CN108918141B - Differential self-coding method based on strain type intelligent gear - Google Patents

Differential self-coding method based on strain type intelligent gear Download PDF

Info

Publication number
CN108918141B
CN108918141B CN201810840065.5A CN201810840065A CN108918141B CN 108918141 B CN108918141 B CN 108918141B CN 201810840065 A CN201810840065 A CN 201810840065A CN 108918141 B CN108918141 B CN 108918141B
Authority
CN
China
Prior art keywords
strain
gear
matrix
differential
self
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
CN201810840065.5A
Other languages
Chinese (zh)
Other versions
CN108918141A (en
Inventor
黄文彬
丁晓喜
李泉昌
邵毅敏
揭达斌
钟思平
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.)
Chongqing University
Original Assignee
Chongqing 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 Chongqing University filed Critical Chongqing University
Priority to CN201810840065.5A priority Critical patent/CN108918141B/en
Publication of CN108918141A publication Critical patent/CN108918141A/en
Application granted granted Critical
Publication of CN108918141B publication Critical patent/CN108918141B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • G01M13/00—Testing of machine parts
    • G01M13/02—Gearings; Transmission mechanisms
    • G01M13/021—Gearings
    • 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
    • G01M13/02—Gearings; Transmission mechanisms
    • G01M13/028—Acoustic or vibration analysis

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Retarders (AREA)

Abstract

The invention discloses a differential self-coding method based on a strain type intelligent gear, which specifically comprises the following steps: a, manufacturing a strain type intelligent gear; b, obtaining a local strain response matrix X; c, aligning the strain position matrix D; d, short-time strain matrix C; e, encoding a strain information matrix U; f, strain differential self-coding sequence y; the method overcomes the defects that weak features cannot be extracted for fault diagnosis due to energy dissipation and burying effects in the traditional external sensor acquisition, a large amount of historical information is needed to complete component state analysis and evaluation, and self-evaluation and self-diagnosis cannot be realized; by uniformly distributing the micro strain sensors at the tooth root positions of the gear teeth, the detail information of the local meshing rigidity change and the strain evolution of the gear can be better observed, and a strain-type differential self-coding method is provided based on the strain-type intelligent gear, so that the self-monitoring and self-diagnosis analysis of the health state of the gear can be quickly and reasonably realized.

Description

Differential self-coding method based on strain type intelligent gear
Technical Field
The invention relates to the field of mechanical transmission, in particular to a differential self-coding method based on a strain type intelligent gear.
Background
The gear is an important component of mechanical transmission equipment, and the running state of the gear is directly related to the safety, reliability and maintainability of a mechanical transmission system and the running of the whole machine. With the automation, systematization, complication and higher precision degree of modern mechanical transmission equipment, on one hand, the gearbox body is closed, the internal temperature is high, lubricating oil and oil gas exist at high rotating speed, and the installation space is small; on the other hand, in the actual operation of the box body, various complex working conditions such as speed change, load change, operation change and the like exist, so that a system and a method for monitoring the state of the mechanical transmission equipment under the special mechanical environment are urgently needed to be developed.
At present, for health monitoring and fault diagnosis of gears, scholars at home and abroad have carried out a great deal of research work, and the following defects generally exist on the basis of structural vibration, sound signals or external sounds as research objects:
1. the existing sensor is expensive and has high requirements on installation environment.
2. In the prior art, a sensor is arranged outside a mechanical transmission equipment box body, so that the weakening and attenuation effects of energy dissipation on global vibration signals/sound in the multi-interface contact transmission process cannot be avoided, and the vibration/sound signals have the burying effect on weak fault signals of gears.
3. The prior art can not track the excitation response of the actual meshing position of the gear in real time, and can not realize the accurate positioning diagnosis of the local fault of the gear and the direct pickup of the structural information change.
4. In the prior art, a large amount of historical operation information and external working condition auxiliary information are needed for diagnosing and monitoring gear parts as a basis, however, due to the fact that actual operation working conditions are complex and operation measures are variable, the gear health state monitoring and fault diagnosis analysis are not facilitated, and online rapid self-monitoring and diagnosis analysis cannot be achieved.
It is therefore imperative to those skilled in the art to solve the corresponding technical problems.
Disclosure of Invention
The invention aims to at least solve the technical problems in the prior art, and particularly provides a differential self-encoding method based on a strain type intelligent gear, which is applied to self-monitoring and self-diagnosis analysis of a gear part.
The invention discloses a differential self-coding method based on a strain type intelligent gear, which is characterized in that the design and information self-coding of the strain type intelligent gear comprise the following steps: the intelligent gear for monitoring and diagnosis is characterized in that N tooth roots of the intelligent gear are respectively provided with a flexible strain micro sensor, wherein a fixed tooth number z or an angle is formed between the flexible strain micro sensors, and two gear tooth strain units with approximate 180 degrees are adopted as a group of strain coding pairs and are coded and output; according to the gear meshing mechanism, the rigidity of meshing teeth can be changed when the gears are meshed, and stress change is generated on tooth roots correspondingly; further, it can be found that under the healthy state, each gear can generate similar meshing strain phenomena at different moments, and the meshing response of the local fault gear is greatly different from the meshing response of the healthy gear; based on the facts, the differential output monitoring and diagnosis of local strong strain of the intelligent gear are realized by adopting gear strain coding self-differential output;
the device comprises the following specific steps of sampling, coding, monitoring and diagnosing:
a, manufacturing a strain type intelligent gear: z film type strain micro sensors (Z is the gear tooth number) are uniformly distributed at the root of the gear tooth root, and a strain type intelligent gear is constructed through an embedded sensing system;
b, obtaining a local strain response matrix X: when the gears are actually meshed and operated, adjacent gear teeth are sequentially subjected to tension and compression micro-strain when meshed with another gear and are sensed by the local strain micro-sensor to generate a corresponding local strong strain signal sequence;
c, aligning the strain position matrix D: giving a fixed tooth number z or an angle, realizing alignment of related strain pairs through a phase position matrix according to a phase delay effect of a strain coding pair, and acquiring a corrected local strong strain matrix;
d, short-time strain matrix C: based on a gear meshing principle, extracting local strong strain impact at each tooth root position by setting a local meshing window function W, and constructing and outputting a short-time strain matrix;
e, a strain information encoding matrix U: completing the statistical description of the short-time strain matrix C by utilizing the kurtosis, and acquiring the coded output of strain information;
f, strain differential self-encoding sequence y: and based on the similarity and difference of gear tooth meshing, self-differential encoding output is carried out on the strain information encoding matrix U, accurate description of gear teeth is realized, and the running state and the fault position of the gear are determined.
Further: with first sensor H engaging during rotation of the gear1The position is used as a position starting point, and other sensors are defined as H respectively in the same sequence of the rotating direction2，H3，…，Hi，…，HZ(the actual collected strain data may be in the order of first engagement impact adjustment) where Z is the number of teeth, H1And H1+Δ，H2And H2+Δ，HiAnd Hi+Δ(i-1, 2, …) and so on into a set of strain encoded pairs (using dashed lines)Connected representation) where Δ is a given difference in tooth number
Figure BDA0001745370050000031
Here, the
Figure BDA0001745370050000032
To round-down operations.
Further: by Z miniature strain sensors H1，H2，H3，…，Hi，…，HZThe obtained local strain response matrix X:
Figure BDA0001745370050000033
wherein XiVector is sensor HiThe strain signal sequence (i ═ 1,2, …, Z) is collected, L is the length of the signal vector, Xi＝[xi(1) xi(2) ... xi(L)]。
Further: definition of Xi+ΔVector is and sensor HiPosition sensor H approximated to θ 180 °i+ΔA sequence of acquired strain signals, where Δ is a given difference in tooth number
Figure BDA0001745370050000034
Here, the
Figure BDA0001745370050000035
Is a rounding-down operation; the two are combined into a differential encoding pair matrix Ei：
Figure BDA0001745370050000041
Wherein the selection matrix
Figure BDA0001745370050000042
⊙ denotes a selection operation, based on the principle of phase compensation, using a position matrix alignment to align XiAnd Xi+ΔProducing same-sequence strong strain impact when engaged at different positions to obtain the alignment strain position momentArray Di|2×L：
Figure BDA0001745370050000043
Wherein DiAn alignment matrix, P, for the ith root micro strain celliA strain encoded position matrix corresponding to the ith root micro strain cell,
Figure BDA0001745370050000044
represents the position compensation operation:
Figure BDA0001745370050000045
Figure BDA0001745370050000046
wherein Δ is a given difference in tooth number
Figure BDA0001745370050000047
λ is the number of position compensation points (i.e. timing position compensation)
Figure BDA0001745370050000048
Where f issIs the sampling frequency, frIs the shaft rotation frequency, in the above-mentioned alignment position strain matrix Di|2×LPicking up local strong strain to obtain short-time strain matrix Ci|2×K：
Figure BDA0001745370050000051
Where W (m) is a short time translated rectangular window with a window length of w + 1. According to the gear meshing principle, the following steps are known:
Figure BDA0001745370050000052
wherein α is the window width gain factor, and
Figure BDA0001745370050000053
representing a translation intercept operation, where dzIs the translation step size; where K is 0,1,2, …, K is the number of strain matrix columns, i.e. the number of short-time translation window steps in the strain signal translation:
Figure BDA0001745370050000054
further: the short-time strain matrix can be further obtained by substituting equation (7) into equation (6) according to the gear meshing principle:
Figure BDA0001745370050000055
Figure BDA0001745370050000056
wherein xi,kIs a truncated short-time strain sequence; the strain impulse response of meshing of each tooth can be obtained through the formula, and the local strong strain impact extraction of each tooth root position can be realized through aligning the position strain matrix.
Further: based on a gear meshing mechanism, each gear tooth in a healthy state can generate a similar meshing strain phenomenon at different moments, and a local fault gear meshing response and a healthy gear meshing strain response have a large difference, so that short-time strain information described in a formula (9) is evaluated in a statistical evaluation mode, and therefore the strain information coding of a short-time strain matrix is realized:
Figure BDA0001745370050000057
based on the short-time strain kurtosis code element U in the formula (10)i(k)|2×1And i is 1,2, 1, K, and a short-time kurtosis differential distribution △ Ku is defined, and a strong strain distribution condition reflecting tooth root meshing is obtained through multi-strain information fusion, so that a sampling strain sequence is realizedColumn X-Z×LTo strain differential sequence y-1×KIs output, particularly as shown in fig. 3, 4, where the short-time strain kurtosis matrix u (k) for the k-th step is outputZ×[2×1]Short time differential distribution of (2):
Figure BDA0001745370050000061
finally, sequentially outputting a strain differential self-coding sequence y (k) through a gear meshing principle:
y(k)＝Y(ik,k)s.t.ik＝mod(k,Z)+1 (12)
where mod denotes the modulus. The strain differential self-coding sequence y (k) has good response and detection capability on difference information, the dependence on the gear operation history information is weakened, and self-monitoring and self-diagnosis on each tooth of the gear are realized.
Further: based on the principle of gear mesh impact, a localized failure of the gear roots may produce greater strain and displacement excitation than when the gears are in healthy engagement. Therefore, when the differential level of the strain differential self-coding sequence y (k) exceeds a certain threshold value (the meshing strain statistical distribution level) and has a periodic differential value at the same differential position, the condition that the pair of meshing teeth has a certain fault is indicated, and self-monitoring is realized.
Further: assuming that the corresponding ith gear tooth group has abnormal differential value distribution, namely a fault gear tooth group, judging the position of the fault gear tooth according to the positive and negative values of y (k):
Figure BDA0001745370050000062
where thre is the strain differential sequence statistical distribution level that mainly reflects the strain fluctuation level at healthy smooth engagement. Therefore, accurate positioning and diagnosis and analysis of the gear tooth fault can be realized through the formula (13).
In summary, due to the adoption of the technical scheme, the invention has the beneficial effects that:
1. aiming at the energy dissipation and burying functions existing in external acquisition, the strain gauge is directly installed at the tooth root of the gear to sense the meshing vibration and the running state of the gear, and the method has the advantages of low cost, low power consumption, high efficiency and high precision;
2. aiming at the characteristics of superposition of fault signals and uncertainty of fault positions, the meshing positions of the gears are tracked in real time based on a position strain sensing technology, an encoder and a differential processing technology, so that the fault positions of the gears are accurately positioned;
3. aiming at the defect that the traditional fault diagnosis and monitoring needs healthy gear state data, the local meshing state is directly judged through kurtosis differential indexes between gear tooth pairs, and finally, global self-monitoring and self-diagnosis of the gear equipment are realized.
Drawings
In order to more clearly illustrate the detailed description of the invention or the technical solutions in the prior art, the drawings that are needed in the detailed description of the invention or the prior art will be briefly described below. In the drawings, like elements or portions are generally identified by like reference numerals. In the drawings, elements or portions are not necessarily drawn to scale.
FIG. 1 is a schematic diagram of the differential self-encoding method based on a strain-type intelligent gear of the invention;
FIG. 2 is a detailed flow chart of the differential self-encoding method based on the strain-type intelligent gear of the present invention;
FIG. 3 is a schematic diagram of a strain sampling sequence of the differential self-encoding method based on the strain intelligent gear of the present invention;
FIG. 4 is a schematic diagram of self-differential encoding of the differential self-encoding method based on the strain-type intelligent gear.
Detailed Description
Embodiments of the present invention will be described in detail below with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and therefore are only examples, and the protection scope of the present invention is not limited thereby.
It is to be noted that, unless otherwise specified, technical or scientific terms used herein shall have the ordinary meaning as understood by those skilled in the art to which the invention pertains.
The differential self-encoding method based on the strain type intelligent gear comprises the following steps:
a, manufacturing a strain type intelligent gear: z film type strain micro sensors (Z is the gear tooth number) are uniformly distributed at the root of the gear tooth root, and a strain type intelligent gear is constructed through an embedded sensing system;
b, obtaining a local strain response matrix X: when the gears are actually meshed and operated, adjacent gear teeth are sequentially subjected to tension and compression micro-strain when meshed with another gear and are sensed by the local strain micro-sensor to generate a corresponding local strong strain signal sequence;
c, aligning the strain position matrix D: giving a fixed tooth number z or an angle, realizing alignment of related strain pairs through a phase position matrix according to a phase delay effect of a strain coding pair, and acquiring a corrected local strong strain matrix;
d, short-time strain matrix C: based on a gear meshing principle, extracting local strong strain impact at each tooth root position by setting a local meshing window function W, and constructing and outputting a short-time strain matrix;
e, a strain information encoding matrix U: completing the statistical description of the short-time strain matrix C by utilizing the kurtosis, and acquiring the coded output of strain information;
f, strain differential self-encoding sequence y: and based on the similarity and difference of gear tooth meshing, self-differential encoding output is carried out on the strain information encoding matrix U, accurate description of gear teeth is realized, and the running state and the fault position of the gear are determined.
As shown in FIG. 1, the strain type intelligent gear is constructed in such a way that A is a gear, 1-i + △ is a strain microsensor, and a strain sensor is installed at each tooth root of the gear, so that a first engaged sensor H is engaged when the gear rotates1The position is used as a position starting point, and other sensors are defined as H respectively in the same sequence of the rotating direction2，H3，…，Hi，…，HZ(the actual collected strain data may be in the order of first engagement impact adjustment) where Z is the number of teeth, H1And H1+Δ，H2And H2+Δ，HiAnd Hi+Δ(i-1, 2, …) and so on into a set of strain encoded pairs (indicated by the dashed connection), where Δ is the difference in the given number of teeth
Figure BDA0001745370050000081
Here, the
Figure BDA0001745370050000082
To round-down operations.
As shown in fig. 2, the differential self-encoding method based on the strain-type intelligent gear is as follows: defined by Z micro strain sensors H1，H2，H3，…，Hi，…，HZThe obtained local strain response matrix X:
Figure BDA0001745370050000091
wherein XiVector is sensor HiThe strain signal sequence (i ═ 1,2, …, Z) is collected, L is the length of the signal vector, Xi＝[xi(1) xi(2) ... xi(L)]。
In this example, X is definedi+ΔVector is and sensor HiPosition sensor H approximated to θ 180 °i+ΔA sequence of acquired strain signals, where Δ is a given difference in tooth number
Figure BDA0001745370050000092
Here, the
Figure BDA0001745370050000093
Is a rounding-down operation; the two are combined into a differential encoding pair matrix Ei：
Figure BDA0001745370050000094
Wherein the selection matrix
Figure BDA0001745370050000095
⊙ denotes a selection operation, based on the principle of phase compensation, using a position matrix alignment to align XiAnd Xi+ΔGenerating same-sequence strong strain impact when engaged at different positions to obtain an aligned strain position matrix Di|2×L：
Figure BDA0001745370050000096
Wherein DiAn alignment matrix, P, for the ith root micro strain celliA strain encoded position matrix corresponding to the ith root micro strain cell,
Figure BDA0001745370050000097
represents the position compensation operation:
Figure BDA0001745370050000098
Figure BDA0001745370050000101
wherein Δ is a given difference in tooth number
Figure BDA0001745370050000102
λ is the number of position compensation points (i.e. timing position compensation)
Figure BDA0001745370050000103
Where f issIs the sampling frequency, frIs the shaft rotation frequency, in the above-mentioned alignment position strain matrix Di|2×LPicking up local strong strain to obtain short-time strain matrix Ci|2×K：
Figure BDA0001745370050000104
Where W (m) is a short time translated rectangular window with a window length of w + 1. According to the gear meshing principle, the following steps are known:
Figure BDA0001745370050000105
wherein α is the window width gain factor, and
Figure BDA0001745370050000106
representing a translation intercept operation, where dzIs the translation step size; where K is 0,1,2, …, K is the number of strain matrix columns, i.e. the number of short-time translation window steps in the strain signal translation:
Figure BDA0001745370050000107
in addition, similarly, by substituting equation (7) into equation (6) according to the gear meshing principle, the short-time strain matrix can be further obtained:
Figure BDA0001745370050000108
Figure BDA0001745370050000109
wherein xi,kIs a truncated short-time strain sequence; the strain impulse response of each tooth mesh can be obtained through the above formula, and as shown in fig. 3, by aligning the position strain matrix, local strong strain impulse extraction at each tooth root position can be realized.
Based on a gear meshing mechanism, each gear tooth in a healthy state can generate a similar meshing strain phenomenon at different moments, and a local fault gear meshing response and a healthy gear meshing strain response have a large difference, so that short-time strain information described in a formula (9) is evaluated in a statistical evaluation mode, and therefore the strain information coding of a short-time strain matrix is realized:
Figure BDA0001745370050000111
based on the short-time strain kurtosis code in the formula (10)Yuan Ui(k)|2×1And i is 1,2, is Z, and K is 1,2, is K, a short-time kurtosis differential distribution △ Ku is defined, and strong strain distribution conditions reflecting tooth root meshing are obtained through multi-strain information fusion, so that a sampling strain sequence X is realizedZ×LTo strain differential sequence y-1×KIs output, particularly as shown in fig. 3, 4, where the short-time strain kurtosis matrix u (k) for the k-th step is outputZ×[2×1]Short time differential distribution of (2):
Figure BDA0001745370050000112
finally, sequentially outputting a strain differential self-coding sequence y (k) through a gear meshing principle:
y(k)＝Y(ik,k)s.t.ik＝mod(k,Z)+1 (12)
where mod denotes the modulus. The strain differential self-coding sequence y (k) has good response and detection capability on difference information, the dependence on the gear operation history information is weakened, and self-monitoring and self-diagnosis on each tooth of the gear are realized.
Based on the principle of gear mesh impact, a localized failure of the gear roots may produce greater strain and displacement excitation than when the gears are in healthy engagement. Therefore, when the differential level of the strain differential self-coding sequence y (k) exceeds a certain threshold value (the meshing strain statistical distribution level) and has a periodic differential value at the same differential position, the condition that the pair of meshing teeth has a certain fault is indicated, and self-monitoring is realized. Further, assuming that an abnormal differential value distribution exists in the ith gear tooth group, namely, the fault gear tooth group, judging the position of the fault gear tooth according to the positive and negative values of y (k):
Figure BDA0001745370050000121
where thre is the strain differential sequence statistical distribution level that mainly reflects the strain fluctuation level at healthy smooth engagement. Therefore, accurate positioning and diagnosis and analysis of the gear tooth fault can be realized through the formula (13).
The invention discloses a differential self-coding method based on a strain type intelligent gear, which overcomes the defects that weak characteristics cannot be extracted for fault diagnosis due to energy dissipation and burying effects in the traditional external sensor acquisition, a large amount of historical information is needed to complete component state analysis and evaluation, and self evaluation and self diagnosis cannot be realized; by uniformly distributing the micro strain sensors at the tooth root positions of the gear teeth, the detail information of the local meshing rigidity change and the strain evolution of the gear can be better observed, and a strain-type differential self-coding method is provided based on the strain-type intelligent gear, so that the self-monitoring and self-diagnosis analysis of the health state of the gear can be quickly and reasonably realized.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; such modifications and substitutions do not depart from the spirit and scope of the present invention, and they should be construed as being included in the following claims and description.

Claims (7)

1. A differential self-coding method based on a strain type intelligent gear is characterized in that: the method specifically comprises the following steps:
a, manufacturing a strain type intelligent gear: z film type strain micro sensors are uniformly distributed at the root of a gear root, and a strain type intelligent gear is constructed through an embedded sensing system;
b, obtaining a local strain response matrix X: when the gears are actually meshed and operated, adjacent gear teeth are sequentially subjected to tension and compression micro-strain when meshed with another gear and are sensed by the local strain micro-sensor to generate a corresponding local strong strain signal sequence;
c, aligning the strain position matrix D: giving a fixed tooth number z or an angle, realizing alignment of related strain pairs through a phase position matrix according to a phase delay effect of a strain coding pair, and acquiring a corrected local strong strain matrix;
d, short-time strain matrix C: based on a gear meshing principle, extracting local strong strain impact at each tooth root position by setting a local meshing window function W, and constructing and outputting a short-time strain matrix;
e, a strain information encoding matrix U: completing the statistical description of the short-time strain matrix C by utilizing the kurtosis, and acquiring the coded output of strain information;
f, strain differential self-encoding sequence y: and based on the similarity and difference of gear tooth meshing, self-differential encoding output is carried out on the strain information encoding matrix U, accurate description of gear teeth is realized, and the running state and the fault position of the gear are determined.
2. The differential self-coding method based on the strain type intelligent gear according to claim 1, characterized in that: with first sensor H engaging during rotation of the gear1The position is used as a position starting point, and other sensors are defined as H respectively in the same sequence of the rotating direction2，H3，…，Hi，…，HZWherein Z is the number of teeth, H1And H1+Δ，H2And H2+Δ，HiAnd Hi+ΔThe like to form a group of strain encoding pairs, wherein delta is the difference of given tooth number
Figure FDA0002610278700000011
Here, the
Figure FDA0002610278700000012
To round-down operations.
3. The differential self-coding method based on the strain type intelligent gear according to claim 2, characterized in that: by Z miniature strain sensors H1，H2，H3，…，Hi，…，HZThe obtained local strain response matrix X:
Figure FDA0002610278700000021
wherein XiVector is sensor HiThe sequence of the acquired strain signals, L being the length of the signal vector, Xi＝[xi(1) xi(2)... xi(L)]。
4. The differential self-coding method based on the strain type intelligent gear according to claim 3, characterized in that: definition of Xi+ΔVector is and sensor HiPosition sensor H approximated to θ 180 °i+ΔA sequence of acquired strain signals, where Δ is a given difference in tooth number
Figure FDA0002610278700000022
Here, the
Figure FDA0002610278700000023
Is a rounding-down operation; the two are combined into a differential encoding pair matrix Ei：
Figure FDA0002610278700000024
Figure FDA0002610278700000025
Wherein the selection matrix
Figure FDA0002610278700000026
e denotes a selection operation, based on the principle of phase compensation, using a position matrix alignment to align XiAnd Xi+ΔGenerating same-sequence strong strain impact when engaged at different positions to obtain an aligned strain position matrix Di|2×L：
Figure FDA0002610278700000027
Wherein DiIs the ith tooth root micro strain unitCorresponding alignment position matrix, PiA strain encoded position matrix corresponding to the ith root micro strain cell,
Figure FDA0002610278700000028
represents the position compensation operation:
Figure FDA0002610278700000029
Figure FDA0002610278700000031
wherein Δ is a given difference in tooth number
Figure FDA0002610278700000032
λ is the number of position compensation points
Figure FDA0002610278700000033
Where f issIs the sampling frequency, frIs the shaft rotation frequency, in the above-mentioned alignment strain position matrix Di|2×LPicking up local strong strain to obtain short-time strain matrix Ci|2×K：
Figure FDA0002610278700000034
Wherein W (m) is a short-time translation rectangular window, the window length is w +1, and the following can be obtained according to the gear meshing principle:
Figure FDA0002610278700000035
wherein α is the window width gain factor, and
Figure FDA0002610278700000036
representing a translation intercept operation, where dzIs the translation step size; where K is 0,1,2, …, K is the number of strain matrix columns, and K is alsoNamely the translation steps of the short-time translation window in the strain signal:
Figure FDA0002610278700000037
5. the differential self-coding method based on the strain type intelligent gear is characterized in that: the short-time strain matrix can be further obtained by substituting equation (7) into equation (6) according to the gear meshing principle:
Figure FDA0002610278700000041
wherein xi,kIs a truncated short-time strain sequence; the strain impulse response of meshing of each tooth can be obtained through the formula, and the local strong strain impact extraction of each tooth root position can be realized through aligning the strain position matrix.
6. The differential self-coding method based on the strain type intelligent gear is characterized in that: evaluating the short-time strain information described in the formula (9) by adopting a statistical evaluation mode, thereby realizing the encoding of the strain information of the short-time strain matrix:
Figure FDA0002610278700000042
based on the short-time strain kurtosis code element U in the formula (10)i(k)|2×1And i is 1,2, is Z, and K is 1,2, is K, a short-time kurtosis differential distribution △ Ku is defined, and strong strain distribution conditions reflecting tooth root meshing are obtained through multi-strain information fusion, so that a sampling strain sequence X is realizedZ×LTo strain differential sequence y-1×KThe self-encoded output of (1), wherein the short-time strain kurtosis matrix U (k) for the k-th stepZ×[2×1]Short time differential distribution of (2):
Figure FDA0002610278700000043
finally, sequentially outputting a strain differential self-coding sequence y (k) through a gear meshing principle:
y(k)＝Y(ik,k)s.t.ik＝mod(k,Z)+1 (12)
where mod denotes the modulus.
7. The differential self-coding method based on the strain type intelligent gear is characterized in that: assuming that the corresponding ith gear tooth group has abnormal differential value distribution, namely a fault gear tooth group, judging the position of the fault gear tooth according to the positive and negative values of y (k):
Figure FDA0002610278700000051
wherein, thre is the statistical distribution level of the strain differential sequence, and the accurate positioning and diagnosis analysis of the gear tooth fault can be realized through the formula (13).
CN201810840065.5A 2018-07-27 2018-07-27 Differential self-coding method based on strain type intelligent gear Active CN108918141B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810840065.5A CN108918141B (en) 2018-07-27 2018-07-27 Differential self-coding method based on strain type intelligent gear

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810840065.5A CN108918141B (en) 2018-07-27 2018-07-27 Differential self-coding method based on strain type intelligent gear

Publications (2)

Publication Number Publication Date
CN108918141A CN108918141A (en) 2018-11-30
CN108918141B true CN108918141B (en) 2020-09-29

Family

ID=64418562

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810840065.5A Active CN108918141B (en) 2018-07-27 2018-07-27 Differential self-coding method based on strain type intelligent gear

Country Status (1)

Country Link
CN (1) CN108918141B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114001890A (en) * 2021-10-25 2022-02-01 河北白沙烟草有限责任公司 Method, system, terminal and storage medium for monitoring condition of rolling connection equipment based on vibration and noise data analysis

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS5624545A (en) * 1979-08-08 1981-03-09 Hitachi Ltd Predicting method of trouble in gear transmission
CN104460654B (en) * 2014-11-04 2017-08-25 哈尔滨工业大学 A kind of gear imperfection failure diagnosis Rules extraction method based on quantization characteristic relation
CN104457856B (en) * 2014-12-24 2017-01-11 重庆大学 Gearbox position sequence sampling device and method based on complex information sensors
CN105675113B (en) * 2016-03-16 2018-08-17 重庆大学 Rotating machinery angular domain vibration signal acquisition device based on microsensor and method
CN105841792B (en) * 2016-03-16 2018-12-11 重庆大学 Gear pressure angular direction local vibration signal acquisition methods based on microsensor
CN106441867B (en) * 2016-09-22 2019-01-29 北京航空航天大学 Based on the considerations of the spiral bevel gear Dedenda's bending stress test method of similarity theory dynamic loading

Also Published As

Publication number Publication date
CN108918141A (en) 2018-11-30

Similar Documents

Publication Publication Date Title
CN110866314B (en) Method for predicting residual life of rotating machinery of multilayer bidirectional gate control circulation unit network
CN101835974B (en) Method for determining fatigue damage in a power train of a wind turbine
CN108227676A (en) The online fault detect of valve-controlled cylinder electrohydraulic servo system, estimation and localization method
US20200272139A1 (en) Method and System for Data Driven Machine Diagnostics
US6888262B2 (en) Method and apparatus for wind turbine rotor load control
CN108918141B (en) Differential self-coding method based on strain type intelligent gear
CN103185109A (en) Intelligent drive device
CN105823503A (en) Improved gray prediction GM(1,1) model-based autonomous underwater vehicle (AUV) sensor fault diagnosis method
CN113486868A (en) Motor fault diagnosis method and system
CN109596349A (en) A kind of decelerator trouble diagnostic method based on VMD and PCT
CN110987396B (en) Intelligent fault diagnosis and service life prediction method for coal mining machine rocker arm
KR102156858B1 (en) Method for diagnosing and predicting robot arm's failure
CN103090834A (en) Gear train backlash measuring device and measuring method thereof
CN116662920A (en) Abnormal data identification method, system, equipment and medium for drilling and blasting method construction equipment
CN103410918B (en) Intelligent drive device
CN116793666A (en) Wind turbine generator system gearbox fault diagnosis method based on LSTM-MLP-LSGAN model
JP2000501157A (en) Armature operation standby inspection method
Mehlan et al. Estimation of wind turbine gearbox loads for online fatigue monitoring using inverse methods
Shi et al. A novel diagnostic scheme for gear pitting fault using fiber Bragg grating based strain sensors
CN112270081B (en) Wind driven generator fault detection method based on parallel Elman-NN
CN115171930A (en) Rod position monitoring method, device, equipment and storage medium for high-temperature gas cooled reactor control rod
CN110569478A (en) Improved variational modal decomposition method for encoder signal analysis
CN110686066A (en) Transmission system and gear strain measuring point acquisition method
Wu Vibration Signal Analysis of Complex Mechanical Systems and Early Wear Detection and Forecasting for Gears.
CN116449717B (en) Extruder reduction gearbox state monitoring system based on digital twin

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