CN115728396A - Acoustic emission signal feature extraction method based on pressure-bearing equipment activity defect monitoring - Google Patents
Acoustic emission signal feature extraction method based on pressure-bearing equipment activity defect monitoring Download PDFInfo
- Publication number
- CN115728396A CN115728396A CN202211534524.XA CN202211534524A CN115728396A CN 115728396 A CN115728396 A CN 115728396A CN 202211534524 A CN202211534524 A CN 202211534524A CN 115728396 A CN115728396 A CN 115728396A
- Authority
- CN
- China
- Prior art keywords
- acoustic emission
- distance
- class
- characteristic
- sample
- 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
Abstract
The invention relates to an acoustic emission signal characteristic extraction method based on pressure equipment active defect monitoring, which is used for extracting acoustic emission waveform characteristic parameters aiming at the actual parameter characteristics of pressure equipment active defect monitoring acoustic emission signals; using the extracted characteristic parameters as characteristic vectors, and defining the distance between variables by using similarity coefficients; performing hierarchical clustering, and classifying the parameters into one class one by one according to similarity coefficients among the parameters to obtain a hierarchical clustering tree of acoustic emission characteristic parameter correlation; setting a correlation threshold, and keeping parameters of the correlation coefficient below the correlation threshold as characteristic vectors of cluster analysis; and performing K-means clustering analysis on the selected vector on the basis of determining the characteristic vector of the clustering analysis. The method reduces the effective signals deleted by mistake, and performs effective identification, thereby improving the accuracy of the analysis and prediction result of the active defect monitoring acoustic emission signals of the pressure-bearing equipment.
Description
Technical Field
The invention relates to an acoustic emission signal feature extraction method based on hierarchical clustering and a K-means algorithm, in particular to an acoustic emission signal feature extraction method based on pressure equipment activity defect monitoring.
Background
An acoustic emission signal is an elastic stress wave from the inside of a material due to a sudden release of strain energy, which contains information about the damage inside the material, so that acoustic emission techniques can be used to identify different types of damage occurring in the loaded material. In the process of carrying out acoustic emission online detection on the normal-pressure vertical storage tank, because a certain amount of interference caused by environment and electromagnetic noise exists in original acoustic emission data, the original data of a storage tank bottom plate is generally analyzed and processed by adopting a traditional clustering method, a large number of interference signals in the original data are eliminated, and acoustic emission correlation analysis comparison is carried out on data files before and after processing to obtain new storage tank detection data. However, in terms of extracting the characteristic parameters of the acoustic emission waveform, although the amplitude and the energy can well reflect the intensity characteristics of the acoustic emission signal, the duration and the counting cannot reflect the shape characteristics of the acoustic emission waveform; in addition, the conventional signal processing method inevitably deletes the effective signal by mistake in the process of signal noise reduction, and as a result, the signal analysis is inaccurate. Therefore, more advanced signal analysis techniques are needed to identify the different acoustic emission sources in the determination of the damage category.
Disclosure of Invention
The invention aims to overcome the defects and provide the acoustic emission signal feature extraction method based on the monitoring of the activity defect of the pressure-bearing equipment, so as to effectively identify the acoustic emission signal feature extraction method and further improve the accuracy of the analysis and prediction result of the monitoring acoustic emission signal of the activity defect of the pressure-bearing equipment.
The purpose of the invention is realized as follows:
a preparation method of an acoustic emission signal feature extraction method based on pressure equipment activity defect monitoring comprises the following steps:
s1: monitoring actual parameter characteristics of an acoustic emission signal aiming at the active defect of the pressure-bearing equipment, and extracting characteristic parameters of the acoustic emission waveform;
s2: using the extracted characteristic parameters as characteristic vectors, and defining the distance between variables by using similarity coefficients;
s3: performing hierarchical clustering, and classifying the parameters into one class one by one according to similarity coefficients among the parameters to obtain a hierarchical clustering tree of acoustic emission characteristic parameter correlation;
s4: setting a correlation threshold, and keeping parameters of the correlation coefficient below the correlation threshold as characteristic vectors of cluster analysis;
s5: and performing K-means clustering analysis on the selected vector on the basis of determining the characteristic vector of the clustering analysis.
Further, in the above scheme, the extracting acoustic emission waveform characteristic parameters in the first step mainly includes:
a. selecting amplitude and energy to reflect the waveform intensity characteristics of the acoustic emission signal;
b. determining A% of the maximum amplitude as a soft threshold value, and recalculating the characteristic parameters of the waveform to enable the obtained characteristic parameters to reflect the shape characteristics of the acoustic emission waveform;
c. and reflecting the shape characteristics of the acoustic emission waveform by using a margin factor, wherein the expression of the margin factor is shown as formula 1:
where N represents the number of data points of the analysis sample, x n Represents the time history of the sample signal, and T represents the time of all samples;
d. through wavelet decomposition, a wavelet characteristic energy spectrum coefficient vector of the acoustic emission signals is calculated in batch according to wavelet characteristic energy spectrum coefficient definition and is used for reflecting waveform frequency distribution characteristics.
Further, the similarity coefficient in step two in the above scheme is defined as follows:
let C ij Is X i And X j The similarity coefficient between the two groups is limited as follows: is provided with|C ij The | is less than or equal to 1, and the j is true for all the i and j; c ij =C ji For all i, j holds; absolute value of similarity coefficient | C ij The closer | is to 1,X i And X j The closer to C ij Close to 0, the farther away the relationship between the two is;
for quantitative variables, the similarity factor used is X i And X j The correlation coefficient of (a); variable X i And X j Is usually given by r ij Is shown, here denoted as C ij I.e. by
When C is present ij If =1, it means that two variables are related, and in general, | C ij |≤1;
Defining the distance between variables by means of similarity coefficients, i.e.
Further, in the third step in the above scheme, the hierarchical clustering step is as follows:
SS3.1, calculating the distance between each two of the n parameters to obtain a distance matrix D between the samples (0) ;
SS3.2, the initial (first step: i = 1) n samples each constitute one class, the number of classes k = n, the ith class G i ={X (i) } (i =1,.., n); the distance between such species is the distance between the samples (i.e., D) (1) =D (0) ) (ii) a Then for sample X (i) (i = 1.. Multidot.n) steps SS3.3 and SS3.4 of the execution and classification process;
SS3.3, distance matrix D obtained in step S2 (i-1) And merging the two classes with the minimum distance between the classes into a new class. The total number k of classes at this time is reduced by 1 class, i.e. k = n-i +1;
SS3.4, calculating the distance between the new class and other classes to obtain a new distance matrix D (i) (ii) a If the total number of the merged classes is still larger than 1, repeating the steps SS3.3 and SS3.4, and turning to the step (5) until the total number of the classes is l;
SS3.5, drawing a spectrum series cluster diagram;
SS3.6, determining the number of the classes and the members of the classes.
Further, in the fifth step in the above scheme, the K-means clustering step is as follows:
SS5.1, specifying the distance between the samples. Three numbers are artificially defined: k (number of classes), C (minimum of inter-class distance) and R (maximum of intra-class distance); taking the first k sample points as condensation points;
SS5.2, calculating the distance between every two k condensation points, if the minimum distance is smaller than C, combining the corresponding two condensation points, using the gravity centers of the two condensation points as new condensation points, and repeating the step (2) until the distances between all the condensation points are larger than C;
SS5.3, classifying the remaining n-k samples one by one, and calculating the distance between each sample and all condensation points for each mascot, wherein if the minimum distance is greater than R, the sample is taken as a new condensation point; if the minimum distance is less than or equal to R, classifying the sample into the class of the condensation point closest to the sample; then, the gravity centers of the samples are recalculated, the gravity centers are used as new condensation points, if the distances between the condensation points are all larger than or equal to C, the next sample is considered, otherwise, the samples are merged in the step (2) and then the next sample is considered until all the samples are classified;
SS5.4, classifying the samples from beginning to end one by one according to the step (3), wherein the difference is as follows: after a certain sample is classified, if the classification is consistent with the original classification, the gravity center does not need to be calculated; if the classification is different from the original classification, the gravity centers of the two types need to be counted again;
if the new classification is the same as the last classification, the clustering process is ended, otherwise, the step SS5.4 is repeated;
the optimal clustering number k in the k-means clustering is determined by Dacies & Bouldin criterion, and the criterion is defined as formula 3:
wherein d is i And d j The average distances within classes i and j, respectively, D ij Representing the distance between class i and j, the number of clusters is optimal when DB reaches a minimum value.
The new storage tank detection data obtained by the method is applied to deducing the property, the approximate region and the severity of the acoustic emission source, so that the active defect monitoring state of the pressure-bearing equipment is evaluated, and maintenance decision is guided.
Compared with the prior art, the invention has the beneficial effects that:
(1) The extraction method of the invention determines the margin factor of the acoustic emission signal according to the waveform characteristic of the acoustic emission source, and determines A% of the maximum amplitude of the waveform as a soft threshold to recalculate the characteristic parameter of the waveform, so that the obtained characteristic parameter can reflect the shape characteristic of the acoustic emission waveform, and the parameter space is easy to classify.
(2) According to the method, the hierarchical clustering algorithm and the K-means algorithm are used for processing the active defect monitoring acoustic emission data of the pressure-bearing equipment, so that effective signals of mistaken deletion are reduced, various signals are effectively classified and effectively identified, and the accuracy of the analysis and prediction result of the active defect monitoring acoustic emission signals of the pressure-bearing equipment is improved.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
Detailed Description
For a better understanding of the technical aspects of the present invention, reference will now be made in detail to the accompanying drawings. It should be understood that the following specific examples are not intended to limit the embodiments of the present invention, but are merely exemplary embodiments of the present invention. It should be noted that the description of the positional relationship of the components, such as the component a is located above the component B, is based on the description of the relative positions of the components in the drawings, and is not intended to limit the actual positional relationship of the components.
Example 1:
referring to fig. 1, fig. 1 is a schematic flow chart of an acoustic emission signal feature extraction method based on pressure equipment activity defect monitoring. As shown in the figure, the acoustic emission signal feature extraction method based on pressure equipment activity defect monitoring comprises the following steps:
step one, characteristic parameter extraction:
through wavelet decomposition, defining a wavelet characteristic energy spectrum coefficient vector of the acoustic emission signals in batch according to the wavelet characteristic energy spectrum coefficient, and reflecting waveform frequency distribution characteristics;
selecting amplitude and energy to reflect the waveform intensity characteristics of the acoustic emission signals;
determining A% of the maximum amplitude as a soft threshold value to recalculate the characteristic parameters of the waveform, so that the obtained characteristic parameters reflect the shape characteristics of the acoustic emission waveform;
and reflecting the shape characteristics of the acoustic emission waveform by using a margin factor, wherein the expression of the margin factor is shown as formula 1:
where N represents the number of data points of the analysis sample, x n Representing the time history of the sample signal and T representing the time of all samples.
Step two, defining the distance between variables:
utilizing the parameters and vectors as feature vectors describing the acoustic emission signals,
let C ij Is X i And X j The similarity coefficient between the two groups is limited as follows: is provided with|C ij The | is less than or equal to 1, and the j is true for all the i and j; c ij =C ji For all i, j holds; absolute value of similarity coefficient | C ij The closer | is to 1,X i And X j The closer to C ij Close to 0, the more distant the relationship.
For quantitative variables, the similarity factor used is X i And X j The correlation coefficient of (a); the correlation coefficient is the cosine of the included angle after the data are standardized; variable X i And X j Is usually given by r ij Denotes here we denote C ij I.e. by
When C is present ij The expression "= 1" means that two variables are related, in general, | C ij |≤1。
Defining the distance between variables by means of similarity coefficients, i.e.
Step three, hierarchical clustering:
too many redundant parameters lead to different categories tending to be similar, hierarchical clustering is used for analyzing the similarity among the acoustic emission parameters, and is a method for changing the categories from more to less, so hierarchical clustering is adopted, and the hierarchical clustering steps are as follows:
(1) Calculating the distance between every two of the n parameters to obtain a distance matrix D between the samples (0) 。
(2) The initial (first step: i = 1) n samples each constitute one class, the number of classes k = n, the i-th class G i ={X (i) }(i=1,...,n)。The distance between such species is the distance between the samples (i.e., D) (1) =D (0) ). Then for sample X (i) (i = 1...., n) performing steps (3) and (4) of the merging process.
(3) For the distance matrix D obtained in the step (2) (i=1) The two classes with the minimum distance between the classes are merged into a new class. The total number of classes k in this case is reduced by 1 class, i.e. k = n-i +1.
(4) Calculating the distance between the new class and other classes to obtain a new distance matrix D (i) . And (5) if the total number of the merged classes is still larger than 1, repeating the steps (3) and (4) until the total number of the classes is l.
(5) The score is a cluster map.
(6) The number of classes and the members of the classes are determined.
And (3) by utilizing hierarchical clustering, classifying the parameters into one class one by one according to the similarity coefficient among the parameters to obtain a hierarchical clustering tree of the acoustic emission characteristic parameter correlation.
Step four, setting a relevant threshold:
and (4) carrying out hierarchical clustering on the parameters obtained by calculation to obtain a hierarchical clustering tree of acoustic emission characteristic parameter correlation. And setting a correlation threshold, and keeping parameters below a correlation coefficient as a characteristic vector of the cluster analysis.
Step five, K mean value clustering analysis:
analyzing the similarity of each parameter of acoustic emission data through hierarchical clustering, and performing clustering analysis by taking the selected feature vector as a parameter of K-means clustering, wherein the method comprises the following steps:
(1) The distance between the samples is specified. Three numbers are artificially defined: k (number of classes), C (minimum of inter-class distance) and R (maximum of intra-class distance); the first k sample points were taken as the condensation points.
(2) And (3) calculating the distance between every two k condensation points, merging the corresponding two condensation points if the minimum distance is smaller than C, taking the gravity center of the two condensation points as a new condensation point, and repeating the step (2) until the distances between all the condensation points are larger than C.
(3) Classifying the rest n-k samples one by one, and calculating the distance between each sample and all condensation points for each mascot, wherein if the minimum distance is more than R, the sample is taken as a new condensation point; if the minimum distance is less than or equal to R, the sample is classified as the one with the condensation point closest to the sample. Then, the barycenter of the class is recalculated, and the barycenter is used as a new condensation point. Such as the distance between the condensation points is more than or equal to C. The next sample is considered, otherwise the next sample is considered after combining in step (2) until all samples are classified.
(4) Classifying the samples from beginning to end one by one according to the step (3), wherein the difference is as follows: after a certain sample is classified, if the classification is consistent with the original classification, the gravity center does not need to be calculated; if the classification is different from the original, the barycenters of the two classes involved are recalculated.
If the new classification is the same as the last one, the clustering process ends, otherwise step (4) is repeated.
When k-means clustering is performed, the optimal clustering number of clustering analysis is determined by Dacies & Bouldin criteria (a clustering algorithm for evaluation metric proposed by grand satellite L-davis and tang nard-Bouldin), and is defined as shown in formula 3:
wherein d is i And d j The average distances within classes i and j, respectively, D ij Representing the distance between class i and j, the number of clusters is optimal when DB reaches a minimum value.
The above is only a specific application example of the present invention, and the protection scope of the present invention is not limited in any way. All the technical solutions formed by equivalent transformation or equivalent replacement fall within the protection scope of the present invention.
Claims (8)
1. An acoustic emission signal feature extraction method based on pressure-bearing equipment active defect monitoring is characterized by comprising the following steps:
s1, monitoring actual parameter characteristics of an acoustic emission signal aiming at active defects of pressure-bearing equipment, and extracting acoustic emission waveform characteristic parameters;
s2, the extracted characteristic parameters are used as characteristic vectors, and the distance between variables is defined by using similarity coefficients;
s3, hierarchical clustering is carried out, the parameters are classified into one class one by one according to similarity coefficients among the parameters, and a hierarchical clustering tree of acoustic emission characteristic parameter correlation is obtained;
s4, setting a threshold, and keeping parameters with the correlation coefficient below the threshold as a characteristic vector of the cluster analysis;
and S5, performing K-means clustering analysis on the selected vector on the basis of determining the characteristic vector of the clustering analysis.
2. The acoustic emission signal feature extraction method based on pressure bearing equipment activity defect monitoring according to claim 1, characterized in that: the step S1 of extracting acoustic emission waveform characteristic parameters includes the following steps:
selecting amplitude and energy to reflect the intensity characteristics of the acoustic emission signal;
determining A% of the maximum amplitude as a soft threshold value, and recalculating the characteristic parameters of the waveform to enable the obtained characteristic parameters to reflect the shape characteristics of the acoustic emission waveform;
reflecting the shape characteristics of the acoustic emission waveform by using a margin factor;
and through wavelet decomposition, defining a wavelet characteristic energy spectrum coefficient vector of the acoustic emission signals in batch according to the wavelet characteristic energy spectrum coefficient to reflect the waveform frequency distribution characteristics.
3. The acoustic emission signal feature extraction method based on pressure bearing equipment activity defect monitoring according to claim 2, characterized in that: the margin factor expression is shown in formula 1:
where N represents the number of data points of the analysis sample, x n Representing the time history of the sample signal, T representing the time of all samplesAnd (3) removing the solvent.
4. The acoustic emission signal feature extraction method based on pressure bearing equipment activity defect monitoring according to claim 1, characterized in that: the similarity coefficient in step S2 is defined as follows:
let C ij Is X i And X j The similarity coefficient between the two groups is limited as follows: is provided with|C ij The | is less than or equal to 1, and the j is true for all the i and j; c ij =C ji For all i, j holds;
absolute value of similarity coefficient | C ij The closer to 1,X | is i And X j The closer to C ij Close to 0, the farther away the relationship between the two is; for quantitative variables, the similarity factor used is X i And X j The correlation coefficient of (a);
5. The acoustic emission signal feature extraction method based on pressure bearing equipment activity defect monitoring according to claim 4, characterized in that: the correlation coefficient is the cosine of an included angle after data are subjected to standardization processing; variable X i And X j Is usually given by r ij Is represented by C ij I.e. by
When C is ij If =1, it means that two variables are related, and in general, | C ij |≤1。
6. The acoustic emission signal feature extraction method based on pressure bearing equipment activity defect monitoring of claim 1, characterized by comprising the following steps: the hierarchical clustering in step S3 includes the following:
SS3.1, calculating the distance between each two of the n parameters to obtain a distance matrix D between the samples (0) ;
SS3.2, the initial (first step: i = 1) n samples each constitute one class, the number of classes k = n, the ith class G i ={X (i) } (i =1,.., n); the distance between such species is the distance between the samples (i.e., D) (1) =D (0) ) (ii) a Then to sample X (i) (i = 1...., n) performing steps SS3.3 and SS3.4 of the parallel process;
SS3.3, distance matrix D obtained in step SS3.2 (i-1) Combining the two classes with the minimum distance between the classes into a new class; the total number k of classes at this time is reduced by 1 class, i.e. k = n-i +1;
SS3.4, calculating the distance between the new class and other classes to obtain a new distance matrix D (i) If the total number of the merged classes is still larger than 1, repeating the steps SS3.3 and SS3.4, and turning to the step SS3.5 until the total number of the classes is l;
SS3.5, drawing a spectrum series clustering chart;
SS3.6, determining the number of the classes and the members of the classes.
7. The acoustic emission signal feature extraction method based on pressure bearing equipment activity defect monitoring according to claim 1, characterized in that: the K-means clustering in step S5 includes the following:
SS5.1, define the distance between samples, define three numbers: k is the classification number, C is the minimum value of the inter-class distance, and R is the maximum value of the intra-class distance; taking the first k sample points as condensation points;
SS5.2, calculating the distance between every two k condensation points, if the minimum distance is less than C, combining the corresponding two condensation points, using the gravity centers of the two condensation points as new condensation points, and repeating the step SS5.2 until the distances between all the condensation points are more than C;
SS5.3, classifying the remaining n-k samples one by one, and calculating the distance between each sample and all condensation points for each auspicious sample, wherein if the minimum distance is greater than R, the sample is taken as a new condensation point; if the minimum distance is less than or equal to R, classifying the sample into the class of the condensation point closest to the sample; then, the gravity center of one type is recalculated, and the gravity center is used as a new condensation point; if the distances between the condensation points are all larger than or equal to C, considering the next sample, otherwise, combining the samples in the step SS5.2 and then considering the next sample until all the samples are classified;
SS5.4, classifying the samples from beginning to end one by one according to the step SS5.3, wherein the difference is as follows: after a certain sample is classified, if the classification is consistent with the original classification, the gravity center does not need to be calculated; if the classification is different from the original classification, the gravity centers of the two types are recalculated; if the new classification is the same as the last classification, the clustering process is ended, otherwise, the step SS5.4 is repeated;
the optimal clustering number k in the k-means clustering is determined by Dacies & Bouldin criterion.
8. The acoustic emission signal feature extraction method based on pressure bearing equipment activity defect monitoring according to claim 7, characterized in that: the Dacies & Bouldin criterion is defined as shown in equation 3:
wherein d is i And d j The average distances in class i and class j, respectively, D ij Representing the distance between the i-class and the j-class, and the clustering number is optimal when DB reaches the minimum value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211534524.XA CN115728396A (en) | 2022-12-02 | 2022-12-02 | Acoustic emission signal feature extraction method based on pressure-bearing equipment activity defect monitoring |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211534524.XA CN115728396A (en) | 2022-12-02 | 2022-12-02 | Acoustic emission signal feature extraction method based on pressure-bearing equipment activity defect monitoring |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115728396A true CN115728396A (en) | 2023-03-03 |
Family
ID=85301247
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211534524.XA Pending CN115728396A (en) | 2022-12-02 | 2022-12-02 | Acoustic emission signal feature extraction method based on pressure-bearing equipment activity defect monitoring |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115728396A (en) |
-
2022
- 2022-12-02 CN CN202211534524.XA patent/CN115728396A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109543763B (en) | Raman spectrum analysis method based on convolutional neural network | |
CN109657011B (en) | Data mining system for screening terrorist attack event crime groups | |
CN110826618A (en) | Personal credit risk assessment method based on random forest | |
CN112036476A (en) | Data feature selection method and device based on two-classification service and computer equipment | |
CN111104398A (en) | Detection method and elimination method for approximate repeated record of intelligent ship | |
CN114548272A (en) | Centrifugal pump cavitation state identification method | |
CN111967535A (en) | Fault diagnosis method and device for temperature sensor in grain storage management scene | |
CN110837874A (en) | Service data abnormity detection method based on time series classification | |
CN113364751A (en) | Network attack prediction method, computer-readable storage medium, and electronic device | |
CN115600194A (en) | Intrusion detection method, storage medium and device based on XGboost and LGBM | |
CN112926592B (en) | Trademark retrieval method and device based on improved Fast algorithm | |
CN108090635B (en) | Road performance prediction method based on cluster classification | |
CN117033912B (en) | Equipment fault prediction method and device, readable storage medium and electronic equipment | |
CN114167237A (en) | GIS partial discharge fault identification method and system, computer equipment and storage medium | |
CN111882554B (en) | SK-YOLOv 3-based intelligent power line fault detection method | |
CN113468538A (en) | Vulnerability attack database construction method based on similarity measurement | |
CN109508350B (en) | Method and device for sampling data | |
CN115728396A (en) | Acoustic emission signal feature extraction method based on pressure-bearing equipment activity defect monitoring | |
CN113523904A (en) | Cutter wear detection method | |
Biju et al. | A genetic algorithm based fuzzy C mean clustering model for segmenting microarray images | |
CN115563500A (en) | Power distribution equipment partial discharge mode identification method, device and system based on data enhancement technology | |
CN115331661A (en) | Voiceprint recognition backdoor attack defense method based on feature clustering analysis and feature dimension reduction | |
CN115438239A (en) | Abnormity detection method and device for automatic abnormal sample screening | |
KR101408345B1 (en) | Data processing apparatus and method for classifying data, media for writing the method | |
CN110766087A (en) | Method for improving data clustering quality of k-means based on dispersion maximization method |
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 |