CN103500267A - Method for judging assembling reliability degree of bolt connection device with state information - Google Patents

Method for judging assembling reliability degree of bolt connection device with state information Download PDF

Info

Publication number
CN103500267A
CN103500267A CN201310403584.2A CN201310403584A CN103500267A CN 103500267 A CN103500267 A CN 103500267A CN 201310403584 A CN201310403584 A CN 201310403584A CN 103500267 A CN103500267 A CN 103500267A
Authority
CN
China
Prior art keywords
subspace
index
utilize
connection device
bolt connection
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.)
Pending
Application number
CN201310403584.2A
Other languages
Chinese (zh)
Inventor
何正嘉
孙闯
曹宏瑞
张周锁
陈雪峰
李兵
訾艳阳
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.)
Xian Jiaotong University
Original Assignee
Xian Jiaotong University
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 Xian Jiaotong University filed Critical Xian Jiaotong University
Priority to CN201310403584.2A priority Critical patent/CN103500267A/en
Publication of CN103500267A publication Critical patent/CN103500267A/en
Pending legal-status Critical Current

Links

Images

Landscapes

  • Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)

Abstract

The invention discloses a method for judging the assembling reliability degree of a bolt connection device with state information. According to the method, firstly, the state information of the device is acquired through a motivation test; then, the acquired state information is preprocessed through a feature extraction method to obtain a state feature matrix; the inherent mode of the feature matrix is analyzed through a mode recognition algorithm, and a state subspace corresponding to the state feature matrix is constructed; the similarity index of a normal state subspace and the current state subspace is calculated, and the index reflects the similarity of the two state subspaces, and can describe the closeness degree of the device performance at the current moment and a normal state; finally, the similarity index is used as a reliability index to quantitatively reflect the assembling reliability degree of the device. The method does not depend on a large number of failure samples, is simple and easy to implement, has the advantages of being reliable in result, strong in practicality and the like, is suitable for evaluating the assembling reliability degree of the bolt connection device in real time on site, and has engineering application value.

Description

Utilize status information to judge the method for bolt connection device assembling fiduciary level
Technical field
The present invention relates to the decision method of plant equipment assembling fiduciary level, particularly a kind of method of utilizing status information to judge bolt connection device assembling fiduciary level.
Background technology
It is a kind of mechanical part type of attachment of extensive employing that bolt connects, and is applied in the crucial plant equipment such as aeromotor, gas turbine.This version utilizes bolt to connect independently mechanical part, forms a plant equipment with certain dynamics.The tight power deficiency caused by bolt looseness will change the assembly quality of equipment, and then the equipment that affects completes its predetermined function.Therefore, the assembling fiduciary level judges that reliability and the security of support equipment are significant to the identification confined state.
Reliability analyzing method commonly used is supposed based on Conditions of General Samples at present, according to batch device fiduciary level statistical information, obtains the global failure rule.Use traditional reliability analyzing method assessment plant equipment fiduciary level to have the difficulty of following two aspects: 1) traditional reliability analyzing method, based on large sample theory, is difficult to effectively process small sample and single sample fail data; 2) traditional reliability analyzing method individual character that consideration equipment is not degenerated, lack the accuracy to the single device Reliability assessment.
Status information is the external manifestation of the inherent health status of equipment, can reflect the degree of reliability of equipment, for the equipment reliability assessment provides foundation.When bolt connection device assembly quality is degenerated, the dynamic characteristic of equipment, as rigidity, damping etc. will change, can show as the change that the features such as the Time Domain Amplitude of equipment status information, frequency-domain structure are descended in externally excitation.Therefore, analyze the difference of normal confined state and current confined state status information of equipment, and then the structure RELIABILITY INDEX, for single device assembling Reliability assessment provides approach.
Summary of the invention
The purpose of this invention is to provide a kind of method of rationally, effectively judging bolt connection device assembling fiduciary level, the method can solve judges separate unit bolt connection device assembling fiduciary level under the condition of inefficacy sample deficiency.
For reaching above purpose, the present invention takes following technical scheme to be achieved:
A kind of method of utilizing status information to judge bolt connection device assembling fiduciary level, is characterized in that, comprises following steps:
(1) structural regime eigenmatrix
The bolt connection device is carried out to the stand Knock test, utilize the status information of vibration acceleration sensor collecting device under externally encouraging, the time domain of computing mode information and frequency domain statistical nature, utilize the latent structure status flag matrix extracted;
(2) structural regime subspace
At first, utilize the Nonlinear Mapping function that eigenmatrix is mapped to high-dimensional feature space, and utilize core locality preserving projections method characteristics of decomposition matrix to obtain projection matrix; Secondly, utilize orthogonalization method to the vectorial orthogonalization in projection matrix, obtain one group of unit orthogonal vector; Finally, utilize these group unit's orthogonal vector to form subspace method;
(3) computing mode subspace index of similarity
At first, construct respectively normal condition subspace and current state subspace; Then, utilize matrix norm to define the index of similarity of two class subspace methods; Described normal condition subspace is:
Figure BDA0000378338520000021
Figure BDA0000378338520000022
Wherein i=1 ..., r 1mean subspace base vector, mean status flag matrix when assembling is good, γ i, i=1 ..., r 1mean weight vector, r 1mean the subspace dimension; Described current state subspace is:
Figure BDA0000378338520000025
Figure BDA0000378338520000026
Wherein
Figure BDA0000378338520000027
i=1 ..., r 2mean subspace base vector,
Figure BDA0000378338520000028
the status flag matrix that means current time,
Figure BDA0000378338520000029
i=1 ..., r 2mean weight vector, r 2mean the subspace dimension; Then, utilize matrix norm to define the index of similarity SI of two class subspace methods, be expressed as:
SI = | | ( S 1 ) T S 2 | | 2 | | ( S 1 ) T | | 2 | | S 2 | | 2 ;
(4) definition RELIABILITY INDEX
Because the variation range of index of similarity SI is [0,1], and the similarity degree that can reflect current time bolt connection device state and its normal condition, therefore the assembling fiduciary level using this index as RELIABILITY INDEX R quantitative response bolt connection device is defined as RELIABILITY INDEX:
R = SI = | | ( S 1 ) T S 2 | | 2 | | ( S 1 ) T | | 2 | | S 2 | | 2 ;
(5) fiduciary level is judged
Can judge that with RELIABILITY INDEX R the assembling fiduciary level of bolt connection device: R approaches 1 and means that equipment assembly quality and normal condition approach this moment; Otherwise R approaches 0 and means that the equipment assembly quality departs from normal condition this moment.
With based on Conditions of General Samples hypothesis, according to the classic method of batch device fiduciary level statistical information, compare, the inventive method is by the status information of exciting test equipment; Utilize feature extracting method to obtain the status flag matrix to obtained status information pre-service; The inherent natural mode of reuse mode recognizer analytical characteristic matrix, structure is corresponding to the subspace method of status flag matrix; Then calculate the index of similarity of normal condition subspace and current state subspace, this index has reflected the similarity of two class subspace methods, can depict the degree of closeness of current time equipment performance and normal condition; The assembling level of reliability using this index of similarity as RELIABILITY INDEX quantitative response equipment, its advantage is, assembling fiduciary level for single device is judged, do not rely on a large amount of inefficacy samples, simple, there are reliable results, the characteristics such as real-time, be applicable to the assembling fiduciary level of on-the-spot real-time assessment bolt connection device, be conducive to improve equipment safety in operation and reliability, there is the engineering using value.
The accompanying drawing explanation
Below in conjunction with the drawings and the specific embodiments, the present invention is described in further detail.
Fig. 1 is certain bolt connection device status information time domain waveform under different confined states.Wherein, the time domain waveform figure that Fig. 1 (a) is status information when normal confined state; Fig. 1 (b) is the status information time domain waveform figure when the loosening half-turn of 3 stay-bolts; Fig. 1 (c) is the status information time domain waveform figure when the loosening half-turn of 12 stay-bolts; Fig. 1 (d) is the status information time domain waveform figure when the loosening half-turn of 24 stay-bolts; Fig. 1 (e) is the status information time domain waveform figure when the loosening circle of 24 stay-bolts; Fig. 1 (f) is the status information time domain waveform figure at the loosening circle half of 24 stay-bolts.In figure, horizontal ordinate means the time, and unit is second; Ordinate means the vibration signal amplitude, and unit is g.
Fig. 2 is certain bolt connection device status information spectrogram under different confined states.Wherein, the spectrogram that Fig. 2 (a) is status information when normal confined state; Fig. 2 (b) is the status information spectrogram when the loosening half-turn of 3 stay-bolts; Fig. 2 (c) is the status information spectrogram when the loosening half-turn of 12 stay-bolts; Fig. 2 (d) is the status information spectrogram when the loosening half-turn of 24 stay-bolts; Fig. 2 (e) is the status information spectrogram when the loosening circle of 24 stay-bolts; Fig. 2 (f) is the status information spectrogram at the loosening circle half of 24 stay-bolts.In figure, horizontal ordinate means frequency content, and unit is Hz; Ordinate means frequency amplitude, and unit is g.
The typical temporal signatures changing trend diagram that Fig. 3 is certain bolt connection device status information.In figure, horizontal ordinate means the sample sequence number; Ordinate representation feature amplitude; S1-S6 means six kinds of confined states.
The typical frequency domain character changing trend diagram that Fig. 4 is certain bolt connection device status information.In figure, horizontal ordinate means the sample sequence number; Ordinate representation feature amplitude; S1-S6 means six kinds of confined states.
The result of determination figure that Fig. 5 is certain bolt connection device assembling fiduciary level.In figure, horizontal ordinate means six kinds of confined states, and ordinate means every kind of equipment assembling fiduciary level R that confined state is corresponding.
Its embodiment
With reference to Fig. 5, when confined state is good, the fiduciary level result of determination of equipment is 1, and along with assembly quality is backward, assembling RELIABILITY INDEX R descends gradually.The variation tendency of RELIABILITY INDEX R has reflected the backward process of equipment assembly quality well.
The present invention utilizes status information to judge that the bolt connection device assembles fiduciary level and implements by following concrete steps:
1) status flag matrix construction:
The bolt connection device is carried out to the stand Knock test, utilize the status information of vibration acceleration sensor collecting device under externally encouraging.The Time-domain Statistics feature of computing mode information and frequency domain statistical nature, utilize the latent structure status flag matrix extracted.
2) subspace method structure:
At first, utilize the Nonlinear Mapping function that eigenmatrix is mapped to high-dimensional feature space, and utilize core locality preserving projections method characteristics of decomposition matrix to obtain projection matrix; Secondly, utilize orthogonalization method to the vectorial orthogonalization in projection matrix, obtain one group of unit orthogonal vector; Finally, utilize these group unit's orthogonal vector to form subspace method.
3) the subspace index of similarity calculates:
At first, structure bolt connection device assembles subspace method S when good 1, be expressed as:
Figure BDA0000378338520000041
Be called the normal condition subspace, wherein
Figure BDA0000378338520000043
i=1 ..., r 1mean subspace base vector,
Figure BDA0000378338520000044
mean status flag matrix when assembling is good, γ i, i=1 ..., r 1mean weight vector, r 1mean the subspace dimension.The subspace method S of structure current time 2, be expressed as:
Figure BDA0000378338520000046
Be called the current state subspace, wherein
Figure BDA0000378338520000047
i=1 ..., r 2mean subspace base vector,
Figure BDA0000378338520000048
the status flag matrix that means current time,
Figure BDA0000378338520000051
i=1 ..., r 2mean weight vector, r 2mean the subspace dimension.Then, utilize matrix norm to define the index of similarity SI of two class subspace methods, be expressed as:
SI = | | ( S 1 ) T S 2 | | 2 | | ( S 1 ) T | | 2 | | S 2 | | 2 .
4) RELIABILITY INDEX definition:
From the character of matrix norm, || (S 1) ts 2|| 2, || (S 1) t|| 2, || S 2|| 2all be not less than 0, and || (S 1) ts 2|| 2≤ || (S 1) t|| 2|| S 2|| 2, therefore, the variation range of index of similarity is [0,1], that is:
0≤SI≤1
Index of similarity has reflected the similarity between two class subspace methods, and variation range is [0,1], the degree of reliability that can be using it as RELIABILITY INDEX R quantitative response equipment state.Therefore RELIABILITY INDEX is defined as:
R = SI = | | ( S 1 ) T S 2 | | 2 | | ( S 1 ) T | | 2 | | S 2 | | 2 .
5) the assembling fiduciary level is judged:
Can judge that with RELIABILITY INDEX R the assembling fiduciary level of bolt connection device: R approaches 1 and means that equipment assembly quality and normal condition approach this moment, the assembling fiduciary level approaches 1; Otherwise R approaches 0 and means that the equipment assembly quality departs from normal condition this moment, the assembling fiduciary level approaches 0.
Below provide a concrete application example process, simultaneous verification the validity of the present invention in engineering application.
Assembling fiduciary level to a kind of removable disk drum type aeroengine rotor judged, this rotor mainly comprises the parts such as wheel disc, drum barrel and stay-bolt, utilizes stay-bolt that wheel discs at different levels are strained, and forms a rotor structure integral body.Simulated six kinds of rotor confined states by loosening stay-bolt, definition and the method for expressing of every kind of confined state are as shown in table 1.
Method for expressing, loose bolts number and the aeration level of every kind of confined state have been defined in table 1.Therefrom can find out, state S1 means there is no bolt looseness, is normal (benchmark) state; In state S2, have 3 stay-bolts loosening, each bolt looseness degree is 0.5 circle; For state S3-S6 can the rest may be inferred.From state S1 to S6, the assembling aeration level of rotor is deepened successively.
The definition of table 1 confined state and method for expressing
Figure BDA0000378338520000061
Under six kinds of confined states of rotor, use vibrator excitation rotor, and utilize the vibration acceleration response of vibration acceleration sensor and data acquisition equipment collection storage rotor structure, i.e. status information.
Under six kinds of confined states, time domain waveform, frequency-domain waveform, time domain characteristic feature and the frequency domain characteristic feature of this rotor status information are respectively referring to Fig. 1, Fig. 2, Fig. 3 and Fig. 4.As can be seen from Figure 1, there is the concussion decay characteristics in status information, and decay is very fast.Not significantly difference of the time domain waveform of status information under six kinds of confined states, can not directly judge the aeration level of rotor according to time domain waveform.As can be seen from Figure 2, spectrum energy concentrates on high band, i.e. the 1600Hz-3200Hz frequency range.Similar with time domain waveform, the difference of spectrogram is also not obvious, can not directly judge the loosening state of rotor from frequency-domain waveform.Can find out time domain characteristic feature T from Fig. 3 and Fig. 4 2(root-mean-square value) and frequency domain characteristic feature F 3though can distinguish preferably state S1, S2 and S6, can not distinguish well state S3, S4 and S5, and feature there is no the variation tendency of obvious variation tendency and aeration level inconsistent.
The status information recorded under normal confined state is divided into to two parts, and wherein a part is as normal condition information, and the information recorded under the loosening state of a part and assembling in addition is as current state information.Utilize the method for the invention to calculate the similarity of normal condition and current state, and then obtain the assembling fiduciary level of every kind of state lower rotor part.The fiduciary level result of determination is in Table 2 and Fig. 5.Therefrom can find out, when confined state is good, RELIABILITY INDEX R equals 1; Along with the stay-bolt aeration level is deepened gradually, assembly quality is backward, and RELIABILITY INDEX R also descends thereupon, and the downtrending of fiduciary level is consistent with the intensification process of bolt looseness degree.The fiduciary level result of determination is consistent with aeroengine rotor practical set state, has verified the validity of the method for the invention.
Table 2 removable disk drum type aeroengine rotor assembling fiduciary level result of determination
Figure BDA0000378338520000062

Claims (1)

1. a method of utilizing status information to judge bolt connection device assembling fiduciary level, is characterized in that, comprises following steps:
(1) structural regime eigenmatrix
The bolt connection device is carried out to the stand Knock test, utilize the status information of vibration acceleration sensor collecting device under externally encouraging, the time domain of computing mode information and frequency domain statistical nature, utilize the latent structure status flag matrix extracted;
(2) structural regime subspace
At first, utilize the Nonlinear Mapping function that eigenmatrix is mapped to high-dimensional feature space, and utilize core locality preserving projections method characteristics of decomposition matrix to obtain projection matrix; Secondly, utilize orthogonalization method to the vectorial orthogonalization in projection matrix, obtain one group of unit orthogonal vector; Finally, utilize these group unit's orthogonal vector to form subspace method;
(3) computing mode subspace index of similarity
At first, construct respectively normal condition subspace and current state subspace; Then, utilize matrix norm to define the index of similarity of two class subspace methods; Described normal condition subspace is:
Figure FDA0000378338510000011
Figure FDA0000378338510000012
Wherein
Figure FDA0000378338510000013
i=1 ..., r 1mean subspace base vector, mean status flag matrix when assembling is good, γ i, i=1 ..., r 1mean weight vector, r 1mean the subspace dimension; Described current state subspace is:
Figure FDA0000378338510000016
Wherein
Figure FDA0000378338510000017
i=1 ..., r 2mean subspace base vector,
Figure FDA0000378338510000018
the status flag matrix that means current time,
Figure FDA0000378338510000019
i=1 ..., r 2mean weight vector, r 2mean the subspace dimension; Then, utilize matrix norm to define the index of similarity SI of two class subspace methods, be expressed as:
SI = | | ( S 1 ) T S 2 | | 2 | | ( S 1 ) T | | 2 | | S 2 | | 2 ;
(4) definition RELIABILITY INDEX
Because the variation range of index of similarity SI is [0,1], and the similarity degree that can reflect current time bolt connection device state and its normal condition, therefore the assembling fiduciary level using this index as RELIABILITY INDEX R quantitative response bolt connection device is defined as RELIABILITY INDEX:
R = SI = | | ( S 1 ) T S 2 | | 2 | | ( S 1 ) T | | 2 | | S 2 | | 2 ;
(5) fiduciary level is judged
Can judge that with RELIABILITY INDEX R the assembling fiduciary level of bolt connection device: R approaches 1 and means that equipment assembly quality and normal condition approach this moment; Otherwise R approaches 0 and means that the equipment assembly quality departs from normal condition this moment.
CN201310403584.2A 2013-09-06 2013-09-06 Method for judging assembling reliability degree of bolt connection device with state information Pending CN103500267A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310403584.2A CN103500267A (en) 2013-09-06 2013-09-06 Method for judging assembling reliability degree of bolt connection device with state information

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310403584.2A CN103500267A (en) 2013-09-06 2013-09-06 Method for judging assembling reliability degree of bolt connection device with state information

Publications (1)

Publication Number Publication Date
CN103500267A true CN103500267A (en) 2014-01-08

Family

ID=49865475

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310403584.2A Pending CN103500267A (en) 2013-09-06 2013-09-06 Method for judging assembling reliability degree of bolt connection device with state information

Country Status (1)

Country Link
CN (1) CN103500267A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103745108A (en) * 2014-01-10 2014-04-23 电子科技大学 Reliability degree assessment method for multilevel state monitoring data fusion
CN109522631A (en) * 2018-11-06 2019-03-26 西安交通大学 It is a kind of to be bolted the mathematical character method being distributed in conjunction with surface pressure
CN111122135A (en) * 2019-12-05 2020-05-08 西安交通大学 Method for evaluating looseness degree of flange bolt connection structure

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5495409A (en) * 1993-01-29 1996-02-27 Matsushita Electric Industrial Co., Ltd. Constructing method of finite-state machine performing transitions according to a partial type of success function and a failure function
CN102607829A (en) * 2012-02-22 2012-07-25 西安交通大学 Quantitative recognition method for assembling loosening fault of dismountable disk drum type rotor

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5495409A (en) * 1993-01-29 1996-02-27 Matsushita Electric Industrial Co., Ltd. Constructing method of finite-state machine performing transitions according to a partial type of success function and a failure function
CN102607829A (en) * 2012-02-22 2012-07-25 西安交通大学 Quantitative recognition method for assembling loosening fault of dismountable disk drum type rotor

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
孙闯,何正嘉,张周锁等: "基于状态信息的航空发动机运行可靠性评估", 《机械工程学报》 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103745108A (en) * 2014-01-10 2014-04-23 电子科技大学 Reliability degree assessment method for multilevel state monitoring data fusion
CN103745108B (en) * 2014-01-10 2016-08-17 电子科技大学 The Reliability assessment method that multi-level Condition Monitoring Data merges
CN109522631A (en) * 2018-11-06 2019-03-26 西安交通大学 It is a kind of to be bolted the mathematical character method being distributed in conjunction with surface pressure
CN109522631B (en) * 2018-11-06 2021-07-13 西安交通大学 Mathematical characterization method for pressure distribution of bolt connection joint surface
CN111122135A (en) * 2019-12-05 2020-05-08 西安交通大学 Method for evaluating looseness degree of flange bolt connection structure

Similar Documents

Publication Publication Date Title
CN108181107B (en) The Wind turbines bearing mechanical method for diagnosing faults of meter and more class objects
CN109357822A (en) A kind of quick test and evaluation method of bridge changed based on Vehicle-Bridge Coupling System time-varying dynamic characteristic
CN104239736A (en) Structure damage diagnosis method based on power spectrum and intelligent algorithms
CN103048137A (en) Fault diagnosis method of rolling bearing under variable working conditions
CN104165925B (en) The centrifugal compressor half-opened impeller crack fault detection method of accidental resonance
Alvanitopoulos et al. Seismic intensity feature construction based on the Hilbert–Huang transform
CN103217213A (en) Modal parameter identification method based on response signal time-frequency joint distribution characteristics
CN103500267A (en) Method for judging assembling reliability degree of bolt connection device with state information
CN102175449A (en) Blade fault diagnostic method based on strain energy response of wind-driven generator
CN103776901B (en) Based on the sticky cartridge clip Rotating fields ageing state recognition methods of vibratory response information
CN106918389A (en) It is a kind of based on the vibration modal analysis method of doppler optical displacement method and its application
Fritzen et al. Vibration-based damage detection under changing environmental and operational conditions
CN104852392A (en) Calculation method of sub-synchronous oscillation mode attenuation coefficients based on Prony algorithm
CN202693304U (en) Gear fault diagnosis platform
Estekanchi et al. Characteristics of second generation endurance time acceleration functions
CN102818686B (en) Grid-control TWT metal grid mesh Modal Experimental Method
CN106446320B (en) Derrick steel structures damnification recognition method based on IMF instantaneous energy curvature difference
Mollineaux et al. Damage detection methods on wind turbine blade testing with wired and wireless accelerometer sensors
James et al. Damping measurements using operational data
Liu et al. Discrepancy study of modal parameters of a scale jacket-type supporting structure of 3.0-MW offshore wind turbine in water and in air
CN105890738A (en) Conflux vortex impact vibration identification method
CN207794158U (en) A kind of pile measurement equipment based on vibration responding
Zheng et al. Classification recognition of anchor rod based on PSO-SVM
CN106841859A (en) The electric energy quality monitoring method for early warning and system of feature based data Optimum Matching
Xue et al. Time‐Varying Wind Load Identification Based on Minimum‐Variance Unbiased Estimation

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20140108

WD01 Invention patent application deemed withdrawn after publication