CN115269870A - Method for realizing classification and early warning of data link faults in data based on knowledge graph - Google Patents
Method for realizing classification and early warning of data link faults in data based on knowledge graph Download PDFInfo
- Publication number
- CN115269870A CN115269870A CN202210884956.7A CN202210884956A CN115269870A CN 115269870 A CN115269870 A CN 115269870A CN 202210884956 A CN202210884956 A CN 202210884956A CN 115269870 A CN115269870 A CN 115269870A
- Authority
- CN
- China
- Prior art keywords
- data
- fault
- knowledge graph
- knowledge
- svm
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/237—Lexical tools
- G06F40/242—Dictionaries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/237—Lexical tools
- G06F40/247—Thesauruses; Synonyms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to a method for realizing classification and early warning of faults of a data link in data based on a knowledge graph. Through the mutual cooperation of the machine learning fault classification module and the knowledge map fault early warning module, the intelligent analysis of the data link is realized, a series of problems existing in an expert system and machine learning are solved, the maintenance efficiency of the data link is effectively improved, and the fault classification accuracy is greatly improved.
Description
Technical Field
The invention belongs to the field of power Internet of things, relates to a data center station technology, and particularly relates to a method for realizing data link fault classification early warning of a data center station based on a knowledge graph.
Background
With the rapid development of the power internet of things and the promotion of data platforms, the data volume of power resources is continuously enlarged, and more data manufacturers and data users appear. But at the same time, the problems of dispersed data distribution, huge data scale, complex data interaction, low data transmission efficiency, difficult diagnosis of data link faults and the like occur.
Most of the existing methods for analyzing data link failures focus on failure analysis in terms of physical hardware, and analysis at a software level is relatively few. The fault classification method for entity hardware mainly comprises the following steps: expert systems, and machine learning, etc., among which the most widely used are expert systems, which are classified into expert knowledge based on a shallow knowledge field and model knowledge based on a deep knowledge analysis object. The machine learning does not need to manually arrange and summarize knowledge, and only needs to use a related data set for training to obtain a classification model of the fault, thereby obtaining a better effect in the field of fault diagnosis.
Although the expert system can effectively simulate the fault diagnosis process completed by the fault diagnosis expert, in practical application, the problems of difficulty in obtaining a complete knowledge base, very slow diagnosis speed, high operation and maintenance difficulty, no learning capability, poor fault tolerance and the like still exist. Therefore, machine learning models are increasingly widely applied to fault classification problems, but not only a large number of labeled data sets are required for fault prediction by using a single machine learning method, but also the classification effect greatly depends on the setting of training time, training parameters and the like, and the fault diagnosis effect may be poor.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, provides a method for realizing classification and early warning of data link faults of a data center station based on a knowledge graph, mainly designs and realizes a deep learning-based data link fault classification method, classifies the existing faults in the data link, and early warns the faults which are likely to occur in the future, thereby effectively improving the maintenance efficiency of the data link.
The technical problem to be solved by the invention is realized by adopting the following technical scheme:
a method for realizing fault classification early warning of a data link of a data center station based on a knowledge graph comprises the following specific steps:
(1) Establishing a Kmeans-SVM fault classification model
Firstly, using a PCA method to reduce the dimension of data, using a K-means method to perform dimension reduction processing on the data, clustering unlabeled substation data by using the unsupervised learning characteristic of the data, wherein the cluster number K is set to be 2, the distance standard used in the clustering process is Euclidean distance, and the sum of squares of errors is used as a representative of the similarity of samples in clusters, so that a certain cluster G is divideduSmaller sum of squared errors in (c) indicates greater similarity of samples within the cluster, and conversely, GuThe larger the sum of squared errors (c) is, the smaller the similarity of the samples in the cluster is, and the calculation formula of the sum of squared errors (c) is as follows:
the input unlabelled data are changed into labeled data through a K-means method, the input data are finally divided into two types, the two types are respectively represented by numbers 1 and 2, then an SVM model is trained by using the labeled data to obtain a maximum interval hyperplane, the two types of data 1 and 2 are respectively divided on two sides of the hyperplane, the accuracy of a classification result obtained by the K-means is tested by using the maximum interval hyperplane after SVM training is finished, the correctly predicted data is used for retraining the hyperplane of the SVM, the hyperplane of the SVM is iteratively updated by the method until the error rate of the predicted data of the SVM is not changed any more, and the final maximum interval hyperplane is obtained, and the specific algorithm steps are as follows:
the method comprises the following steps: preprocessing the non-tag data, dividing the preprocessed data into two clusters by using a K-means algorithm, respectively marking the two clusters as 1 and 2, and converting the non-tag data into tagged data;
step two: training an SVM classifier model by using the labeled data obtained in the step one to obtain a maximum interval hyperplane;
step three: testing K-means clustering by using the trained SVM classification model to obtain two sample data, and taking the accurately predicted data as the input data of the second step again to obtain a new maximum interval hyperplane until the error rate of the data predicted by the SVM is not changed any more;
(2) Constructing fault domain knowledge graph
The method mainly comprises the following steps of constructing a knowledge graph in the link fault field, wherein the knowledge graph is mainly divided into four parts, namely entity and attribute extraction, coreference resolution, knowledge processing and data integration;
(3) Data link failure early warning
And carrying out knowledge reasoning according to the logical relationship between the fault node information and the fault reason in the link fault knowledge graph.
Moreover, the extraction of entities and attributes includes: firstly, performing word segmentation operation on a corpus in a fault field, extracting entities and attributes by using a Markov model, taking the extracted entities and attributes as the entities and attributes finally applied to a knowledge graph, and then performing part-of-speech tagging on all words, wherein the words are divided into the following categories: the system comprises a fault noun entity, a fault phenomenon verb, a fault degree adverb, a fault degree quantifier and an original dictionary of non-extracted words.
And finding out synonyms representing the entities and the attributes, classifying words with high similarity into the same class, and representing the synonyms in a form of a synonym table.
The main purpose of the knowledge processing step is to identify the correspondence between the entity and the attribute, and delete redundant inclusion relations using as a criterion whether there is an inclusion or non-inclusion relation between parts of speech.
And in the data integration step, the fault entity, the attribute and the relation triple are combined, and the final map is constructed through the node updating of the concept layer and the entity attribute layer to finally form the knowledge map in the link fault field.
Moreover, the process of carrying out knowledge reasoning on the logical relationship between the fault node information and the fault reason in the link fault knowledge map is as follows: firstly, extracting rules of stored known knowledge, matching the rules after extracting the rules, adding the rules to a rule execution area if the matching is successful, resolving conflicts if the rule conflicts are established, obtaining an unrealized reasoning result if the rule conflicts are not established, and similarly, obtaining a reasoning result without reasoning new knowledge if the rule matching is unsuccessful.
The invention has the advantages and positive effects that:
1. the method trains a Kmeans-SVM fault classification model, automatically verifies, evaluates and adjusts parameters of the model through an optimization algorithm, tests the model by using a test set, finally classifies the current fault condition of a data link, transmits a diagnosis result to a knowledge graph fault early warning module, and provides a current information basis for fault early warning. And converting the unlabeled data into labeled data by using a K-means method, and iteratively training a final maximum interval hyperplane by using an SVM algorithm. The method has the advantages that the K-means method can save the cost of manual marking, and the fault classification accuracy is greatly improved by iteratively solving the maximum interval hyperplane.
2. According to the invention, through constructing the knowledge graph in the fault field, the classification result data of the machine learning fault classification module is input into the knowledge graph fault early warning module, and the associated fault can be found out, so that effective early warning is carried out. The knowledge graph can link massive different kinds of information together and form a relationship network, so that a user can analyze problems through the angle of relationship. The invention not only can classify the current fault type, but also can early warn future faults through the established fault domain knowledge graph, thereby greatly improving the maintenance efficiency of the data link.
3. Aiming at the problems existing in an expert system and a single machine learning method in the prior art, the invention firstly uses a Kmeans-SVM combined machine learning method to construct a classification model, then carries out early warning of associated faults by constructing a fault domain knowledge graph, realizes intelligent analysis of a data link through mutual cooperation of a machine learning fault classification module and a knowledge graph fault early warning module, solves a series of problems existing in the expert system and machine learning, and effectively improves the maintenance efficiency of the data link.
Drawings
FIG. 1 is a flow chart of the method for establishing a Kmeans-SVM fault classification model;
fig. 2 is a logical relationship inference diagram between failure node information and failure cause in the link failure knowledge graph of the present invention.
Detailed Description
The present invention is further described in the following examples, which are intended to be illustrative, not limiting and are not intended to limit the scope of the invention.
The method comprises the steps of firstly, taking fault classification in a data link as a target, and obtaining a fault classification model through Kmeans-SVM model training. And then constructing a knowledge graph of the fault types by using a Markov model and the like. The fault classification basis is obtained based on a fault domain knowledge graph and a Kmeans-SVM method, and then the incidence relation between faults is analyzed through fault reason similarity, so that the incidence faults possibly occurring in the data link are predicted.
The invention provides a method for realizing classification and early warning of data link faults in data based on a knowledge graph, which comprises the following steps:
(1) Establishing a Kmeans-SVM fault classification model
The specific modeling method flow is shown in fig. 1, and the dimensionality reduction is performed on the data by using a PCA method. And clustering the unlabeled substation data by using a K-means method and utilizing the unsupervised learning characteristic of the data after the dimensionality reduction, wherein the cluster number K is set to be 2, and the distance standard used in the clustering process is Euclidean distance. Using the sum of squared errors as a representative of the similarity of the samples in the cluster, and dividing a certain cluster GuThe smaller the sum of squared errors, the greater the similarity of samples within the cluster; and, conversely, GuThe larger the sum of squared errors in (b) indicates the smaller the similarity of samples within the cluster. The equation for the sum of the squares of the errors is as follows:
the input label-free data is changed into label data through a K-means method. The input data is finally divided into two categories, denoted by the numbers 1 and 2, respectively. And training the SVM model by using the labeled data to obtain a maximum interval hyperplane, and respectively dividing the two types of data 1 and 2 on two sides of the hyperplane. And after the SVM training is finished, testing the accuracy of the classification result obtained by K-means by using a maximum interval hyperplane, using the correctly predicted data to retrain the hyperplane of the SVM, and iteratively updating the hyperplane of the SVM by the method until the error rate of the predicted data by using the SVM is not changed any more to obtain the final maximum interval hyperplane. The algorithm comprises the following steps:
the method comprises the following steps: preprocessing the non-tag data, dividing the preprocessed data into two clusters by using a K-means algorithm, respectively marking the two clusters as 1 and 2, and converting the non-tag data into tagged data.
Step two: and (4) training an SVM classifier model by using the labeled data obtained in the step one to obtain a maximum interval hyperplane.
Step three: and (3) testing K-means clustering by using the trained SVM classification model to obtain two sample data, and taking the accurately predicted data as the input data of the second step again to obtain a new maximum interval hyperplane until the error rate of the data predicted by the SVM is not changed any more.
(2) Constructing fault domain knowledge graphs
The method mainly comprises four parts of entity and attribute extraction, coreference resolution, knowledge processing and data integration when the knowledge graph in the link fault field is constructed.
Extraction of entities and attributes: firstly, performing word segmentation operation on a corpus in a fault field, extracting entities and attributes by using a Markov model, and taking the extracted entities and attributes as the entities and attributes finally applied to a knowledge graph. Then, part-of-speech tagging is performed on all words, and the words are divided into the following categories: the system comprises a fault noun entity, a fault phenomenon verb, a fault degree adverb, a fault degree quantifier and an original dictionary of non-extracted words.
Performing coreference resolution: the method mainly aims to find out synonyms representing entities and attributes, classify words with high similarity into the same class, and use a form of a synonym table to represent the synonyms.
Knowledge processing: the main purpose of the step is to identify the corresponding relationship between the entity and the attribute, and delete the redundant inclusion relationship by taking the relationship whether each part of speech contains or does not contain as a standard.
Data integration: the step combines the fault entity, the attribute and the relation triple, and realizes the final map construction through the node updating of the concept layer and the entity attribute layer, and finally forms the knowledge map of the link fault field.
(3) Data link failure early warning
Performing knowledge inference according to a logical relationship between the fault node information and the fault reason in the link fault knowledge graph, wherein an inference rule is shown in fig. 2, and a process of performing knowledge inference on the logical relationship between the fault node information and the fault reason in the link fault knowledge graph is as follows: firstly, rule extraction is carried out on stored known knowledge, rule matching is carried out after rule extraction, if matching is successful, the rule is added to a rule execution area, conflict resolution is carried out if rule conflict is established, if rule conflict is not established, an unrealistic reasoning result is obtained, and similarly, if rule matching is unsuccessful, a reasoning result without reasoning new knowledge is obtained.
Although the embodiments of the present invention have been disclosed for illustrative purposes, those skilled in the art will appreciate that: various substitutions, alterations and modifications are possible without departing from the spirit and scope of this disclosure and appended claims, and accordingly, the scope of this disclosure is not limited to the embodiments disclosed.
Claims (6)
1. A method for realizing classification early warning of data link faults in data based on a knowledge graph is characterized by comprising the following steps: the method comprises the following specific steps:
(1) Establishing a Kmeans-SVM fault classification model
Firstly, using PCA method to carry out dimensionality reduction and dimensionality reduction processing on dataClustering the unlabeled substation data by using a K-means method and using the characteristics of unsupervised learning of the unlabeled substation data, wherein the number K of clusters is set to be 2, the distance standard used in the clustering process is Euclidean distance, the sum of squares of errors is used as a representative of the similarity of samples in the clusters, and one of the well-divided clusters G isuSmaller sum of squared errors in (c) indicates greater similarity of samples within the cluster, and conversely, GuThe larger the sum of squared errors, the smaller the similarity of the samples in the cluster, and the calculation formula of the sum of squared errors is as follows:
the input unlabelled data are changed into labeled data through a K-means method, the input data are finally divided into two types, the two types are respectively represented by numbers 1 and 2, an SVM model is trained by using the labeled data to obtain a maximum interval hyperplane, the two types of data 1 and 2 are respectively divided on two sides of the hyperplane, the accuracy of a classification result obtained by K-means is tested by using the maximum interval hyperplane after SVM training is finished, the correctly predicted data is used for retraining the hyperplane of the SVM, the hyperplane of the SVM is iteratively updated by the method until the error rate of the data predicted by using the SVM is not changed any more, and the final maximum interval hyperplane is obtained, and the specific algorithm steps are as follows:
the method comprises the following steps: preprocessing the non-tag data, dividing the preprocessed data into two clusters by using a K-means algorithm, respectively marking the two clusters as 1 and 2, and converting the non-tag data into tagged data;
step two: training an SVM classifier model by using the labeled data obtained in the step one to obtain a maximum interval hyperplane;
step three: testing K-means clustering by using the trained SVM classification model to obtain two sample data, taking the accurately predicted data as the input data of the second step again to obtain a new maximum interval hyperplane until the error rate of the data predicted by the SVM is not changed any more;
(2) Constructing fault domain knowledge graph
The method mainly comprises the following steps of constructing a knowledge graph in the link fault field, wherein the knowledge graph is mainly divided into four parts, namely entity and attribute extraction, coreference resolution, knowledge processing and data integration;
(3) Data link failure early warning
And carrying out knowledge reasoning according to the logical relationship between the fault node information and the fault reason in the link fault knowledge graph.
2. The extraction of entities and attributes includes: firstly, performing word segmentation operation on a corpus in a fault field, extracting entities and attributes by using a Markov model, taking the extracted entities and attributes as the entities and attributes finally applied to a knowledge graph, and then performing part-of-speech tagging on all words, wherein the words are divided into the following categories: the original dictionary comprises fault noun entities, fault verb, fault degree adverb, fault degree quantifier and unextracted words.
3. The method for realizing classification and early warning of faults of data link of data center station based on knowledge graph as claimed in claim 1, wherein: the common reference resolution step mainly aims to find out synonyms representing entities and attributes, classify words with high similarity into the same class, and use a form of a synonym table to realize representation of the synonyms.
4. The method for realizing classification and early warning of data link faults in data based on the knowledge graph as claimed in claim 1, wherein the method comprises the following steps: the main purpose of the knowledge processing step is to identify the corresponding relationship between the entity and the attribute, and delete the redundant inclusion relationship by taking the relationship whether the part of speech contains or does not contain as a standard.
5. The method for realizing classification and early warning of data link faults in data based on the knowledge graph as claimed in claim 1, wherein the method comprises the following steps: and in the data integration step, the triples of the fault entities, the attributes and the relations are combined, and the final map is constructed through the node updating of the concept layer and the entity attribute layer to finally form the knowledge map in the link fault field.
6. The method for realizing classification and early warning of data link faults in data based on the knowledge graph as claimed in claim 1, wherein the method comprises the following steps: the process of carrying out knowledge reasoning on the logical relationship between the fault node information and the fault reason in the link fault knowledge graph is as follows: firstly, extracting rules of stored known knowledge, matching the rules after extracting the rules, adding the rules to a rule execution area if the matching is successful, resolving conflicts if the rule conflicts are established, obtaining an unrealized reasoning result if the rule conflicts are not established, and similarly, obtaining a reasoning result without reasoning new knowledge if the rule matching is unsuccessful.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210884956.7A CN115269870A (en) | 2022-07-26 | 2022-07-26 | Method for realizing classification and early warning of data link faults in data based on knowledge graph |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210884956.7A CN115269870A (en) | 2022-07-26 | 2022-07-26 | Method for realizing classification and early warning of data link faults in data based on knowledge graph |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115269870A true CN115269870A (en) | 2022-11-01 |
Family
ID=83769530
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210884956.7A Pending CN115269870A (en) | 2022-07-26 | 2022-07-26 | Method for realizing classification and early warning of data link faults in data based on knowledge graph |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115269870A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116827817A (en) * | 2023-04-12 | 2023-09-29 | 国网河北省电力有限公司信息通信分公司 | Data link state monitoring method, device, monitoring system and storage medium |
| CN117647697A (en) * | 2023-11-21 | 2024-03-05 | 广东电网有限责任公司江门供电局 | Knowledge graph-based fault positioning method and system for electric power metering assembly line |
-
2022
- 2022-07-26 CN CN202210884956.7A patent/CN115269870A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116827817A (en) * | 2023-04-12 | 2023-09-29 | 国网河北省电力有限公司信息通信分公司 | Data link state monitoring method, device, monitoring system and storage medium |
| CN117647697A (en) * | 2023-11-21 | 2024-03-05 | 广东电网有限责任公司江门供电局 | Knowledge graph-based fault positioning method and system for electric power metering assembly line |
| CN117647697B (en) * | 2023-11-21 | 2024-05-14 | 广东电网有限责任公司江门供电局 | Knowledge graph-based fault positioning method and system for electric power metering assembly line |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yu et al. | Beyond Word Attention: Using Segment Attention in Neural Relation Extraction. | |
| WO2018218708A1 (en) | Deep-learning-based public opinion hotspot category classification method | |
| CN115269870A (en) | Method for realizing classification and early warning of data link faults in data based on knowledge graph | |
| CN113191148B (en) | Rail transit entity identification method based on semi-supervised learning and clustering | |
| CN112560432A (en) | Text emotion analysis method based on graph attention network | |
| CN113138920B (en) | Software defect report allocation method and device based on knowledge graph and semantic role labeling | |
| CN112487822A (en) | Cross-modal retrieval method based on deep learning | |
| US20230014904A1 (en) | Searchable data structure for electronic documents | |
| CN112417893A (en) | Software function demand classification method and system based on semantic hierarchical clustering | |
| CN116842194A (en) | Electric power semantic knowledge graph system and method | |
| CN115146062A (en) | Intelligent event analysis method and system fusing expert recommendation and text clustering | |
| CN116611071A (en) | Function-level vulnerability detection method based on multiple modes | |
| CN113392191B (en) | Text matching method and device based on multi-dimensional semantic joint learning | |
| CN114880307A (en) | Structured modeling method for knowledge in open education field | |
| CN113312907B (en) | Remote supervision relation extraction method and device based on hybrid neural network | |
| CN111339258B (en) | University computer basic exercise recommendation method based on knowledge graph | |
| CN112685374B (en) | Log classification method and device and electronic equipment | |
| CN110377690B (en) | Information acquisition method and system based on remote relationship extraction | |
| CN111160756A (en) | Scenic spot assessment method and model based on secondary artificial intelligence algorithm | |
| CN116680407A (en) | Knowledge graph construction method and device | |
| WO2023093116A1 (en) | Method and apparatus for determining industrial chain node of enterprise, and terminal and storage medium | |
| CN114757097B (en) | Line fault diagnosis method and device | |
| Hou et al. | Automatic Classification of Basic Nursing Teaching Resources Based on the Fusion of Multiple Neural Networks. | |
| Chang et al. | Code Transform Model Producing High-Performance Program. | |
| Zhao et al. | Test case classification via few-shot learning |
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 |