CN105184364B - A kind of pulling force compensation method for non-destructive tests - Google Patents
A kind of pulling force compensation method for non-destructive tests Download PDFInfo
- Publication number
- CN105184364B CN105184364B CN201510627442.3A CN201510627442A CN105184364B CN 105184364 B CN105184364 B CN 105184364B CN 201510627442 A CN201510627442 A CN 201510627442A CN 105184364 B CN105184364 B CN 105184364B
- Authority
- CN
- China
- Prior art keywords
- pulling force
- admittance
- admittance value
- data
- rbf neural
- 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
Links
Landscapes
- Investigating Strength Of Materials By Application Of Mechanical Stress (AREA)
Abstract
The invention discloses a kind of pulling force compensation method for non-destructive tests, comprise the following steps：(1) swept frequency and different stresses are formed into RBF neural network structure as input layer, admittance information as output layer；(2) use and the RBF neural is trained comprising frequency, pulling force and the sample data of admittance；(3) using the frequency of the test data after the completion of training and pulling force as input, contrasted by the corresponding admittance data of simulation data, and to the admittance data of the admittance data and emulation gained of actual measurement, the effect of compensation is weighed according to RMSD damage criterions；Compensate influence of the external pull to admittance data；This method is damaged tension girder steel and is verified by a lossless tension girder steel experiment and one；By method provided by the invention, the influence that pulling force suffered by structure identifies to EMI regulations for technical operations is eliminated, there is good practicality.
Description
Technical field
The invention belongs to structural health monitoring technology field, is mended more particularly, to a kind of pulling force for non-destructive tests
Compensation method.
Background technology
For civil engineering structure due to being used continuously, material degeneration, environment influence, earthquake etc. cause structural damage, and big absolutely
The potential microlesion of part-structure is difficult to be identified by naked eyes.If the microlesion of structure early stage is not found in time, can trigger
The progressive damage of structure, so as to cause the unexpected loss of structure, timely and accurately identify structure early stage faint and potential damage
Wound, it is the problem of structural engineering field；Existing structural health monitoring technology has visual inspection method, low-frequency vibration technology, static state
Structural response technology, and local non-destructive testing technology；The defects of these technologies, essentially consists in：Engineering structure system is huge, inspection
Survey is difficult to reach each position, takes time and effort；Because its low frequency characteristic is invalid to local damage, and easily by Environmental Noise Influence；
The static characteristic (such as displacement, speed, acceleration) of measurement bulky structure does not have operability.
Piezoelectric Impedance (EMI) technology based on PZT is based primarily upon local high-frequency excitation, and sensor is used as simultaneously by the use of PZT
And driver, the information for obtaining structural behaviour change is encouraged to structure partial, so as to realize the identification to structure microlesion；Its
General principle be using high-strength binding agent by inside PZT sticking structures surface or implant infrastructure, using the positive inverse piezoelectric effects of PZT,
Voltage is applied to structure partial exciting by Piezoelectric Impedance instrument, obtains the prison related with structural behaviour (quality, rigidity, damping etc.)
Signal is surveyed, the benchmark that this signal is weighed as structural health, identifies whether structure occurs by the change of observation signal in the future
Damage.Due to its high frequency characteristics, have it is sensitive to structure microlesion, and the advantages of Environmental Noise Influence can be avoided；But the skill
Art is still limited to testpieces, does not there is the influence for being included in external pull to workpiece.
And under actual condition, engineering structure is in stress all the time；And during the whole military service of engineering structure, institute
The load born is continually changing；When structure is damaged, local stress can great changes will take place；EMI technologies pass through measurement
Admittance signal and by differentiating that Damage Assessment Method is realized in the skew of mechanical admittance curves；Experiment shows that structure is in the state by pulling force
Under Piezoelectric Impedance admittance signal and mechanical admittance curves can not caused to offset by the difference of state of tension, structure tension, so as to influence
The degree of accuracy of identification of the EMI technologies to structural damage；Therefore, the interference that tension variations identify to engineering structure damage is excluded, is built
Vertical effective pulling force compensation method has to be solved.
The content of the invention
For the disadvantages described above or Improvement requirement of prior art, the invention provides a kind of pulling force benefit for non-destructive tests
Compensation method, its object is to be trained by using comprising frequency, pulling force and the sample data of admittance to RBF neural,
The compensation to pulling force suffered by structure is realized, is the accurate non-destructive tests exclusive PCR of structure.
To achieve the above object, according to one aspect of the present invention, there is provided a kind of pulling force for non-destructive tests compensates
Method, the pulling force compensation method are based on RBF (RBF of multivariate interpolation, Radical Basis Function) nerves
Network specifically comprises the following steps：
(1) using swept frequency and pulling force as input layer, using admittance value as output layer, RBF is formed together with hidden layer
Neutral net；
(2) RBF neural is trained using the sample data comprising swept frequency, pulling force and admittance value,
Until the difference between the admittance value of output admittance value and respective sample data is within ± the 5% of sample data admittance value, knot
Beam RBF neural is trained；
(3) using the swept frequency of test data and pulling force as input, the RBF nerve nets obtained are trained using step (2)
Network, obtain admittance value corresponding with the swept frequency and pulling force；The admittance value contains external pull information, compensate for drawing
Influence of the power to admittance data；
(4) admittance value obtained according to step (3), the RMSD damage criterions after compensation are obtained；
The above-mentioned pulling force compensation method for non-destructive tests is first using the sample number for including swept frequency, pulling force and admittance value
It is trained according to RBF neural, using the swept frequency and pulling force of test data as RBF neural after the completion of training
Input, obtains the admittance value accordingly exported, and the admittance value contains pulling force information, compensate for influence of the pulling force to admittance data.
Preferably, in the RBF neural training described in step (2), hidden layer neuron is determined according to following methods
Quantity：
A, using the input vector corresponding to mean square error caused by RBF neural (MSE) as weight vector, generation one
Individual new hidden layer neuron；
B, it is new with step a generations using the admittance value of emulation gained as output using test gained admittance value as input
Hidden layer neuron, form new RBF neural；
C, the mean square error of RBF neural new described in obtaining step b；
D, repeat step a~c, it is implicit with now until the mean square error of RBF neural reaches default mean square error
Quantity of the layer neuronal quantity as RBF neural hidden layer neuron；Wherein, mean square error is preset 5%~10%；
In RBF neural training, prior art is the quantity for making hidden layer neuron with the element phase of input vector
Deng；But when input vector is a lot, excessive hidden layer unit number can increase the complexity of calculating；On provided by the invention
Method is stated, progressively the optimum value for reaching hidden layer neuron quantity of convergence, optimal benefit can be obtained in the amount of calculation of minimum
Repay effect.
Preferably, described in step (2) the step of being trained to RBF neural, including following sub-step：
(2.1) admittance value of swept frequency, pulling force and desired output is given；
(2.2) using above-mentioned swept frequency and pulling force as input vector, the RBF neural obtained through step (1) is handled,
Obtain admittance value；
(2.3) admittance value of the admittance value and the desired output that are obtained in step (2.2) is made the difference, obtains difference；Sentence
Whether the difference of breaking is within ± the 5% of desired output admittance value；If so, then terminate to train；If it is not, then increase neuron,
And return to step (2.2).
Preferably, after above-mentioned steps (2.3) terminate to the training of RBF neural, RBF god is determined using following steps
Through network training effect, it is specially：
(2.4) data of training sample be will differ from as input vector, imitated with the RBF neural trained
Very, the admittance value of simulation data is obtained；
(2.5) RMSD values are obtained according to the admittance value of step (2.4) inner simulation data, if obtained under each pulling force operating mode
RMSD values fluctuate within 5%, then into step (3)；Otherwise, return to step (2.1), using new sample data re -training
RBF neural；
Wherein, RMSD (root mean-square deviation) is the signal measured under different conditions structure
Root mean square index, PZT (piezoelectric ceramics, is pressed before and after can weighing admittance signal change according to the statistical indicator
Electroceramics) sensor electric signal intensity of variation, obtained according to following formula：
Wherein, n refers to sample frequency number, and i refers to the admittance value of each stepped-frequency signal, xiMeasured before representing structural damage
Impedance value；yiRepresent the impedance value measured after structural damage；Wherein, i=1,2,3 ..., n.
In general, by the contemplated above technical scheme of the present invention compared with prior art, it can obtain down and show
Beneficial effect：
(1) provided by the present invention for the pulling force compensation method of non-destructive tests, using influence of the RBF neural to pulling force
Compensate；Using swept frequency and different stresses as input layer, admittance information is as output layer；Neutral net is defeated
Enter and specific mapping relations are formed between layer and output layer；By training, emulated using the neutral net formed；
After giving some input, corresponding output is gone out by Simulation of Neural Network；So as to reach the influence for eliminating pulling force to admittance value,
Reduce interference of the pulling force to Piezoelectric Impedance Technology.
(2) pulling force compensation method provided by the invention, in its preferred scheme, in addition to whether analysis network training effect closes
The step of reason, have can constantly regulate emulation data flexibility, the compensation effect of actual condition is met until reaching.
Brief description of the drawings
Fig. 1 is RBF neural schematic diagram；
Fig. 2 is RBF neural training schematic flow sheet；
Fig. 3 is mechanical admittance curves figure and partial enlarged drawing under each pulling force operating modes of PZT2 for measuring test specimen one；
Fig. 4 is mechanical admittance curves figure and partial enlarged drawing under each pulling force operating modes of PZT3 for measuring test specimen two；
Fig. 5 is PZT2RBF emulation and the actual measurement admittance information comparison diagram for measuring test specimen one；
Fig. 6 is PZT3RBF emulation and the actual measurement admittance information comparison diagram for measuring test specimen two；
Fig. 7 is that the PZT2 admittance of test specimen one compensates front and rear RMSD indexs comparison diagram through RBF；
Fig. 8 is that PZT3 admittance compensates front and rear RMSD indexs comparison diagram through RBF when test specimen two is lossless；
Fig. 9 is that PZT3 admittance compensates front and rear RMSD indexs comparison diagram through RBF when test specimen two has damage.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, it is right below in conjunction with drawings and Examples
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.As long as in addition, technical characteristic involved in each embodiment of invention described below
Conflict can is not formed each other to be mutually combined.
Provided by the present invention for the pulling force compensation method of non-destructive tests, damage inspection when being influenceed for structure by pulling force
Survey, based on to type RBF neural, its first layer be input layer before a kind of three layers, be made up of signal source node, the second layer is hidden
Containing layer, third layer is output layer；The neuronal quantity of hidden layer according to description problem it needs to be determined that, the transforming function transformation function of hidden layer
It is the non-negative nonlinear function of central point radial symmetric and decay；The conversion of input layer to hidden layer control is nonlinear, and
Change from hidden layer control to output layer is linear；
Hidden layer control is formed by the use of RBF as " base " of hidden layer, output vector (need not directly be connected) by weighing
It is mapped to latent space；And implicit sheaf space is linear to the mapping relations in output space, the output of network is hidden unit output
Linear weighted function and；Power herein be network tunable parameter in general, network is non-thread by the mapping for being input to output
Property, and network output is linear to adjustable parameter；The power of network obtains according to linear equation, so as to provide RBF significantly
The pace of learning of neutral net；The RBF of RBF neural uses Gaussian function in embodiment；Embodiment 1 its use
RBF neural network structure as schematically shown in Figure 1, using swept frequency and pulling force as input layer, output admittance data, it is implicit
The neuronal quantity of layer determines according to the mean square error of RBF neural with the convergence degree of default error.
It is RBF neural training schematic flow sheet shown in Fig. 2, the flow comprises the following steps：
(2.1) swept frequency and pulling force are given as input vector, gives the admittance of desired output；
(2.2) hidden layer, the output of output layer each unit are obtained；
(2.3) output error is obtained；Judge whether output error meets the requirement of predeterminated target, if so, then terminating to train；
If it is not, then increase neuron, and return to step (2.3).
In embodiment, test specimen 1 is a Q235 girder steel, and girder steel size is 500mm*35mm*4mm, not damaged；Test specimen 2 is
Q235 girder steels, girder steel size are 500mm*35mm*4mm, have the crack damage of 8mm length；Using stretching instrument to girder steel test specimen 1
Apply pulling force, and the admittance signal using electric impedance analyzer test PZT under different pulling force operating modes with test specimen 2.
Using the pulling force compensation method provided by the present invention for non-destructive tests, stretching survey is carried out to test specimen 1 and test specimen 2
Examination；Shown in Fig. 3, the mechanical admittance curves figure and part that are measured under the PZT2 each pulling force the operating mode 0kN, 2kN, 4kN, 6kN, 8kN that are test specimen 1
Enlarged drawing；
Shown in Fig. 4, the mechanical admittance curves that are measured under the PZT3 each pulling force the operating mode 0kN, 4kN, 8kN, 12kN, 16kN that are test specimen 2
Figure and partial enlarged drawing；
From figure 3, it can be seen that the mechanical admittance curves that test specimen 1 is measured in 300kHz to 400kHz frequency bands are as pulling force gradually increases
It is big that downtrend is presented, especially become apparent at the peak value of mechanical admittance curves.
Using the swept frequency of the admittance signal of measurement and different stresses as input layer, admittance information is as output
Layer, form RBF neural network structure；Hidden layer neuron quantity is 10.
By the 1/2 of PZT3 is measured in PZT2 in test specimen 1 and test specimen 2 data as training sample, 1/2 as test specimens
This；Admittance information of No. 2 piezoelectric patches after RBF neural emulates under different pulling force operating modes and the admittance of actual measurement in test specimen 1
Information contrasts；Be test specimen 1PZT2 in value of thrust it is 0kN, 2kN, 4kN shown in Fig. 5, the data comparison under 6kN pulling force operating modes；Fig. 6
It is shown, be test specimen 2PZT3 in value of thrust it is 0kN, 4kN, 8kN, the data comparison under 12kN pulling force operating modes, shows in different pulling force
Admittance data is emulated under operating mode to coincide substantially with actual measurement admittance value；
As can be seen that after the data by different pulling force operating modes are trained, RBF neural can simulate different pulling force works
Admittance data under condition；Contrasted with measured data and can be seen that the admittance data emulated and the admittance data of actual measurement coincide substantially,
The admittance data under different pulling force operating modes under health status can be predicted very well.
Analyzed again with the admittance data of actual measurement with the admittance data predicted, judge admittance data and the actual measurement of prediction
Whether the difference of data meets that default difference it is expected, default difference is desired for 5%~10%, is 6% in embodiment, to mend
Repay influence of the pulling force to admittance data.
Quantitative analysis test specimen not damaged and in the case of having Crack Damage tension RBF compensation effects；Fig. 7 is represented lossless
Under state, the RMSD damage criterions after No. 2 piezoelectric patches compensation of gained test specimen 1 are calculated after RBF is compensated in different pulling force works
0.1 or so under condition, tended towards stability substantially relative to the RMSD indexs of initial data；
Thereby determine that, the health status of test specimen 1 does not change, i.e., not damaged occurs；Because pulling force shadow in initial data
The erroneous judgement to structural health conditions is rung to be corrected after RBF is compensated.
Fig. 8 represents RMSD index comparison diagrams before and after RBF compensation of No. 3 piezoelectric patches of test specimen 2 under nondestructive state；In test specimen
After 2 damage, as shown in figure 9, damage criterion is changed into 0.44 compared to the RMSD indexs after nondestructive state compensation from 0.26, hence it is evident that
Increase, shows that structure is damaged；
With the increase of pulling force, RMSD tends towards stability no significant change, shows that structural damage degree does not change；Thus
It can be seen that the judgement of health status and the judgement of degree of injury after compensation to structure are corrected, EMI is damaged and examined
The pulling force compensation of survey method is effective；Using method provided by the invention, after being compensated to pulling force suffered by structure again
Whether identification structure is damaged, and eliminates influence of the pulling force suffered by structure to non-destructive tests accuracy.
As it will be easily appreciated by one skilled in the art that the foregoing is merely illustrative of the preferred embodiments of the present invention, not to
The limitation present invention, all any modification, equivalent and improvement made within the spirit and principles of the invention etc., all should be included
Within protection scope of the present invention.
Claims (3)
1. a kind of pulling force compensation method for non-destructive tests, it is characterised in that the pulling force compensation method is based on RBF nerve nets
Network, comprise the following steps：
(1) using swept frequency and pulling force as input layer, using admittance value as output layer, RBF nerves are formed together with hidden layer
Network；
(2) RBF neural is trained using the sample data comprising swept frequency, pulling force and admittance value, until
Difference between the admittance value of output admittance value and respective sample data terminates RBF within ± the 5% of sample data admittance value
Neural metwork training；
(3) using the swept frequency of test data and pulling force as input, the RBF neural obtained is trained using step (2), is obtained
Take admittance value corresponding with the swept frequency and pulling force；The admittance value contains external pull information, compensate for pulling force pair
The influence of admittance data；
(4) admittance value obtained according to step (3), the RMSD damage criterions after compensation are obtained；
In RBF neural training described in step (2), the quantity of hidden layer neuron is determined according to following methods：
A, using the input vector corresponding to mean square error caused by RBF neural as weight vector, generate one it is new hidden
Neuron containing layer；
B, using test gained admittance value as input, using emulation gained admittance value as output, with step a generate newly it is hidden
Neuron containing layer, form new RBF neural；
C, the mean square error of RBF neural new described in obtaining step b；
D, repeat step a~c, until the mean square error of RBF neural reaches default mean square error, with hidden layer god now
Quantity through first quantity as RBF neural hidden layer neuron；Wherein, mean square error is preset 5%~10%.
2. the pulling force compensation method described in claim 1, it is characterised in that being instructed to RBF neural described in step (2)
Experienced step, including following sub-step：
(2.1) admittance value of swept frequency, pulling force and desired output is given；
(2.2) using the swept frequency and pulling force as input vector, the RBF neural obtained through step (1) is handled, and is obtained
Admittance value；
(2.3) admittance value of the admittance value and the desired output that are obtained in step (2.2) is made the difference, obtains difference；Judge institute
Difference is stated whether within ± the 5% of desired output admittance value；If so, then terminate to train；If it is not, then increasing neuron, and return
Return step (2.2).
3. pulling force compensation method as claimed in claim 2, it is characterised in that after step (2.3) terminates training, using as follows
Step determines RBF neural training effect, is specially：
(2.4) data of training sample be will differ from as input vector, emulated, obtained with the RBF neural trained
Take the admittance value of simulation data；
(2.5) RMSD values are obtained according to the admittance value of step (2.4) inner simulation data, if the RMSD obtained under each pulling force operating mode
Value fluctuates within 5%, then into step (3)；Otherwise, return to step (2.1), using new sample data re -training RBF
Neutral net.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510627442.3A CN105184364B (en) | 2015-09-28 | 2015-09-28 | A kind of pulling force compensation method for non-destructive tests |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510627442.3A CN105184364B (en) | 2015-09-28 | 2015-09-28 | A kind of pulling force compensation method for non-destructive tests |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105184364A CN105184364A (en) | 2015-12-23 |
| CN105184364B true CN105184364B (en) | 2017-12-12 |
Family
ID=54906427
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510627442.3A Active CN105184364B (en) | 2015-09-28 | 2015-09-28 | A kind of pulling force compensation method for non-destructive tests |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105184364B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106353245B (en) * | 2016-10-13 | 2019-08-30 | 国家电网公司 | The strand extent of corrosion lossless detection method of steel strand wires as aerial earth wire |
| CN108226230B (en) * | 2018-01-05 | 2020-12-01 | 宁波大学 | Method for monitoring compactness defect of grouting material of steel bar sleeve based on piezoelectric impedance effect |
| CN114186586B (en) * | 2021-12-08 | 2024-08-09 | 华中科技大学 | Damage identification method and device based on two-dimensional convolutional neural network |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1450704A (en) * | 2002-04-05 | 2003-10-22 | 清华大学 | Method for compensating dynamic three-phase imbalance load and compensator |
| CN104200005A (en) * | 2014-07-28 | 2014-12-10 | 东北大学 | Bridge damage identification method based on neural network |
| CN104316341A (en) * | 2014-11-17 | 2015-01-28 | 金陵科技学院 | Underground structure damage identification method based on BP neural network |
-
2015
- 2015-09-28 CN CN201510627442.3A patent/CN105184364B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1450704A (en) * | 2002-04-05 | 2003-10-22 | 清华大学 | Method for compensating dynamic three-phase imbalance load and compensator |
| CN104200005A (en) * | 2014-07-28 | 2014-12-10 | 东北大学 | Bridge damage identification method based on neural network |
| CN104316341A (en) * | 2014-11-17 | 2015-01-28 | 金陵科技学院 | Underground structure damage identification method based on BP neural network |
Non-Patent Citations (2)
| Title |
|---|
| 基于EMI损伤检测技术的温度补偿研究;杨景文等;《土木工程与管理学报》;20140930;第7-11页及图1-9 * |
| 基于压电阻抗技术和BP网络的结构健康监测;沈星等;《南京航空航天大学学报》;20100831;第418-422页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105184364A (en) | 2015-12-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Rosales et al. | Crack detection in beam-like structures | |
| Jung et al. | Advanced deep learning model-based impact characterization method for composite laminates | |
| Kewalramani et al. | Concrete compressive strength prediction using ultrasonic pulse velocity through artificial neural networks | |
| Hasançebi et al. | Linear and nonlinear model updating of reinforced concrete T-beam bridges using artificial neural networks | |
| US20240218626A1 (en) | Device and method for health diagnosis of subgrade service performance | |
| CN109325263A (en) | Truss bridge damage position neural network based and damage extent identification method | |
| Law et al. | Crack identification in beam from dynamic responses | |
| Jang et al. | Corrosion estimation of a historic truss bridge using model updating | |
| CN105184364B (en) | A kind of pulling force compensation method for non-destructive tests | |
| Mao et al. | The construction and comparison of damage detection index based on the nonlinear output frequency response function and experimental analysis | |
| Liu et al. | Diagnosis of structural cracks using wavelet transform and neural networks | |
| Wu et al. | Crack diagnosis method for a cantilevered beam structure based on modal parameters | |
| Zhao et al. | Quantitative diagnosis method of beam defects based on laser Doppler non-contact random vibration measurement | |
| Saltan et al. | Hybrid neural network and finite element modeling of sub-base layer material properties in flexible pavements | |
| Vanniamparambil et al. | An active–passive acoustics approach for bond-line condition monitoring in aerospace skin stiffener panels | |
| Zhao et al. | An indirect comparison quasi-static calibration method for piezoelectric pressure sensors based on an inverse model | |
| Gonsalez-Bueno | An investigation into the way in which longitudinal and flexural waves interact with corrosion-like damage | |
| CN103399974B (en) | Quantize the method comparing random vibration emulated data and experimental data | |
| Nguyen | A deep learning platform for evaluating energy loss parameter in engineering structures | |
| Xu et al. | Damage identification of simply-supported bridges using impact response-based recurrence graph | |
| Gupta et al. | Multiple damage identification in a beam using artificial neural network-based modified mode shape curvature | |
| Vijayan | Prediction of displacement in Reinforced concrete based on artificial neural networks using sensors | |
| Sofge | Structural health monitoring using neural network based vibrational system identification | |
| Al-Rahmani et al. | A combined soft computing-mechanics approach to inversely predict damage in bridges | |
| Pekcan et al. | Artificial neural network based backcalculation of conventional flexible pavements on lime stabilized soils |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |