CN101000276A - Rotary mechanical failure diagnosis method based on dimensionless index immunity tester - Google Patents
Rotary mechanical failure diagnosis method based on dimensionless index immunity tester Download PDFInfo
- Publication number
- CN101000276A CN101000276A CN 200610132503 CN200610132503A CN101000276A CN 101000276 A CN101000276 A CN 101000276A CN 200610132503 CN200610132503 CN 200610132503 CN 200610132503 A CN200610132503 A CN 200610132503A CN 101000276 A CN101000276 A CN 101000276A
- Authority
- CN
- China
- Prior art keywords
- infin
- index
- dimensionless
- fault
- integral
- 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
Links
Images
Landscapes
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention discloses rotary machine fault diagnosis method based on dimensionless index immunity detector. It is made up of off-line training generating detector and online fault detection. According to immune system vaccination, clone selection mechanism off-line training dimensionless immunity detector, immune evolution, study, memory, and response, detector is classified to apply to realize online fast diagnosis. The invention can be used in rotary machine online state detecting and fault diagnosis to increase precision.
Description
Technical field
The present invention relates to a kind of rotary machinery fault diagnosis method, be used for online fault detect, the identification of rotating machinery, belong to the rotary machinery fault diagnosis technical field based on dimensionless index immunity tester.
Background technology
The vibration processes of rotating machinery is actually a stochastic process.The rotatory mechanical system of operate as normal, the random signal that characterizes its process are steady and weakly stationary; When carrying out signal Processing, temporal analysis (characteristic parameter analytical approach, correlation analysis method, the Domain Synchronous method of average) all is based on the major premise that the random signal that will handle is approximately the weakly stationary random signal.Often there is information nonlinear in a large number, at random, that can not travel through in vibration signal when breaking down, extract for the fault signature time domain and bring great difficulty, and the vibration time-domain signal is the most basic, the most original, and employing Fourier transform etc. often brings bigger personal error.
What can reflect failure message preferably in time-domain analysis is exactly probability density function, based on probability density function, in the amplitude domain that derives dimension index (as average, root-mean-square value, kurtosis etc.) and dimensionless index arranged, the dimension index is arranged generally to the fault signature sensitivity, but very easily be subjected to disturbing effect, dimensionless index is insensitive to disturbing, but seeks the corresponding relation of fault signature sensitivity is still had difficulties.
The research of artificial immune system is risen, and injects new vitality to temporal analysis.Particularly negative selection algorithm is in Study on application of fault diagnosis, for fault diagnosis technology provides strong means at aspects such as fault signature extraction, online quick diagnosis.But, at present, the work that domestic and international experts and scholars are carried out is the basis of research, analysis with the dimension index that has in the time domain vibration signal often, still exist the time domain vibration signal of being analyzed often to change because of the variation of external conditions such as load, rotating speed, be difficult in practice distinguish normal with malfunction, be difficult to extract effectively and have fault signature one to one, make and really effectively use seldom.
Summary of the invention
The purpose of this invention is to provide a kind of rotary machinery fault diagnosis method based on dimensionless index immunity tester, this detecting device is the online detection abnormal failure of energy on the one hand, on the other hand constantly to known fault off-line learning, training, memory, when reaching online quick diagnosis, the fault detect accuracy increases substantially, thereby is providing strong means aspect the fault signature extraction for fault diagnosis technology.
Purpose of the present invention can realize by following technical measures:
A kind of rotary machinery fault diagnosis method based on dimensionless index immunity tester, based on probability density function, derive several dimensionless indexs in the amplitude domain, make up the fault diagnosis that the dimensionless immune detectors is rotated machinery in conjunction with the negative principle mechanisms of selecting of artificial immune system, this method comprises the steps:
(1), fully obtain the span of each dimensionless index when the rotating machinery operate as normal, be defined as the own space of certain dimensionless index;
(2), according to sizes such as rotating machinery capacity, rotating speeds, the size of each dimensionless index when estimating than catastrophic failure generates the rough defective space of each dimensionless index;
(3), from the own space of dimensionless index each dimensionless index immunity tester of off-line training, constitute each dimensionless index immunity tester collection.
(4), produce a certain fault that rotating machinery provides, each detecting device of being concentrated by each dimensionless index immunity tester carries out incomplete matching detection successively; If detect unusually, write down this fault signature string and detecting device string, and this detecting device is put into outstanding detector library, produce next fault again, carry out same coupling, training process; Until at last, generate the outstanding detecting device of fault, these outstanding detecting devices have constituted the non-space that can shine upon one by one with exclusive fault signature.
2, a kind of rotary machinery fault diagnosis method according to claim 1 based on dimensionless index immunity tester, it is characterized in that described dimensionless index immunity tester, comprise peak value index immune detectors, nargin index immune detectors, pulse index immune detectors, kurtosis index immune detectors, waveform index immune detectors;
The dimensionless index that constitutes dimensionless index immunity tester is defined as follows:
In the formula, x represents vibration amplitude, the probability density function of p (x) expression vibration amplitude.On the basis of formula (1):
If l=2, m=1 has the waveform index:
If l → ∞, m=1 has the pulse index:
If l → ∞,
The nargin index is arranged:
If l → ∞, m=2 has the peak value index:
In addition, also have the kurtosis index:
In the formula (6),
Be kurtosis.
The present invention is better reflecting the probability density function of failure message in time-domain analysis, and the dimensionless index in the amplitude domain that derives is selected principle mechanisms to make up to form in conjunction with artificial immune system is negative.This method mainly generates detecting device by off-line training and the online detecting device that utilizes carries out the realization of fault detect two parts.According to immune system vaccine inoculation, Immune Clone Selection mechanism off-line training dimensionless immune detectors, notion according to immunoevolution, immunological learning, immunological memory and immune response etc., the detecting device classification is detected application, reached the purpose of online quick diagnosis.The presence that the present invention can be used for rotating machinery detects and fault diagnosis, can improve greatly and detect and the fault diagnosis precision.
Description of drawings
Fig. 1 unit operate as normal displacement time domain waveform figure;
Fig. 2 base flexible fault displacement time domain waveform figure;
Fig. 3 splits a fault displacement time domain waveform figure;
Fig. 4 cambered axle fault displacement time domain waveform figure;
Three kinds of fault waveform desired values of Fig. 5 synoptic diagram;
Fig. 6 fault of eccentricity displacement time domain waveform figure;
Fig. 7 misaligns fault displacement time domain waveform figure;
Fig. 8-1 peak value desired value synoptic diagram;
Fig. 8-2 nargin desired value synoptic diagram;
Fig. 8-3 pulse desired value synoptic diagram;
Fig. 8-4 waveform desired value synoptic diagram;
Fig. 8-5 kurtosis desired value synoptic diagram.
Embodiment
The platform of this experiment is the motor that is provided by Maoming College---compressor simulation unit.
The division of own space and defective space and coding
At first carry out balance and demarcate on testing table, making the axle of whole test platform, particularly unit is the state that is in centering, balance, and the rotating speed of unit at first is set in 1860r/min.Consider the collection condition of priori, this example is with the poorest waveform index object as an example of diagnosis capability.Fully collect the test unit in normal operation, the data of the waveform index of corresponding displacement time domain waveform.
(1) determines that span is 1.117~1.121 under the normal condition;
(2) pass through with reference to pertinent literature, preestablishing fault coverage is 1.25~1.40;
(3) by 1.117~2.25 overall spans, MIN=1.117, MAX=1.40 adopts 6 binary codings, and each length of an interval degree is
As follows between the code area:
1.117~1.1214~1.1258~1.1302~1.1346~1.139~1.1434~1.1478~1.1522~
1.1566~1.161~1.1654~1.1698~1.1742~1.1786~1.183~1.1874~1.1918~
1.1962~1.2006~1.205~1.2094~1.2138~1.2182~1.2226~1.227~1.2314~
1.2358~1.2402~1.2446~1.249~1.2534~1.2578~1.2622~1.2666~1.271~
1.2754~1.2798~1.2842~1.2886~1.293~1.2974~1.3018~1.3062~1.3106~
1.315~1.3194~1.3238~1.3282~1.3326~1.337~1.3414~1.3458~1.3502~
1.3546~1.3590~1.3634~1.3678~1.3722~1.3766~1.381~1.3854~1.3898~
1.3942~1.40
More than the corresponding binary coding in each section interval be in order: 000000~000001~...~111110~111111, the detecting device of considering own space needs certain code length, refetching the normal condition span is 1.117~1.1258, and promptly own space string encoding is 000000~000001; Other coded strings is 000010~000011~...~111110~111111, wherein, syndrome serial is encoded to 000111~0001000~...~111110~111111; Here need to prove at own space string and the some strings between the defective space string and can do flexible processing, near MIN value, go here and there the be thought of as syndrome serial of close fault value lower limit if need can be used as oneself.
2. carry out matching operation with own space, generate initial dimensionless index immunity tester collection R
Own space gone here and there 000,000 000001 carry out feminine gender and select computing, produce and 000,000 000001 unmatched initial detector, matching domain r=8 adopts variation search procedure method to search for, and formation can cover the initial detector collection in the whole space of controlling oneself substantially.
3. carry out yojan, cluster matching computing, generate ripe dimensionless index immunity tester
Press matching domain r=8, each detecting device and whenever going here and there in twos of defective space string that detecting device is concentrated are mated training, promptly be to carry out yojan, cluster computing, remove the detecting device that is complementary with any two or more syndrome serial in twos, generate ripe detecting device collection.
4. train, generate outstanding dimensionless index immunity tester in real time
The ripe detecting device that trains is detected in real time, in the test, carry out base flexible successively respectively, split axle, the test of three kinds of simulated failures of cambered axle.After the simulated failure test each time, between testing, need again shaft system of unit to be carried out centering, balance demarcation next time.Unit operate as normal, base flexible fault, split a fault, cambered axle fault displacement time domain waveform figure sees shown in the accompanying drawing 1,2,3,4.The waveform desired value is referring to table 1 separately.In the test, be one group by 1024 and sample that each index is respectively got 10 groups, by the minimum value in 10 groups of each index and maximal value span as this index.
Table 1 waveform desired value scope
As can be seen from Table 1, three kinds of malfunctions itself and corresponding above-mentioned coding span all exist between the code area of repetition in the table, that is have the different faults that has common trait, can know from Fig. 5 and find out this relation.
For obtaining there is the outstanding detecting device of mapping relations one by one, must carry out yojan, cluster computing to original ripe detecting device collection R with exclusive fault signature.Its process directly adopts coding interval division coding section to explain for convenience of description.
The loose fault in basis: 1.291~1.344; Corresponding binary code is: 100111,101000,101001,101010,101011,101100,101101,101110,101111,110000,110001,110010,110011 totally ten three section 6 bit code.
Split a fault: 1.283~1.363; Corresponding binary code is: 100101,100110,100111,101000,101001,101010,101011,101100,101101,101110,101111,110000,110001,110010,110011,110100,110101,110110,110111 totally ten nine section 6 bit code.
Cambered axle fault: 1.311~1.339; Corresponding binary code is: 101,100 101,101 101,110 101,111 110,000,110,001 110010 totally seven section 6 bit code.
In corresponding the 1st step between the code area as seen, pass through yojan, cluster, it is as follows to obtain the exclusive feature of each fault.
Split the exclusive feature of a fault: 1.2798~1.2842~1.2886 (corresponding coding 100,101 100110) and 1.3414~1.3458~1.3502~1.3546~1.3590~1.3634 (correspondence is encoded to 110,100 110,101,110,110 110111).The exclusive feature of the loose fault in basis: 1.2886~1.293~1.2974~1.3018~1.3062~1.3106 (correspondence is encoded to 100,111 101,000 101,001 101,010 101011) and 1.3414~1.3458 (corresponding coding 110011).
The exclusive feature of cambered axle fault: 1.3106~1.315~1.3194~1.3238~1.3282~1.3326~1.337~1.3414 (correspondence is encoded to 101100,101101,101110,101111,110000,110001,110010).
Whenever make up the dimensionless index immunity tester that exclusive feature string mates in twos with above-mentioned each fault, be outstanding waveform index immune detectors, total set is outstanding dimensionless index immunity tester collection between them.Ripe conversely speaking, detecting device has formed dimensionless index immunity tester collection that can be outstanding one to one with exclusive fault signature through carrying out yojan, cluster with each fault signature, uses when waiting until actual detected.
5. the off-line training of five kinds of dimensionless index immunity testers
With reference to above method and step, carry out eccentric test, misaligned test, the figure of displacement time domain waveform separately is referring to accompanying drawing 6,7, and finish the off-line training of other immune detectors such as nargin index immune detectors, pulse index immune detectors, kurtosis index immune detectors and peak value index immune detectors, set up the immune detectors of maturation separately, outstanding immune detectors.Calculate five kinds of dimensionless index values according to normal condition and five kinds of malfunction time domain waveforms, referring to accompanying drawing 8-1 to 8-5.As seen from the figure, each corresponding five kinds of fault of five kinds of dimensionless index values all have the situation that desired value intersects (every section odd even point line is represented normal, basic loose fault successively, split a fault, fault of eccentricity, cambered axle fault, misalign the fault indices value).Operating process referring to the 4th step can train five kinds of outstanding dimensionless index immunity tester collection.Each self-corresponding concrete outstanding detecting device coding is listed no longer in detail.
6. unit fault detect actual application
(1) when practical application, at first utilize outstanding dimensionless index immunity tester to real time data binary coding string matching detection, if coupling then knows fault has taken place, and know which kind of fault has taken place.This point has embodied the fast response characteristic of artificial immune system immune response.
(2) if real time data and all outstanding dimensionless index immunity tester do not match; then adopt ripe dimensionless index immunity tester that it is carried out matching detection; if coupling; then fault has taken place in prompting; and prompting is new fault; answer shutdown inspection; determine failure mode; understand the dimensionless index value of this fault correspondence; and set up the knowledge base of this fault, consider with existing dimensionless index immunity tester collection R ' unification again, if having points of resemblance; carry out yojan, cluster again, form new dimensionless index immunity tester collection.This process has embodied the study mechanism of artificial immune system, has also embodied the continuous perfecting process of fault diagnosis knowledge base.
(3) if real time data does not match with all dimensionless index immunity tester, then belong to normal condition, continue detection.
Claims (2)
1, a kind of rotary machinery fault diagnosis method based on dimensionless index immunity tester, based on probability density function, derive several dimensionless indexs in the amplitude domain, make up the fault diagnosis that the dimensionless immune detectors is rotated machinery in conjunction with the negative principle mechanisms of selecting of artificial immune system, this method comprises the steps:
(1), fully obtain the span of each dimensionless index when the rotating machinery operate as normal, be defined as the own space of certain dimensionless index;
(2), according to sizes such as rotating machinery capacity, rotating speeds, the size of each dimensionless index when estimating than catastrophic failure generates the rough defective space of each dimensionless index;
(3), from the own space of dimensionless index each dimensionless index immunity tester of off-line training, constitute each dimensionless index immunity tester collection.
(4), produce a certain fault that rotating machinery provides, each detecting device of being concentrated by each dimensionless index immunity tester carries out incomplete matching detection successively; If detect unusually, write down this fault signature string and detecting device string, and this detecting device is put into outstanding detector library, produce next fault again, carry out same coupling, training process; Until at last, generate the outstanding detecting device of fault, these outstanding detecting devices have constituted the non-own space that can shine upon one by one with exclusive fault signature.
2, a kind of rotary machinery fault diagnosis method based on dimensionless index immunity tester according to claim 1 is characterized in that described dimensionless index immunity tester comprises peak value index immune detectors, nargin index immune detectors, pulse index immune detectors, kurtosis index immune detectors, waveform index immune detectors;
The dimensionless index that constitutes dimensionless index immunity tester is defined as follows:
In the formula, x represents vibration amplitude, the probability density function of P (x) expression vibration amplitude.On the basis of formula (1):
If l=2, m=1 has the waveform index:
If l → ∞, m=1 has the pulse index:
If l → ∞,
The nargin index is arranged:
If l → ∞, m=2 has the peak value index:
In addition, also have the kurtosis index:
In the formula (6),
Be kurtosis.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 200610132503 CN101000276A (en) | 2006-12-29 | 2006-12-29 | Rotary mechanical failure diagnosis method based on dimensionless index immunity tester |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 200610132503 CN101000276A (en) | 2006-12-29 | 2006-12-29 | Rotary mechanical failure diagnosis method based on dimensionless index immunity tester |
Publications (1)
Publication Number | Publication Date |
---|---|
CN101000276A true CN101000276A (en) | 2007-07-18 |
Family
ID=38692334
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN 200610132503 Pending CN101000276A (en) | 2006-12-29 | 2006-12-29 | Rotary mechanical failure diagnosis method based on dimensionless index immunity tester |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101000276A (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103234768A (en) * | 2013-04-24 | 2013-08-07 | 广东石油化工学院 | Combinational fan composite failure diagnosis test method |
CN103678881A (en) * | 2013-11-22 | 2014-03-26 | 张清华 | Composite fault diagnosis method based on combination of artificial immunity and evidence theory |
CN103810374A (en) * | 2013-12-09 | 2014-05-21 | 中国矿业大学 | Machine fault prediction method based on MFCC feature extraction |
CN104614166A (en) * | 2015-01-29 | 2015-05-13 | 西北工业大学 | Method for identifying failure state of rotor vibration signal of aircraft engine |
CN105067262A (en) * | 2015-08-04 | 2015-11-18 | 潍坊学院 | Rolling bearing state monitoring method |
CN106017921A (en) * | 2016-07-13 | 2016-10-12 | 广东石油化工学院 | Rotating machine fault diagnosis method based on peak index |
CN106053067A (en) * | 2016-05-24 | 2016-10-26 | 广东石油化工学院 | Bearing fault diagnosis method based on quantum genetic algorithm optimized support vector machine |
CN106248356A (en) * | 2016-07-13 | 2016-12-21 | 广东工业大学 | A kind of rotary machinery fault diagnosis method based on kurtosis index |
CN110017989A (en) * | 2019-05-17 | 2019-07-16 | 山东科技大学 | A kind of method of wind energy conversion system bearing failure diagnosis |
CN110631810A (en) * | 2018-05-31 | 2019-12-31 | 新疆金风科技股份有限公司 | Method and equipment for detecting abnormal vibration of wind generating set and wind generating set |
-
2006
- 2006-12-29 CN CN 200610132503 patent/CN101000276A/en active Pending
Cited By (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103234768A (en) * | 2013-04-24 | 2013-08-07 | 广东石油化工学院 | Combinational fan composite failure diagnosis test method |
CN103678881A (en) * | 2013-11-22 | 2014-03-26 | 张清华 | Composite fault diagnosis method based on combination of artificial immunity and evidence theory |
CN103678881B (en) * | 2013-11-22 | 2017-05-17 | 张清华 | Composite fault diagnosis method based on combination of artificial immunity and evidence theory |
CN103810374B (en) * | 2013-12-09 | 2017-04-05 | 中国矿业大学 | A kind of mechanical disorder Forecasting Methodology based on MFCC feature extractions |
CN103810374A (en) * | 2013-12-09 | 2014-05-21 | 中国矿业大学 | Machine fault prediction method based on MFCC feature extraction |
CN104614166A (en) * | 2015-01-29 | 2015-05-13 | 西北工业大学 | Method for identifying failure state of rotor vibration signal of aircraft engine |
CN104614166B (en) * | 2015-01-29 | 2017-04-19 | 西北工业大学 | Method for identifying failure state of rotor vibration signal of aircraft engine |
CN105067262A (en) * | 2015-08-04 | 2015-11-18 | 潍坊学院 | Rolling bearing state monitoring method |
CN105067262B (en) * | 2015-08-04 | 2017-10-27 | 潍坊学院 | A kind of state monitoring of rolling bearing method |
CN106053067A (en) * | 2016-05-24 | 2016-10-26 | 广东石油化工学院 | Bearing fault diagnosis method based on quantum genetic algorithm optimized support vector machine |
CN106248356A (en) * | 2016-07-13 | 2016-12-21 | 广东工业大学 | A kind of rotary machinery fault diagnosis method based on kurtosis index |
CN106017921A (en) * | 2016-07-13 | 2016-10-12 | 广东石油化工学院 | Rotating machine fault diagnosis method based on peak index |
CN106017921B (en) * | 2016-07-13 | 2018-07-06 | 广东石油化工学院 | A kind of rotary machinery fault diagnosis method based on peak index |
CN106248356B (en) * | 2016-07-13 | 2019-03-26 | 广东工业大学 | A kind of rotary machinery fault diagnosis method based on kurtosis index |
CN110631810A (en) * | 2018-05-31 | 2019-12-31 | 新疆金风科技股份有限公司 | Method and equipment for detecting abnormal vibration of wind generating set and wind generating set |
CN110631810B (en) * | 2018-05-31 | 2021-07-27 | 新疆金风科技股份有限公司 | Method and equipment for detecting abnormal vibration of wind generating set and wind generating set |
CN110017989A (en) * | 2019-05-17 | 2019-07-16 | 山东科技大学 | A kind of method of wind energy conversion system bearing failure diagnosis |
CN110017989B (en) * | 2019-05-17 | 2021-06-08 | 山东科技大学 | Method for diagnosing bearing fault of wind turbine |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101000276A (en) | Rotary mechanical failure diagnosis method based on dimensionless index immunity tester | |
CN106408088B (en) | A kind of rotating machinery method for diagnosing faults based on deep learning theory | |
Zhao et al. | Deep convolutional neural network based planet bearing fault classification | |
CN104849050A (en) | Rolling bearing fault diagnosis method based on composite multi-scale permutation entropies | |
CN101907088B (en) | Fault diagnosis method based on one-class support vector machines | |
CN103776654B (en) | The method for diagnosing faults of multi-sensor information fusion | |
CN110370649B (en) | Online monitoring device and system of 3D printing equipment | |
Wang et al. | Intelligent rolling bearing fault diagnosis via vision ConvNet | |
Li et al. | Refined composite multivariate multiscale dispersion entropy and its application to fault diagnosis of rolling bearing | |
CN105956526A (en) | Method for identifying a microearthquake event with low signal-to-noise ratio based on multi-scale permutation entropy | |
CN112033463B (en) | Nuclear power equipment state evaluation and prediction integrated method and system | |
CN108490345B (en) | Fault simulation diagnosis method for tap switch of converter transformer | |
CN109583386A (en) | A kind of intelligent rotating mechanical breakdown depth network characterization discrimination method | |
CN105488520A (en) | Multi-resolution singular-spectrum entropy and SVM based leakage acoustic emission signal identification method | |
CN103868692A (en) | Rotary machine fault diagnosis method based on kernel density estimation and K-L divergence | |
CN113865868B (en) | Rolling bearing fault diagnosis method based on time-frequency domain expression | |
CN102156873B (en) | Chaos-based method for detecting and classifying early single-point faults of mechanical component | |
CN102944416A (en) | Multi-sensor signal fusion technology-based fault diagnosis method for wind turbine blades | |
CN101382439A (en) | Multi-parameter self-confirming sensor and state self-confirming method thereof | |
CN103398769A (en) | Transformer on-line fault detecting method based on sampling integrated SVM (support vector machine) under wavelet GGD (general Gaussian distribution) feather and unbalanced K-mean value | |
CN105318905B (en) | A kind of Thermal System of Fossil Fuel-Fired Power Station sensor fault diagnosis method | |
CN105044499A (en) | Method for detecting transformer state of electric power system equipment | |
CN110263767A (en) | In conjunction with the intelligent Rotating Shaft Fault method of compressed data acquisition and deep learning | |
CN105388884A (en) | Alarm system for detecting leakage fault of heat supply network based on identification algorithm driven by data and method | |
CN105094118A (en) | Airplane engine air compressor stall detection method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |