CN113988220A - Method for evaluating health state of coal mining machine - Google Patents
Method for evaluating health state of coal mining machine Download PDFInfo
- Publication number
- CN113988220A CN113988220A CN202111416994.1A CN202111416994A CN113988220A CN 113988220 A CN113988220 A CN 113988220A CN 202111416994 A CN202111416994 A CN 202111416994A CN 113988220 A CN113988220 A CN 113988220A
- Authority
- CN
- China
- Prior art keywords
- evaluation
- coal mining
- mining machine
- vector
- weight
- 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
- 239000003245 coal Substances 0.000 title claims abstract description 132
- 238000005065 mining Methods 0.000 title claims abstract description 126
- 238000000034 method Methods 0.000 title claims abstract description 50
- 230000036541 health Effects 0.000 title claims abstract description 44
- 238000011156 evaluation Methods 0.000 claims abstract description 135
- 239000013598 vector Substances 0.000 claims abstract description 103
- 238000012544 monitoring process Methods 0.000 claims abstract description 44
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 24
- 230000008569 process Effects 0.000 claims abstract description 14
- 230000003068 static effect Effects 0.000 claims abstract description 11
- 230000004927 fusion Effects 0.000 claims abstract description 8
- 230000006870 function Effects 0.000 claims description 30
- 230000002159 abnormal effect Effects 0.000 claims description 17
- 239000011159 matrix material Substances 0.000 claims description 13
- 238000001514 detection method Methods 0.000 claims description 12
- 230000003862 health status Effects 0.000 claims description 11
- 238000012549 training Methods 0.000 claims description 11
- 238000010276 construction Methods 0.000 claims description 6
- 230000002087 whitening effect Effects 0.000 claims description 6
- 238000010219 correlation analysis Methods 0.000 claims description 5
- 238000012216 screening Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000004044 response Effects 0.000 claims description 3
- 238000013473 artificial intelligence Methods 0.000 abstract description 4
- 238000012854 evaluation process Methods 0.000 abstract description 3
- 230000008859 change Effects 0.000 description 7
- 230000015556 catabolic process Effects 0.000 description 6
- 238000006731 degradation reaction Methods 0.000 description 6
- 238000013210 evaluation model Methods 0.000 description 5
- 230000006866 deterioration Effects 0.000 description 4
- 238000005520 cutting process Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000009826 distribution Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 206010063385 Intellectualisation Diseases 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000013145 classification model Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000007704 transition Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000021615 conjugation Effects 0.000 description 1
- 230000008602 contraction Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
- G06F18/2148—Generating training patterns; Bootstrap methods, e.g. bagging or boosting characterised by the process organisation or structure, e.g. boosting cascade
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/254—Fusion techniques of classification results, e.g. of results related to same input data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/02—Computing arrangements based on specific mathematical models using fuzzy logic
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Educational Administration (AREA)
- General Health & Medical Sciences (AREA)
- Development Economics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Fuzzy Systems (AREA)
- Mathematical Analysis (AREA)
- Automation & Control Theory (AREA)
- Biomedical Technology (AREA)
- Primary Health Care (AREA)
- Molecular Biology (AREA)
- Animal Husbandry (AREA)
- Algebra (AREA)
- Computational Mathematics (AREA)
- Agronomy & Crop Science (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
Abstract
The invention provides a health state evaluation method of a coal mining machine, which comprises the following steps: calculating comprehensive correlation coefficients among all monitoring parameters in the collected coal mining machine state data, and taking the monitoring parameters lower than a preset threshold value as indexes participating in evaluation; obtaining static weight vectors of all indexes according to an analytic hierarchy process and an entropy weight process, obtaining dynamic weight vectors through a variable weight formula, and obtaining an evaluation vector Q by combining gray clustering; inputting monitoring parameter data to be evaluated into a trained XGboost model to obtain an evaluation probability vector R of the state of a coal mining machine of the XGboost model; carrying out weighted fusion on the evaluation vector Q and the evaluation probability vector R to obtain a comprehensive evaluation vector QR; and obtaining the probability corresponding to each coal mining machine state according to the comprehensive evaluation vector QR. According to the method, the influence of working condition factors and operation data of the coal mining machine is considered in the evaluation process, the empirical model and the artificial intelligence model are combined, the use limit of the two algorithms is weakened, and the rationality of the evaluation result is enhanced.
Description
Technical Field
The invention relates to the technical field of equipment maintenance, in particular to a method for evaluating the health state of a coal mining machine.
Background
Coal mine intellectualization is a new stage of coal comprehensive mechanization development, wherein intelligent coal mining is a key step of coal mine intellectualization development. The coal mining machine is used as core equipment for coal mining, and the working reliability and safety of the coal mining machine are important preconditions for normal operation of fully mechanized coal mining, guarantee of coal production safety and improvement of economic benefits of coal enterprises. The working environment of the coal mining machine is complex and changeable, and the state monitoring is difficult and the faults are frequent due to severe factors such as heavy load operation, frequent starting, narrow operable space and the like. Therefore, the method for actively sensing the state of the coal mining machine, researching data processing of the coal mining machine, index analysis, health state evaluation and the like is closely combined with the coal mining working condition, is an important means for realizing tracking health management and service of the whole life cycle of the coal mining machine, and is a basic guarantee for realizing safe, efficient, green and intelligent coal mining.
The health state evaluation of the existing coal mining machine mainly has the following 4 problems: (1) the coal mining machine has high abnormal data detection difficulty and is not reasonable enough in evaluation index selection. (2) The evaluation algorithm based on experience, such as an analytic hierarchy process, fuzzy comprehensive evaluation and the like, is influenced by the structure and working condition of the coal mining machine, the evaluation process is inevitably affected by human intervention, and meanwhile, related researches, such as evaluation index construction, index weight distribution, health state grade classification and the like, are still insufficient. (3) Common artificial intelligence assessment methods such as a support vector machine have difficulty in solving the multi-classification problem, models such as deep learning and neural networks are easily limited by sample data, assessment models and results are difficult to reasonably explain, and the model generalization capability is poor. (4) The evaluation model established from the experience model or the artificial intelligence evaluation model alone is difficult to take into account the influence of the working condition factors and the operation data of the coal mining machine, and the evaluation result is difficult to accord with the actual working condition. The empirical model considers a large number of evaluation factors, but the deep research on the information provided by the coal mining machine operation data is not enough, and the intelligent evaluation model analyzes the deep meaning expressed by the coal mining machine operation data, but the influence of actual experience and the evaluation factors is difficult to consider.
Therefore, the invention provides a novel method for evaluating the health state of a coal mining machine.
Disclosure of Invention
In order to solve the problems, the invention provides a coal mining machine health state evaluation method integrating variable weight fuzzy comprehensive evaluation and XGboost aiming at the limitation of a single evaluation algorithm, wherein results of a VW-FCE method and a XGboost method are integrated, and a final result is obtained from a multi-model evaluation result. According to the method, the influence of working condition factors and operation data of the coal mining machine is considered in the evaluation process, the empirical model and the artificial intelligence model are combined, the use limit of the two algorithms is weakened, and the rationality of the evaluation result is enhanced.
In order to achieve the above purpose, the present invention provides the following technical solutions.
A health state evaluation method for a coal mining machine comprises the following steps:
calculating comprehensive correlation coefficients among all monitoring parameters in the collected coal mining machine state data, and taking the monitoring parameters lower than a preset threshold value as indexes participating in evaluation;
obtaining static weight vectors of all indexes according to an analytic hierarchy process and an entropy weight process, obtaining dynamic weight vectors by the static weight vectors through a variable weight formula, and obtaining an evaluation vector Q by combining the dynamic weight vectors with gray clustering;
inputting monitoring parameter data to be evaluated into a trained XGboost model, calculating leaf node values corresponding to each state of the coal mining machine, and obtaining an evaluation probability vector R of the state of the coal mining machine of the XGboost model;
carrying out weighted fusion on the evaluation vector Q and the evaluation probability vector R to obtain a comprehensive evaluation vector QR;
and obtaining the probability corresponding to each coal mining machine state according to the comprehensive evaluation vector QR, and obtaining a coal mining machine state result according to the maximum membership degree.
Preferably, the method further comprises the following steps: and carrying out abnormal data detection on the collected coal mining machine state data through an isolated forest algorithm, and removing the abnormal data.
Preferably, the calculating of the comprehensive correlation coefficient between the monitoring parameters in the collected coal mining machine state data includes the following steps:
performing correlation analysis on each monitoring parameter of the coal mining machine through a Pearson correlation coefficient, a Spireman correlation coefficient and a cosine distance respectively:
in the formula: rhop，ρs，ρcRespectively is a Pearson correlation coefficient, a spearman correlation coefficient and a cosine distance among parameters; x, Y are sample vectors, Cov (X, Y) is the covariance of X and Y, and the generations Var (X) and Var (Y)Variance of X and Y; a is the number of samples, N is the total number of samples, daRepresenting the position difference of the two variables which are respectively sequenced into pairs;
integrated correlation coefficient ρcoComprises the following steps:
ρco＝α1×ρs+α2×ρp+α3×ρc
in the formula, alpha1，α2，α3Is a weight and alpha1+α2+α 31 is ═ 1; of which the three correlation coefficients are of equal importance, i.e. alpha1＝α2＝α3＝1/3；
When the comprehensive correlation coefficient is greater than or equal to a threshold TH, the parameters are closely related, and when the comprehensive correlation coefficient is less than the threshold TH, the parameters do not have close correlation;
according to the correlation coefficient rhocoAnd screening out monitoring parameters with comprehensive correlation coefficients lower than TH according to the calculation result of (1), and finishing the construction of evaluation indexes.
Preferably, the obtaining the dynamic weight vector by the static weight vector through a variable weight formula includes the following steps:
determining the number m of indexes participating in evaluation;
respectively obtaining a jth index weight vector u according to an analytic hierarchy process and an entropy weight processj、vjCombining the two weight vectors to obtain a jth index combination weight as follows:
the combining weight vector is obtained as:
w＝(w1,w2,...,wm)
determining the weight-variable formula as follows:
in the formula:
is the weight-varying coefficient of the j-th index, wjTo combine the weight coefficients, xjThe j index operation parameter;
introducing an equalization function into the variable weight formula, wherein the equalization function is as follows:
substituting the equalization function into a variable weight formula, then:
in the formula: omegajThe weight-variable coefficient of the j index after the introduction of the equalization function; according to a variable weight formula introducing an equalization function, the final variable weight, namely a dynamic weight vector is obtained as follows:
ω＝[ω1,ω2,...,ωm]
wherein α is 0.5.
Preferably, the step of combining the dynamic weight vector with the gray cluster to obtain an evaluation vector Q includes the following steps:
the health state of the coal mining machine is evaluated through a central point type triangular whitening weight function, m indexes participating in evaluation are obtained, and the membership degree is as follows:
in the formula: s is the number of ash groups, lambdatIs the center of the t-th gray class, xjA monitoring value for a certain index;
obtaining xjThe conformity vector for s gray classes is:
calculating the conformity vector of the monitoring values of all indexes to s gray classes to obtain a conformity matrix of
F＝[F1,F2,...,Fm]
Constructing a conformity matrix F for the s ash classes, and obtaining an evaluation vector of each state of the coal mining machine according to the dynamic weight vector, wherein the evaluation vector comprises the following steps:
QT＝FωT。
preferably, the obtaining of the evaluation probability vector R of the state of the XGBoost model coal mining machine includes the following steps:
constructing sample data of monitoring parameters of the coal mining machine in each state, and training the sample data by adopting an XGboost model;
inputting monitoring parameter data to be evaluated into a trained XGboost model, and calculating leaf node values corresponding to each state of the coal mining machine;
adding the scores P of certain states obtained under each tree to obtain the total score S of each state corresponding to the sample to be evaluatedOComprises the following steps:
in the formula: k is the number of trees, O is the status level, P (O)kAnd (3) a scoring matrix S is provided for the scores corresponding to different states in each tree:
S＝[S1,S2,...,Ss]
obtaining sigmoid of each score, obtaining the evaluation probability corresponding to each state, and finally obtaining XGboost probability vector, namely evaluation probability vector R:
R＝[R1,R2,...,Rs]
in the formula: s is the health status grade of the coal mining machine, R1,R2,...,RsThe evaluation result under the sample is the probability of each state.
Preferably, the obtaining of the comprehensive assessment vector QR comprises the following steps:
and carrying out weighted fusion on the evaluation vector Q and the evaluation probability vector R to finally synthesize an evaluation vector QR:
QR＝[QR1,QR2,...,QRs]
wherein
QRO＝(β1QO+β2RO)(O＝1,2,...,s)
In the formula: QR (quick response) displayOFor the comprehensive evaluation of the probability, beta, of each state1,β2Respectively representing probability weights of the two algorithms; wherein, beta1,β2Take 0.5.
The invention has the beneficial effects that:
(1) the method adopts the isolated forest algorithm to complete the detection of the abnormal data of the coal mining machine, and can effectively solve the problem of high difficulty in the detection of the abnormal data of the coal mining machine.
(2) The method provides a method for calculating the comprehensive correlation coefficient of the monitoring parameters, completes the construction of the evaluation index of the coal mining machine, and can solve the problem that the evaluation index of the coal mining machine is not reasonably selected.
(3) The variable weight fuzzy comprehensive evaluation provided by the method can better explain the index weight distribution, the establishment of the algorithm model and the evaluation result, and the model has higher applicability when training data is lost under the combination of subjective factors and objective factors.
(4) Aiming at the influence of working condition factors and operation data of the coal mining machine, the variable weight fuzzy comprehensive evaluation method and the XGboost algorithm are combined, the evaluation vectors of the variable weight fuzzy comprehensive evaluation method and the XGboost algorithm are subjected to weighted fusion to obtain a comprehensive evaluation vector, and a final result is obtained from the multi-model evaluation results of the empirical model and the artificial intelligent model. The method weakens the use limit of the two algorithms mutually, and enhances the rationality of the evaluation result.
Drawings
FIG. 1 is a flowchart of evaluation based on a variable weight fuzzy comprehensive evaluation and XGboost method according to an embodiment of the present invention;
FIG. 2 is a diagram of the detection result of the abnormal data of the isolated forest according to the embodiment of the invention;
FIG. 3 is a correlation coefficient thermodynamic diagram of an embodiment of the invention;
FIG. 4 is a graph of index dynamic weight change according to an embodiment of the present invention;
FIG. 5 is a graph of accuracy of different parametric model evaluations according to an embodiment of the present invention;
FIG. 6 is an evaluation confusion matrix according to an embodiment of the present invention;
fig. 7 is a graph of the change of the health state of the shearer in the embodiment of the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Example 1
The invention provides a health state evaluation method of a coal mining machine, a flow chart of the method is shown in figure 1, and the method specifically comprises the following steps:
s1: and carrying out abnormal data detection on the collected coal mining machine state data through an isolated forest algorithm, and removing the abnormal data. The isolated forest algorithm is specially suitable for a high-dimensional data set, and has good detection precision on data with strong mutation and large uncertainty of global outliers.
S2: and calculating comprehensive correlation coefficients among all monitoring parameters in the collected coal mining machine state data, and taking the monitoring parameters lower than a preset threshold value as indexes participating in evaluation.
Specifically, the method comprises the following steps:
performing correlation analysis on each monitoring parameter of the coal mining machine through a Pearson correlation coefficient, a Spireman correlation coefficient and a cosine distance respectively:
in the formula: rhop，ρs，ρcRespectively is a Pearson correlation coefficient, a spearman correlation coefficient and a cosine distance among parameters; x, Y is a sample vector, Cov (X, Y) is the covariance of X and Y, Var (X) and Var (Y) are the variances of X and Y; a is the number of samples, N is the total number of samples, daRepresenting the position difference of the two variables which are respectively sequenced into pairs;
integrated correlation coefficient ρcoComprises the following steps:
ρco＝α1×ρs+α2×ρp+α3×ρc
in the formula, alpha1，α2，α3Is a weight and alpha1+α2+α 31 is ═ 1; of which the three correlation coefficients are of equal importance, i.e. alpha1＝α2＝α3＝1/3；
When the comprehensive correlation coefficient is greater than or equal to a threshold TH, the parameters are closely related, and when the comprehensive correlation coefficient is less than the threshold TH, the parameters do not have close correlation;
according to the correlation coefficient rhocoAnd screening out monitoring parameters with comprehensive correlation coefficients lower than TH according to the calculation result of (1), and finishing the construction of evaluation indexes.
S3: and obtaining static weight vectors of all indexes according to an analytic hierarchy process and an entropy weight process, obtaining dynamic weight vectors by the static weight vectors through a variable weight formula, and obtaining an evaluation vector Q by combining the dynamic weight vectors with gray clustering.
S3.1: obtaining a dynamic weight vector:
determining the number m of indexes participating in evaluation;
respectively obtaining a jth index weight vector u according to an analytic hierarchy process and an entropy weight processj、vjCombining the two weight vectors to obtain a jth index combination weight as follows:
the combining weight vector is obtained as:
w＝(w1,w2,...,wm)
determining the weight-variable formula as follows:
in the formula:
is the weight-varying coefficient of the j-th index, wjTo combine the weight coefficients, xjThe j index operation parameter;
introducing an equalization function into the variable weight formula, wherein the equalization function is as follows:
substituting the equalization function into a variable weight formula, then:
in the formula: omegajThe weight-variable coefficient of the j index after the introduction of the equalization function; according to a variable weight formula introducing an equalization function, the final variable weight, namely a dynamic weight vector is obtained as follows:
ω＝[ω1,ω2,...,ωm]
wherein α is 0.5.
S3.2: obtaining an evaluation vector Q:
the health state of the coal mining machine is evaluated through a central point type triangular whitening weight function, m indexes participating in evaluation are obtained, and the membership degree is as follows:
in the formula: s is the number of ash groups, lambdatIs the center of the t-th gray class, xjA monitoring value for a certain index;
obtaining xjThe conformity vector for s gray classes is:
calculating the conformity vector of the monitoring values of all indexes to s gray classes to obtain a conformity matrix of
F＝[F1,F2,...,Fm]
Constructing a conformity matrix F for the s ash classes, and obtaining an evaluation vector of each state of the coal mining machine according to the dynamic weight vector, wherein the evaluation vector comprises the following steps:
QT＝FωT。
s4: and inputting the monitoring parameter data to be evaluated into the trained XGboost model, calculating leaf node values corresponding to each state of the coal mining machine, and obtaining an evaluation probability vector R of the coal mining machine state of the XGboost model.
S4.1: constructing a prediction model:
the prediction model of XGboost is
In the formula: x is the number ofiIs a feature of the ith sample, fk(xi) For the predicted value of the kth tree,
is the predicted value of the model. The XGboost algorithm is optimized in sequence from the 1 st tree and finally is optimized until the K tree is completed. For a given xiThen, there is model initialization:
add K trees to the model:
the target function of the XGboost consists of a loss function for measuring training errors and a regularization term for controlling complexity, and the formula is as follows:
in the formula:
the loss function is used for measuring the error between the real value and the model predicted value, and the loss function commonly used in the classification model is Mean Square Error (MSE); and Ω is a regularization term used to control the complexity of the model to prevent overfitting of the model, Ω (f)k) Representing the complexity of the kth tree, c is a constant term.
And (3) expanding the loss function according to a Taylor series to obtain an approximate objective function, wherein the constant term c can be ignored.
In the formula: giAs a function of loss
First derivative of, hiTo represent
Wherein:
the complexity in the XGboost classification model is mainly determined by the number of leaf nodes and the smoothness degree of the weight of the corresponding nodes.
In the formula: γ and
are all manually set parameters, T is the number of leaf nodes, P is the weight of each leaf, i.e. the leaf node score,
is a regularization penalty term for the P weight parameters.
By means of IrRepresenting the set of samples in the r-th leaf, let:
obtaining the optimal leaf node weight
And an objective function obj*Comprises the following steps:
s4.2: obtaining an evaluation probability vector R:
constructing sample data of monitoring parameters of the coal mining machine in each state, and training the sample data by adopting an XGboost model;
inputting monitoring parameter data to be evaluated into a trained XGboost model, and calculating leaf node values corresponding to each state of the coal mining machine;
adding the scores P of certain states obtained under each tree to obtain the total score S of each state corresponding to the sample to be evaluatedOComprises the following steps:
in the formula: k is the number of trees, O is the status level, P (O)kAnd (3) a scoring matrix S is provided for the scores corresponding to different states in each tree:
S＝[S1,S2,...,Ss]
obtaining sigmoid of each score, obtaining the evaluation probability corresponding to each state, and finally obtaining XGboost probability vector, namely evaluation probability vector R:
R＝[R1,R2,...,Rs]
in the formula: s is the health status grade of the coal mining machine, R1,R2,...,RsThe evaluation result under the sample is the probability of each state.
S5: and carrying out weighted fusion on the evaluation vector Q and the evaluation probability vector R to obtain a comprehensive evaluation vector QR.
Specifically, the method comprises the following steps:
and carrying out weighted fusion on the evaluation vector Q and the evaluation probability vector R to finally synthesize an evaluation vector QR:
QR＝[QR1,QR2,...,QRs]
wherein
QRO＝(β1QO+β2RO)(O＝1,2,...,s)
In the formula: QR (quick response) displayOFor the comprehensive evaluation of the probability, beta, of each state1,β2Respectively representing probability weights of the two algorithms; wherein, beta1,β2Take 0.5.
S6: and obtaining the probability corresponding to each coal mining machine state according to the comprehensive evaluation vector QR, and obtaining a coal mining machine state result according to the maximum membership degree.
Example 1:
1.1 abnormal Point detection
2 main operation parameters of the coal mining machine, namely main frequency converter current and main frequency converter temperature measured under the working condition are selected as examples, and an isolated forest algorithm is adopted to detect abnormal values of the coal mining machine parameters. 2000 groups of measurement data of 2 parameters, wherein the abnormal data are 50 groups in total and mainly comprise abnormal signals of start and stop, sudden change and the like of the coal mining machine.
Determining algorithm parameters according to the requirement of the isolated forest algorithm, setting the number n _ samples to be 2000, and setting the number n _ estimators of the trees to be 100; the data set has an outlier score of outliers _ fraction of 0.025; the outlier proportion conjugation of the data set was 0.025. The results of the detection are shown in FIG. 2.
In the figure, the isolated forest divides normal points into high-density groups, abnormal points are scattered in the space outside the high-density area, and high detection accuracy is obtained by isolating the abnormal points. Compared with two density-based outlier elimination algorithms of DBSCAN clustering and Local Outlier Factor (LOF), the results are shown in Table 1.
TABLE 1 coal mining machine anomaly data detection algorithm comparison
As can be seen from Table 1, the isolated forest algorithm has higher identification precision and shorter identification time for the abnormal data of the coal mining machine.
1.2 evaluation index construction
Taking monitoring data of a production and operation coal mining machine under a complex geological condition as an example, the correlation of monitoring parameters of the coal mining machine is analyzed. And determining main monitoring parameters of the coal mining machine by combining actual working conditions as shown in the table 2.
TABLE 2 coal mining machine Main monitoring parameters
Selecting 3-month monitoring data of the coal mining machine, taking factors such as shutdown and emergency of the coal mining machine into consideration, and extracting data of 2000 time points capable of representing the running state of the coal mining machine to form a data set with dimensions of 18 multiplied by 2000. The comprehensive correlation coefficient is calculated, and a correlation coefficient thermodynamic diagram is obtained as shown in fig. 3.
As shown in FIG. 3, 4 parameters of the main frequency converter of the coal cutter are closely related to 4 parameters of the auxiliary frequency converter, the temperature of the traction motor is closely related to the temperature of the right cutting motor, and the temperature of the pump motor is closely related to the temperatures of the left cutting motor and the right cutting motor. And setting the threshold TH to be 0.9, screening out monitoring indexes with correlation coefficients lower than 0.9 according to the comprehensive correlation coefficients to obtain the final evaluation indexes of the coal mining machine, wherein the results are shown in Table 3.
TABLE 3 evaluation index selection for coal mining machine
2.1 evaluation method training and implementation
Sample data set establishment
And (3) carrying out correlation analysis on the monitoring parameters of the coal mining machine under the working condition to finally obtain 12-dimensional indexes to participate in health evaluation of the coal mining machine, wherein the results are shown in a table 3. Obtaining 2000 pieces of simulation data of the running state of the coal mining machine based on Weibull distribution according to the determined evaluation indexes and part of actual monitoring data, wherein the number of the training data is 1000, and the number of the healthy, good, general, deterioration and fault data is 300, 380, 250, 50 and 20 respectively; 1000 pieces of test data are 300, 380, 250, 50 and 20 pieces of health, good, general, deterioration and fault data respectively.
Model training and validation
(1) Variable weight fuzzy comprehensive judgment
According to an analytic hierarchy process, manually comparing every two 12-dimensional evaluation indexes (table 3) of the coal mining machine, scoring, determining a discrimination matrix, obtaining subjective weight according to the discrimination matrix, and completing matrix consistency inspection. According to the entropy weight method, objective weight is obtained through the information entropy of each index based on training data. The subjective weight and the objective weight are combined to obtain a combined weight, i.e., a static weight vector, as shown in table 4.
TABLE 4 coal mining machine index weight vector
Further, the change rule of the coal mining machine index weight coefficient along with the monitoring time is obtained, and the result is shown in fig. 4.
As shown in fig. 7, when the coal mining machine is working normally, the index weights are relatively stable, and as the health condition of the coal mining machine deteriorates, the weight of each index weight changes differently, and when the coal mining machine has a primary failure, some original secondary indexes (such as the temperature of the pump motor) occupy progressively larger weights, and become relatively important indexes. Therefore, the dynamic weight can more accurately reflect the real running state of each index of the coal mining machine than the static weight coefficient.
According to previous research and expert experience, the health state of the coal mining machine is divided into O {1, 2, 3, 4, 5} { healthy, good, general, deteriorated, and fault }, the range of the corresponding whitening weight function is determined, and the health state grade and the corresponding whitening weight function are shown in a table 5.
TABLE 5 health status level and corresponding whitening weight function
And evaluating the health state of the coal mining machine based on a variable weight fuzzy comprehensive evaluation method, wherein the result is shown in a table 6.
TABLE 6 fuzzy comprehensive evaluation and evaluation results
According to the result statistics, the health state of the coal mining machine can be evaluated by the method, and the overall evaluation accuracy is 92.4%.
(2) XGboost implementation
Based on the XGboost algorithm, processed data of the coal mining machine is input as characteristic quantity, and five health states of the coal mining machine are output as classification results. The XGBoost contraction step eta is set to 0.1, the random sampling ratio subsample is 0.2, the random column number ratio colosample _ byte is 1, the minimum loss function degradation value gamma is 0, and the classification category number num _ class is 5. The initial minimum leaf node weight and min _ child _ weight are set to 5. And selecting the depths max _ depth of the maximum tree as 1, 2, 4 and 6, setting the depths 1, 3, 5, 7 and 9 as min _ child _ weight values, and carrying out model training to obtain the relations between the parameters and the evaluation accuracy as a, b, c and d in the graph 5.
As can be seen from fig. 5, when the model max _ depth is 2, it is more stable and the evaluation accuracy is higher than when the model max _ depth is 1, 4, and 6; as shown in fig. 5(b), as num _ boost _ round increases, the model min _ child _ weight varies, and the estimation accuracy varies, and when min _ child _ weight increases, the model accuracy tends to increase first and then decrease or become gentle, and when min _ child _ weight is equal to 7 and num _ boost _ round is equal to 80, the model average accuracy is highest.
And (5) evaluating the health state of the coal mining machine by using the optimal parameter model, wherein the evaluation result is shown in a table 7. The evaluation results of 5 states of health, good, general, deterioration and failure are counted, as shown in fig. 6.
TABLE 7 XGboost evaluation results
As can be seen from fig. 6, the XGBoost-based coal mining machine state evaluation model still has high identification accuracy for data with unevenly distributed samples, and still has certain evaluation accuracy for the cases of degradation and insufficient faulty samples, and the overall evaluation accuracy is 97.8%.
(3) VW-FCE-XGboost evaluation model verification
And calculating the health state evaluation vector of the coal mining machine at each moment by using three methods of VW-FCE, XGboost and VW-FCE-XGboost to obtain a health state change curve of the coal mining machine in operation, wherein the health state change curve is shown in figure 7.
From fig. 7, the three methods can accurately evaluate the health status of the device, and it can be seen that the health status is difficult to evaluate when the two statuses alternate (circles in the figure). The VW-FCE method exhibits a good ambiguity in that the probability values of the two states are the same (at the intersection), but it is difficult to clearly determine the state at that time, and the reliability of the obtained result is not high. The XGboost makes a judgment in time during the period of alternation of the two states, but ignores the change of the health state during the transition period, and the judgment is too absolute. From VW-FCE-XGBoost, the comprehensive results of the two methods are used, the state at the moment can be accurately judged, the transition process of the health state can be expressed to a certain degree, and the obtained results are more interpretable.
The health state of the coal mining machine is evaluated by adopting VW-FCE-XGboost, the evaluation accuracy of the method on 5 health states of health, good, general, deterioration and fault of the coal mining machine is respectively 100%, 98.7%, 99.2%, 100% and 80%, and the overall evaluation accuracy of the model is 98.9%. Compared with two methods of variable weight fuzzy comprehensive evaluation and XGboost, the evaluation accuracy of the VW-FCE-XGboost model is improved by 6.5% and 1.1%.
Example 2:
taking an MG400/930-WD type coal mining machine for certain mine production operation as an example, the coal mining machine adopts a single cable for power supply, the power supply voltage is 3300V, the overall structure is multi-motor transverse arrangement, and the traction mode is airborne alternating-current variable-frequency speed regulation-pin track type chainless traction.
And selecting monitoring data of the coal mining machine in 3 months of operation, and performing outlier rejection and index correlation analysis on the experimental data based on the isolated forest and the comprehensive correlation coefficient to determine evaluation indexes as shown in the table 3. The coal mining machine has complex degradation trend, faults are mostly sudden-change faults and accidental faults, the evaluation result is difficult to meet the actual working condition due to the fact that the fault data are not defined obviously, and meanwhile the difference of the evaluation result is caused by the fact that the fault and the degradation standard are different. Processing and analyzing the coal mining machine data, combining with the point inspection record of the coal mining machine, and finally dividing the health state of the coal mining machine into three states of O (1, 2, 3) (health, other and fault), so as to obtain 1000 pieces of training and testing data, wherein the proportion of the health, other and fault data is 16: 3: 1. the health state evaluation results of the coal mining machine are obtained based on VW-FCE-XGboost and are shown in the table 8.
TABLE 8 evaluation results of health status of coal mining machine
As can be seen from Table 8, the accuracy of the evaluation method for the health state of the coal mining machine based on the VW-FCE-XGboost to the 3 states of the health state, the other state and the fault of the coal mining machine is 92.9%, 83.3% and 90.0% respectively. In the experiment, the characteristics of all health data are approximately the same; the fault data characteristics have diversity, but because the conditions are limited, the selected fault data are single; other data have high uncertainty including single index degradation, multi-index degradation, start-stop signal, standby signal and other data characteristics, so the accuracy of evaluation of the data is the worst. In conclusion, the experimental result meets the real situation and has certain accuracy.
The present invention is not limited to the above preferred embodiments, and any modifications, equivalent substitutions and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (7)
1. A health state evaluation method of a coal mining machine is characterized by comprising the following steps:
calculating comprehensive correlation coefficients among all monitoring parameters in the collected coal mining machine state data, and taking the monitoring parameters lower than a preset threshold value as indexes participating in evaluation;
obtaining static weight vectors of all indexes according to an analytic hierarchy process and an entropy weight process, obtaining dynamic weight vectors by the static weight vectors through a variable weight formula, and obtaining an evaluation vector Q by combining the dynamic weight vectors with gray clustering;
inputting monitoring parameter data to be evaluated into a trained XGboost model, calculating leaf node values corresponding to each state of the coal mining machine, and obtaining an evaluation probability vector R of the state of the coal mining machine of the XGboost model;
carrying out weighted fusion on the evaluation vector Q and the evaluation probability vector R to obtain a comprehensive evaluation vector QR;
and obtaining the probability corresponding to each coal mining machine state according to the comprehensive evaluation vector QR, and obtaining a coal mining machine state result according to the maximum membership degree.
2. The shearer health state evaluation method according to claim 1, further comprising: and carrying out abnormal data detection on the collected coal mining machine state data through an isolated forest algorithm, and removing the abnormal data.
3. The method for evaluating the health status of the coal mining machine according to claim 1, wherein the step of calculating the comprehensive correlation coefficient among the monitoring parameters in the collected coal mining machine state data comprises the following steps:
performing correlation analysis on each monitoring parameter of the coal mining machine through a Pearson correlation coefficient, a Spireman correlation coefficient and a cosine distance respectively:
in the formula: rhop，ρs，ρcRespectively is a Pearson correlation coefficient, a spearman correlation coefficient and a cosine distance among parameters; x, Y is a sample vector, Cov (X, Y) is the covariance of X and Y, Var (X) and Var (Y) are the variances of X and Y; a is the number of samples, N is the total number of samples, daRepresenting the position difference of the two variables which are respectively sequenced into pairs;
integrated correlation coefficient ρcoComprises the following steps:
ρco＝α1×ρs+α2×ρp+α3×ρc
in the formula, alpha1，α2，α3Is a weight and alpha1+α2+α31 is ═ 1; of which the three correlation coefficients are of equal importance, i.e. alpha1＝α2＝α3＝1/3；
When the comprehensive correlation coefficient is greater than or equal to a threshold TH, the parameters are closely related, and when the comprehensive correlation coefficient is less than the threshold TH, the parameters do not have close correlation;
according to the correlation coefficient rhocoAnd screening out monitoring parameters with comprehensive correlation coefficients lower than TH according to the calculation result of (1), and finishing the construction of evaluation indexes.
4. The method for evaluating the health status of the coal mining machine according to claim 1, wherein the step of obtaining the dynamic weight vector by the static weight vector through a variable weight formula comprises the following steps:
determining the number m of indexes participating in evaluation;
respectively obtaining a jth index weight vector u according to an analytic hierarchy process and an entropy weight processj、vjCombining the two weight vectors to obtain a jth index combination weight as follows:
the combining weight vector is obtained as:
w＝(w1,w2,...,wm)
determining the weight-variable formula as follows:
in the formula:
is the weight-varying coefficient of the j-th index, wjTo combine the weight coefficients, xjThe j index operation parameter;
introducing an equalization function into the variable weight formula, wherein the equalization function is as follows:
substituting the equalization function into a variable weight formula, then:
in the formula: omegajThe weight-variable coefficient of the j index after the introduction of the equalization function; according to a variable weight formula introducing an equalization function, the final variable weight, namely a dynamic weight vector is obtained as follows:
ω＝[ω1,ω2,...,ωm]
wherein α is 0.5.
5. The method for assessing the health status of a coal mining machine according to claim 4, wherein the step of combining the dynamic weight vector with the gray cluster to obtain an assessment vector Q comprises the following steps:
the health state of the coal mining machine is evaluated through a central point type triangular whitening weight function, m indexes participating in evaluation are obtained, and the membership degree is as follows:
in the formula: s is the number of ash groups, lambdatIs the center of the t-th gray class, xjA monitoring value for a certain index;
obtaining xjThe conformity vector for s gray classes is:
calculating the conformity vector of the monitoring values of all indexes to s gray classes to obtain a conformity matrix of
F＝[F1,F2,...,Fm]
Constructing a conformity matrix F for the s ash classes, and obtaining an evaluation vector of each state of the coal mining machine according to the dynamic weight vector, wherein the evaluation vector comprises the following steps:
QT＝FωT。
6. the method for evaluating the health status of the coal mining machine according to claim 5, wherein the obtaining of the evaluation probability vector R of the XGboost model coal mining machine state comprises the following steps:
constructing sample data of monitoring parameters of the coal mining machine in each state, and training the sample data by adopting an XGboost model;
inputting monitoring parameter data to be evaluated into a trained XGboost model, and calculating leaf node values corresponding to each state of the coal mining machine;
adding the scores P of certain states obtained under each tree to obtain the total score S of each state corresponding to the sample to be evaluatedOComprises the following steps:
in the formula: k is the number of trees, O is the status level, P (O)kFor each tree to be differentAnd (3) the score corresponding to the state is provided with a score matrix S:
S＝[S1,S2,...,Ss]
obtaining sigmoid of each score, obtaining the evaluation probability corresponding to each state, and finally obtaining XGboost probability vector, namely evaluation probability vector R:
R＝[R1,R2,...,Rs]
in the formula: s is the health status grade of the coal mining machine, R1,R2,...,RsThe evaluation result under the sample is the probability of each state.
7. The shearer health state evaluation method according to claim 6, wherein the obtaining of the comprehensive evaluation vector QR comprises the following steps:
and carrying out weighted fusion on the evaluation vector Q and the evaluation probability vector R to finally synthesize an evaluation vector QR:
QR＝[QR1,QR2,...,QRs]
wherein
QRO＝(β1QO+β2RO)(O＝1,2,...,s)
In the formula: QR (quick response) displayOFor the comprehensive evaluation of the probability, beta, of each state1,β2Respectively representing probability weights of the two algorithms; wherein, beta1,β2Take 0.5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111416994.1A CN113988220A (en) | 2021-11-25 | 2021-11-25 | Method for evaluating health state of coal mining machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111416994.1A CN113988220A (en) | 2021-11-25 | 2021-11-25 | Method for evaluating health state of coal mining machine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113988220A true CN113988220A (en) | 2022-01-28 |
Family
ID=79731943
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111416994.1A Pending CN113988220A (en) | 2021-11-25 | 2021-11-25 | Method for evaluating health state of coal mining machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113988220A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114548739A (en) * | 2022-02-15 | 2022-05-27 | 国网河南省电力公司电力科学研究院 | Transformer substation flood prevention risk combination evaluation method |
| CN114595581A (en) * | 2022-03-11 | 2022-06-07 | 重庆地质矿产研究院 | Regional geological disaster danger probability model based on influence factor dynamic weight |
-
2021
- 2021-11-25 CN CN202111416994.1A patent/CN113988220A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114548739A (en) * | 2022-02-15 | 2022-05-27 | 国网河南省电力公司电力科学研究院 | Transformer substation flood prevention risk combination evaluation method |
| CN114595581A (en) * | 2022-03-11 | 2022-06-07 | 重庆地质矿产研究院 | Regional geological disaster danger probability model based on influence factor dynamic weight |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111222549B (en) | Unmanned aerial vehicle fault prediction method based on deep neural network | |
| CN108960303B (en) | Unmanned aerial vehicle flight data anomaly detection method based on LSTM | |
| CN116757534B (en) | Intelligent refrigerator reliability analysis method based on neural training network | |
| CN110807257A (en) | Method for predicting residual life of aircraft engine | |
| CN111062508B (en) | Method for evaluating real-time running state of wind turbine generator based on big data technology | |
| CN107949812A (en) | For detecting the abnormal combined method in water distribution system | |
| CN113988220A (en) | Method for evaluating health state of coal mining machine | |
| CN110852441B (en) | Fire disaster early warning method based on improved naive Bayes algorithm | |
| CN111079348B (en) | Method and device for detecting slowly-varying signal | |
| CN114818313A (en) | Hidden electricity stealing behavior identification method based on synthesized minority oversampling technology | |
| CN114819102A (en) | GRU-based air conditioning equipment fault diagnosis method | |
| CN112149884A (en) | Academic early warning monitoring method for large-scale students | |
| CN113782113B (en) | Method for identifying gas fault in transformer oil based on deep residual error network | |
| CN117312972A (en) | Method for identifying health state of scraper conveyor speed reducer | |
| CN114460481A (en) | Energy storage battery thermal runaway early warning method based on Bi-LSTM and attention mechanism | |
| CN113742983A (en) | Long-span structural damage identification method based on depth self-encoder neural network | |
| CN117556347A (en) | Power equipment fault prediction and health management method based on industrial big data | |
| CN117232809A (en) | Fan main shaft fault pre-diagnosis method based on DEMATEL-ANP-CRITIC combined weighting | |
| Kim et al. | Anomaly pattern detection in streaming data based on the transformation to multiple binary-valued data streams | |
| CN110889186A (en) | Consistency inspection method for degradation data of accelerated storage and natural storage based on equal degradation increment time interval gray correlation analysis | |
| CN115545342A (en) | Risk prediction method and system for enterprise electric charge recovery | |
| CN115423370A (en) | Relay protection equipment health state assessment method and device | |
| Li et al. | Social network change detection using a genetic algorithm based back propagation neural network model | |
| CN114046816A (en) | Sensor signal fault diagnosis method based on lightweight gradient lifting decision tree | |
| CN113849479A (en) | Comprehensive energy supply station oil tank leakage detection method based on instant learning and self-adaptive threshold |
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 |