring and early warning method for output shaft of gearbox of wind turbine generator
CN114061743A (en) Vibration monitoring method, device, equipment and medium for wind generating set
CN113268835A (en) Pumped storage hydroelectric generating set health assessment method and device based on extended isolation forest and cloud model
CN111610024A (en) Gear crack fault diagnosis method based on vibration signals
CN117195121A (en) Wind turbine generator abnormal state identification method and system based on improved countermeasure automatic encoder
CN114971351A (en) Equipment multi-index health condition assessment method based on entropy method
CN109242148B (en) State monitoring method for fault growth trend
CN109543357B (en) Fault degree quantitative evaluation method for multivariate regression model optimization
CN109145773A (en) A kind of failure prediction method of multi-source tendency information fusion
CN115791159A (en) Gear fault online real-time diagnosis method based on vibration signals
Jia et al. A comparison study of hidden Markov model and particle filtering method: Application to fault diagnosis for gearbox
Song et al. Framework of designing an adaptive and multi-regime prognostics and health management for wind turbine reliability and efficiency improvement

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