CN112668749B - Coal mine gas early warning method based on class mark weighting extreme learning machine - Google Patents
Coal mine gas early warning method based on class mark weighting extreme learning machine Download PDFInfo
- Publication number
- CN112668749B CN112668749B CN202011325639.9A CN202011325639A CN112668749B CN 112668749 B CN112668749 B CN 112668749B CN 202011325639 A CN202011325639 A CN 202011325639A CN 112668749 B CN112668749 B CN 112668749B
- Authority
- CN
- China
- Prior art keywords
- data
- gas concentration
- learning machine
- early warning
- class
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 37
- 239000003245 coal Substances 0.000 title claims abstract description 26
- 238000012544 monitoring process Methods 0.000 claims abstract description 49
- 239000011159 matrix material Substances 0.000 claims abstract description 42
- 238000012549 training Methods 0.000 claims abstract description 18
- 230000006870 function Effects 0.000 claims abstract description 16
- 230000004913 activation Effects 0.000 claims abstract description 12
- 238000004519 manufacturing process Methods 0.000 claims abstract description 9
- 238000002790 cross-validation Methods 0.000 claims abstract description 6
- 238000001514 detection method Methods 0.000 claims abstract description 5
- 238000007781 pre-processing Methods 0.000 claims abstract description 3
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 2
- 230000008901 benefit Effects 0.000 abstract description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 230000003472 neutralizing effect Effects 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a coal mine gas early warning method based on a class mark weighted extreme learning machine, which comprises the steps of acquiring historical gas concentration monitoring data, preprocessing the data, filling missing data, generating a sample set, setting a gas concentration early warning threshold value, determining a sample unbalance ratio by adopting the early warning threshold value mu, carrying out weighting operation on a class mark matrix of a sample according to a feedback value of the unbalance ratio, determining the optimal hidden layer node number of the extreme learning machine by utilizing five-fold cross validation, taking a Sigmoid function as an activation function, taking the weighted class mark matrix as an expected output matrix of a model, and training and generating the extreme learning machine model. The invention predicts whether the gas concentration of a single monitoring point exceeds the warning line for 3 hours in the future, and pre-warns the potential production safety risk, considers the influence of class unbalance, and has the advantages of simple realization, high detection rate and low false alarm rate.
Description
Technical Field
The invention relates to a coal mine gas early warning method, and belongs to the technical field of gas concentration prediction.
Background
In the energy industry of China, coal accounts for about 70% of primary energy production and consumption structures of China, and in a relatively long period in the future, coal is still the main energy source of China, but at present, the coal mine production safety situation is very serious, safety accidents still occur, and the monitoring systems of many coal mines in China lack disaster prediction functions. Mine gas is one of the main factors affecting the safety of the coal mine, and a series of safety accidents can be caused by the fact that the gas concentration is too high. Therefore, whether the gas concentration in the future hours exceeds the early warning line is accurately predicted, and potential safety hazards are timely found, so that the method has great significance for safety production of coal mines. However, in the coal mine gas monitoring data, the early warning data are often far less than the non-early warning data, so that the problem of unbalanced category is caused, the detection rate of the traditional prediction model is greatly reduced, the false alarm rate is increased, and the actual use effect is affected.
Disclosure of Invention
The invention aims to: in order to overcome the defects in the prior art, the invention provides the coal mine gas early warning method based on the class mark weighting extreme learning machine, which can effectively solve the problem of class unbalance, accurately predict whether the gas concentration exceeds an early warning line within a few hours in the future according to the historical data of gas, is mainly applied to coal mine safety early warning, and can improve the pre-control capability of coal mine on gas explosion accidents.
The technical scheme is as follows: in order to achieve the above purpose, the invention adopts the following technical scheme:
a coal mine gas early warning method based on class mark weighting extreme learning machine comprises the following steps:
step 1, acquiring historical gas concentration monitoring data of a monitoring point under a mine. A monitoring point is usually arranged under the mine, and gas concentration data are extracted and transmitted in real time.
Step 2, data preprocessing and missing data filling: and calculating a gas concentration mean value every ten minutes at intervals of one monitoring point to obtain regular gas concentration monitoring data, performing linear filling processing on the null value by using the calculated mean value, if the data is continuously lost for more than two hours, not filling, discarding corresponding time data points, and continuously calculating new data points at the occurrence position of new detection data to obtain a regular gas concentration monitoring time sequence.
Step 3, generating a sample set: taking the gas concentration data of 24 continuous hours as a characteristic X, taking the gas concentration data of 3 hours afterwards as a prediction target Y, and generating a training sample set phi one by one in a single-point sequential moving mode.
And 4, setting a gas concentration early-warning threshold mu, setting the gas concentration early-warning threshold mu according to specific requirements, and recognizing that safety production risks occur when the gas concentration exceeds the early-warning threshold mu.
And 5, class mark weighting, namely determining the unbalance ratio of the sample by adopting an early warning threshold mu, and carrying out weighting operation on a class mark matrix of the sample according to the feedback value of the unbalance ratio.
Step 51, scanning the predicted target Y in the sample set phi one by one 1 ,Y 2 ,...,Y N The number of the components larger than or equal to the early warning threshold mu and the number of the components smaller than the early warning threshold mu are respectively recorded as N - And N + And calculates the ratio λ=n - /N + 。
And step 52, performing preliminary assignment on the target output matrix T according to the scanning result of the step 51.
Wherein if Y ij ≥μ,t ij =1, otherwise, t ij -1, i represents the position of the training sample, j represents the predicted point in time, ti j represents the label; .
Step 53, scanning the initial target output matrix T and weighting it, if T ij =1, then it is rewritten asIf t ij If the number is = -1, the operation is not performed, and a class label weighting matrix T is obtained W Lambda represents the unbalance ratio.
And 6, model training, namely determining the optimal hidden layer node number of the extreme learning machine by utilizing five-fold cross validation, taking a Sigmoid function as an activation function, taking a weighted class mark matrix as an expected output matrix of the model, and training and generating the model of the extreme learning machine.
Step 61, setting the number L of hidden layer nodes in the extreme learning machine, and randomly generating the weight and bias sequence from input layer to hidden layer, a= [ a ] 1 ,a 2 ,...,a i2 ,...,a L ]And b= [ b ] 1 ,b 2 ,...,b i2 ,...,b L ]Wherein a is i2 ∈[-1,1],b i2 ∈[-1,1]Converting the training sample set phi into a hidden layer output matrix by using a Sigmoid activation function:
wherein H (a) 1 ,...,a L ,b 1 ,...,b L ,X 1 ,...,X S ) The representation represents a hidden output matrix, G is a Sigmoid activation function, and is calculated as follows:
wherein G (a) i2 ,b i2 ,X j2 ) Representing an activation function.
Step 62, combining hidden layer output matrix H and class label weighting matrix T W Solving the following formula:
Hβ=T w
wherein,,
using the least squares method, the calculation results are as follows:
wherein beta represents a hidden layer weight matrix in an ideal state,indicating meterThe hidden layer weight matrix obtained by calculation, +.>Weight value representing the L-th hidden layer node,,, and>the Moore-Penrose generalized inverse of the hidden layer output matrix results in a least-squares solution of the least-squares of β.
Step 7, data prediction: and predicting whether the future gas concentration exceeds a warning line by using the trained class-weighted extreme learning machine model, and determining whether early warning needs to be given according to a prediction result.
Acquiring gas concentration monitoring data of the past 24 hours in real time, performing normalization processing, marking as XX, and obtaining a= [ a ] through storage 1 ,a 2 ,...,a L ]And b= [ b ] 1 ,b 2 ,...,b L ]And converting X into H, combining the saved hidden layer weight matrix beta, calculating H multiplied by beta to obtain a corresponding future three-hour early warning prediction result Y, checking whether the value larger than 0 exists in Y one by one, and if so, sending out a warning signal.
Preferably: for one monitoring point, the step 1 should include at least 30 days of historical gas concentration monitoring data.
Preferably: the regular gas concentration monitoring time sequence method is obtained in the step 2:
step 21, reading historical gas concentration monitoring data, extracting all data within 10 minutes from the first data, and performing average calculation to generate regular monitoring gas concentration data x 1 And similarly, calculating every 10 minutes to finally obtain regular gas concentration monitoring time series data x 1 ,x 2 ,...,x K Where K is the total number of regular monitoring data.
Step 22, if the gas concentration monitoring time series data x 1 ,x 2 ,...,x K X in the middle t And x t′ And (3) continuously missing data exist between the points, and at the same time, t' -t is less than or equal to 12, filling the missing values by adopting the following formula.
Wherein t, t' ∈ {1, 2., t and x t′ The number of data continuously missing between points, i, represents the missing value position of the current missing sequence. If x t And x t′ And if the number of the data continuously missing between the points is more than 12, setting a breakpoint, and not compensating the missing data.
Preferably: the method for obtaining the sample set phi in the step 3 comprises the following steps: time series x of monitoring data for regular gas concentration 1 , x 2 ,...,x K First, gas concentration data for 24 hours is taken as characteristic X from a starting position 1 =[x 1 ,x 2 ,...,x 144 ]Gas concentration data of 3 hours thereafter was used as the corresponding prediction target Y 1 =[x 145 ,x 146 ,...,x 162 ]A first sample is generated and then sample 2 is generated by means of single point movement, i.e. X 2 =[x 2 ,x 3 ,...,x 145 ],Y 2 =[x 146 ,x 147 ,...,x 163 ]And by analogy, continuously generating samples until the data sequence is exhausted, and obtaining a sample set phi:
where N represents the number of samples generated.
Preferably: step 3, if the time sequence x of the regular gas concentration monitoring data is 1 ,x 2 ,...,x K If a breakpoint is encountered, extraction of the new sample resumes after the breakpoint.
Preferably: and 4, adopting iterative solution to the early warning threshold mu in the step, and setting an initial value according to the coal mine safety production standard and the actual requirement of coal mine safety production.
Preferably: the hidden layer node number L set in the step 61 is obtained by a five-fold cross validation method, that is, preset l= {10,50,100,200}, by dividing the sample set Φ into five parts randomly and averagely, training the marker weighted extreme learning machine model for each preset L value one by using four parts as training sets each time, validating on the remaining part, calculating the root mean square prediction precision of each L parameter after each sample is used as a test set, selecting the L value with the largest precision as the optimal hidden layer node parameter value, and feeding back to the step 61, wherein the square root error can ensure that convergence is steady.
Preferably: the trained model in the step S7 requires that the gas concentration monitoring data of the same monitoring point for the latest 24 hours are read in real time when the model is used.
Compared with the prior art, the invention has the following beneficial effects:
1. the invention adopts the class mark weighting extreme learning machine to realize the prediction of whether the gas concentration of a single monitoring point exceeds the guard line for 3 hours in the future and the early warning of the potential production safety risk, considers the influence of class unbalance, and has simple realization, high detection rate and low false alarm rate.
2. The benefits of the process of computing the class weights and weighting the class weight output matrix and the process of solving for beta at 35 are described.
The step 52 above the process has the advantage of fully considering the problem of unbalanced categories in the data, giving a larger expected output value to the category label with a high concentration gas value, being capable of enlarging the error range, otherwise giving a smaller expected output value to the category label with a low concentration gas value, achieving the prediction effect and neutralizing the output of the two.
Drawings
Fig. 1 is a flow chart of the present invention.
Fig. 2 is a flow chart of the method of the present invention in detail.
Fig. 3 is a flow chart of the method of the present invention running on-line.
Detailed Description
The present invention is further illustrated in the accompanying drawings and detailed description which are to be understood as being merely illustrative of the invention and not limiting of its scope, and various equivalent modifications to the invention will fall within the scope of the appended claims to the skilled person after reading the invention.
A coal mine gas early warning method based on class mark weighting extreme learning machine, as shown in figures 1-3, comprises the following steps:
A. and C, extracting gas concentration monitoring historical data of a certain coal mine monitoring point collected by a sensor, calculating an average value at equal intervals of 10 minutes to obtain normalized gas concentration monitoring data, analyzing whether the monitoring data are missing, and entering the step B if the missing data are missing.
B. And (C) filling the missing data by using a linear filling compensation method, filling if the continuous missing data is less than or equal to 12, and not filling if the continuous missing data exceeds 12 continuous missing values, setting a breakpoint, and then entering the step C.
C. Taking continuous 24-hour gas concentration data as a characteristic X from a normalized data time sequence head, taking the gas concentration data of 3 hours thereafter as a prediction target Y, sequentially moving one by one according to a single point to generate training samples, and restarting the extraction of new samples after break points if break points are encountered in the normalized gas concentration monitoring data time sequence, so as to finally obtain a data set:
where N represents the number of samples generated.
D. Setting a gas concentration early warning threshold mu, counting the number of predicted concentration values smaller than or larger than mu in all sample prediction targets Y, and respectively marking as N - And N + And calculates the ratio λ=n - /N + 。
E. And generating an initial target output matrix T corresponding to the extreme learning machine according to the predicted targets Y of all the samples, if the corresponding value in Y is larger than a threshold value mu, assigning 1 to the corresponding position in T, and otherwise assigning-1.
F. Scanning the initial target output matrix T, weighting, and if the value is 1, rewriting it asIf the value is-1, the operation is not performed, and a class mark weighting matrix T is obtained W 。
G. And training a class label weighting extreme learning machine model.
The specific steps of the step G are as follows:
and setting the optimal hidden layer node number L in the extreme learning machine by a five-fold cross validation method.
Randomly generating a weight and bias sequence between an input layer and a hidden layer, wherein a= [ a ] 1 ,a 2 ,...,a i2 ,...,a L ]And b= [ b ] 1 ,b 2 ,...,b i2 ,..., b L ]Wherein a is i2 ∈[-1,1],b i2 ∈[-1,1]Converting the training sample set phi into a hidden layer output matrix by using a Sigmoid activation function:
wherein H (a) 1 ,...,a L ,b 1 ,...,b L ,X 1 ,...,X S ) The hidden layer output matrix is represented, G is a Sigmoid activation function, and the calculation mode is as follows:
wherein G (a) i2 ,b i2 ,X j2 ) Representing an activation function.
Combining hidden layer output matrix H and class mark weighting matrix T W Solving the following formula:
Hβ=T w (4)
wherein,,
using the least squares method, the calculation results are as follows:
wherein beta represents a hidden layer weight matrix in an ideal state,representing the calculated hidden layer weight matrix, +.>Weight value representing the L-th hidden layer node, < ->Moore-Penrose generalized inverse of the hidden layer output matrix. From its definition, the result is a least-norm least-squares solution of β.
H. Arranging the model into a system, acquiring gas concentration monitoring data of the last 24 hours in real time, carrying out normalization treatment, and marking as X', wherein the gas concentration monitoring data is obtained by storing a= [ a ] 1 ,a 2 ,...,a L ]And b= [ b ] 1 ,b 2 ,...,b L ]And (3) converting X 'into H', combining the saved hidden layer weight matrix beta, calculating H 'X beta to obtain a corresponding future three-hour early warning prediction result Y', checking whether the value larger than 0 exists in Y 'one by one, if yes, sending out an alarm signal, and if not, returning to H' to run again after 10 minutes.
The foregoing is only a preferred embodiment of the invention, it being noted that: it will be apparent to those skilled in the art that various modifications and adaptations can be made without departing from the principles of the present invention, and such modifications and adaptations are intended to be comprehended within the scope of the invention.
Claims (7)
1. The coal mine gas early warning method based on class mark weighting extreme learning machine is characterized by comprising the following steps:
step 1, aiming at a monitoring point under an ore pit, acquiring historical gas concentration monitoring data of the monitoring point;
step 2, data preprocessing and missing data filling: calculating a gas concentration mean value every ten minutes for one monitoring point to obtain regular gas concentration monitoring data, performing linear filling processing on null values by using the calculated mean value, if the data is continuously missing for more than two hours, not filling, discarding corresponding time data points, and continuously calculating new data points at the occurrence position of new detection data to obtain a regular gas concentration monitoring time sequence;
step 3, generating a sample set: taking continuous 24-hour gas concentration data as a characteristic X, taking the gas concentration data of 3 hours thereafter as a prediction target Y, and generating a training sample set phi one by one in a single-point sequential moving mode;
step 4, setting a gas concentration early-warning threshold mu, setting the gas concentration early-warning threshold mu according to specific requirements, and recognizing that safety production risks occur when the gas concentration exceeds the early-warning threshold mu;
step 5, class mark weighting, namely determining the unbalance ratio of the sample by adopting an early warning threshold mu, and carrying out weighting operation on a class mark matrix of the sample according to the feedback value of the unbalance ratio;
step 51, scanning the predicted target Y in the sample set phi one by one 1 ,Y 2 ,...,Y N The number of the components larger than or equal to the early warning threshold mu and the number of the components smaller than the early warning threshold mu are respectively recorded as N - And N + And calculates the ratio λ=n - /N + N represents the number of generated samples;
step 52, performing preliminary assignment on the target output matrix T according to the scanning result in step 51;
wherein if Y i ≥μ,t ij =1, otherwise, t ij = -1, i denotes the position of the training sample, j denotes the predicted point in time, t ij Representing the label;
step 53, scanning the initial target output matrix T and weighting it, if T ij =1, then it is rewritten asIf t ij If the number is = -1, the operation is not performed, and a class label weighting matrix T is obtained W λ represents an imbalance ratio;
step 6, model training, namely determining the optimal hidden layer node number of the extreme learning machine by utilizing five-fold cross validation, taking a Sigmoid function as an activation function, taking a class mark weighting matrix as an expected output matrix of the model, training and generating a class mark weighting extreme learning machine model;
step 61, setting the number L of hidden layer nodes in the class label weighting extreme learning machine model, and randomly generating weights and bias sequences from input layer to hidden layer, wherein a= [ a ] 1 ,a 2 ,...,a i2 ,...,a L ]And b= [ b ] 1 ,b 2 ,...,b i2 ,...,b L ]Wherein a is i2 ∈[-1,1],b i2 ∈[-1,1]Converting the training sample set phi into a hidden layer output matrix by using a Sigmoid activation function:
wherein H (a) 1 ,...,a L ,b 1 ,...,b L ,X 1 ,...,X N ) The hidden layer output matrix is represented, G is a Sigmoid activation function, and the calculation mode is as follows:
wherein G (a) i2 ,b i2 ,X j2 ) Representing an activation function;
step 62, combining the hidden layer output matrix and the class label weighting matrix T W Solving the following formula:
Hβ=T w
wherein,,
using the least squares method, the calculation results are as follows:
wherein beta represents a hidden layer weight matrix in an ideal state,representing the calculated hidden layer weight matrix, < ->Weight value representing the L-th hidden layer node, < ->The Moore-Penrose generalized inverse of the hidden layer output matrix is obtained as a result of the least-norm least-squares solution of beta;
step 7, data prediction: and predicting whether the future gas concentration exceeds a warning line by using the trained class-weighted extreme learning machine model, and determining whether early warning needs to be given according to a prediction result.
2. The coal mine gas early warning method based on class-weighted extreme learning machine according to claim 1, wherein the method is characterized by comprising the following steps of: for one monitoring point, the step 1 should include at least 30 days of historical gas concentration monitoring data.
3. The coal mine gas early warning method based on class-weighted extreme learning machine according to claim 2, wherein the method is characterized by comprising the following steps of: the regular gas concentration monitoring time sequence method is obtained in the step 2:
step 21, reading historical gas concentration monitoring data, extracting all data within 10 minutes from the first data, and performing average calculation to generate regular monitoring gas concentration data x 1 And similarly, calculating every 10 minutes to finally obtain regular gas concentration monitoring time series data x 1 ,x 2 ,...,x K Wherein K is the total number of regular monitoring data;
step 22, if the gas concentration monitoring time series data x 1 ,x 2 ,...,x K X in the middle t And x t' Continuous missing data exist between points, and at the same time, t' -t is less than or equal to 12, the missing values are filled by adopting the following formula;
wherein t, t' ∈ {1, 2., t and x t' The number of data continuously missing between points, i represents the missing value position of the current missing sequence; if x t And x t' And if the number of the data continuously missing between the points is more than 12, setting a breakpoint, and not compensating the missing data.
4. The coal mine gas early warning method based on class-weighted extreme learning machine according to claim 3, wherein the method comprises the following steps of: the method for obtaining the sample set phi in the step 3 comprises the following steps: time series x of monitoring data for regular gas concentration 1 ,x 2 ,...,x K First, gas concentration data for 24 hours is taken as characteristic X from a starting position 1 =[x 1 ,x 2 ,...,x 144 ]Gas concentration data of 3 hours thereafter was used as the corresponding prediction target Y 1 =[x 145 ,x 146 ,...,x 162 ]A first sample is generated and then sample 2 is generated by means of single point movement, i.e. X 2 =[x 2 ,x 3 ,...,x 145 ],Y 2 =[x 146 ,x 147 ,...,x 163 ]And by analogy, continuously generating samples until the data sequence is exhausted, and obtaining a sample set phi:
where N represents the number of samples generated.
5. The coal mine gas early warning method based on class-weighted extreme learning machine according to claim 4, wherein the method is characterized by comprising the following steps of: step 3, if the time sequence x of the regular gas concentration monitoring data is 1 ,x 2 ,...,x K If a breakpoint is encountered, extraction of the new sample resumes after the breakpoint.
6. The coal mine gas early warning method based on class-weighted extreme learning machine according to claim 5, wherein the method is characterized by comprising the following steps of: the hidden layer node number L set in step 61 is obtained by a five-fold cross validation method, that is, preset l= {10,50,100,200}, by dividing the sample set Φ into five parts randomly and averagely, training class labels of preset L values one by using four parts as training sets each time, weighting the model of the extreme learning machine, validating on the remaining part, calculating root mean square prediction precision of each L parameter after each sample is used as a test set, and selecting the L value with the largest precision as the optimal hidden layer node parameter value.
7. The coal mine gas early warning method based on class-weighted extreme learning machine according to claim 6, wherein the method is characterized by comprising the following steps of: the trained model in the step 7 requires that the gas concentration monitoring data of the same monitoring point for 24 hours are read in real time when the model is used.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011325639.9A CN112668749B (en) | 2020-11-24 | 2020-11-24 | Coal mine gas early warning method based on class mark weighting extreme learning machine |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011325639.9A CN112668749B (en) | 2020-11-24 | 2020-11-24 | Coal mine gas early warning method based on class mark weighting extreme learning machine |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112668749A CN112668749A (en) | 2021-04-16 |
CN112668749B true CN112668749B (en) | 2023-07-07 |
Family
ID=75403516
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011325639.9A Active CN112668749B (en) | 2020-11-24 | 2020-11-24 | Coal mine gas early warning method based on class mark weighting extreme learning machine |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112668749B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113605980B (en) * | 2021-07-23 | 2024-03-29 | 上海德衡数据科技有限公司 | Intelligent early warning mine safety method and system based on Internet of things |
CN114648217A (en) * | 2022-03-17 | 2022-06-21 | 重庆邮电大学 | Coal mine safety risk early warning method based on incremental extreme learning machine |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103678886A (en) * | 2013-11-26 | 2014-03-26 | 北京空间飞行器总体设计部 | Satellite Bayesian Network health determination method based on ground test data |
CN105975611A (en) * | 2016-05-18 | 2016-09-28 | 天津大学 | Self-adaptive combined downsampling reinforcing learning machine |
CN108875783A (en) * | 2018-05-09 | 2018-11-23 | 西安工程大学 | A kind of extreme learning machine Diagnosis Method of Transformer Faults towards unbalanced dataset |
CN110084291A (en) * | 2019-04-12 | 2019-08-02 | 湖北工业大学 | A kind of students ' behavior analysis method and device based on the study of the big data limit |
US10510003B1 (en) * | 2019-02-14 | 2019-12-17 | Capital One Services, Llc | Stochastic gradient boosting for deep neural networks |
CN111102005A (en) * | 2019-09-16 | 2020-05-05 | 中煤科工集团重庆研究院有限公司 | Mine environment gas concentration online prediction method |
CN111160461A (en) * | 2019-12-30 | 2020-05-15 | 浙江省机电设计研究院有限公司 | Fuzzy clustering-based weighted online extreme learning machine big data classification method |
-
2020
- 2020-11-24 CN CN202011325639.9A patent/CN112668749B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103678886A (en) * | 2013-11-26 | 2014-03-26 | 北京空间飞行器总体设计部 | Satellite Bayesian Network health determination method based on ground test data |
CN105975611A (en) * | 2016-05-18 | 2016-09-28 | 天津大学 | Self-adaptive combined downsampling reinforcing learning machine |
CN108875783A (en) * | 2018-05-09 | 2018-11-23 | 西安工程大学 | A kind of extreme learning machine Diagnosis Method of Transformer Faults towards unbalanced dataset |
US10510003B1 (en) * | 2019-02-14 | 2019-12-17 | Capital One Services, Llc | Stochastic gradient boosting for deep neural networks |
CN110084291A (en) * | 2019-04-12 | 2019-08-02 | 湖北工业大学 | A kind of students ' behavior analysis method and device based on the study of the big data limit |
CN111102005A (en) * | 2019-09-16 | 2020-05-05 | 中煤科工集团重庆研究院有限公司 | Mine environment gas concentration online prediction method |
CN111160461A (en) * | 2019-12-30 | 2020-05-15 | 浙江省机电设计研究院有限公司 | Fuzzy clustering-based weighted online extreme learning machine big data classification method |
Also Published As
Publication number | Publication date |
---|---|
CN112668749A (en) | 2021-04-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103455682B (en) | A method of prediction hp-ht well corrosion set pipe residue lifetime | |
CN112668749B (en) | Coal mine gas early warning method based on class mark weighting extreme learning machine | |
CN113723010B (en) | Bridge damage early warning method based on LSTM temperature-displacement correlation model | |
US11093314B2 (en) | Time-sequential data diagnosis device, additional learning method, and recording medium | |
CN108921359B (en) | Distributed gas concentration prediction method and device | |
CN112766429B (en) | Method, device, computer equipment and medium for anomaly detection | |
CN108470095B (en) | TBM (tunnel boring machine) propulsion prediction method based on data-driven radial basis function model | |
CN112308306A (en) | Multi-mode input coal and gas outburst risk prediction method | |
CN112257917B (en) | Time sequence abnormal mode detection method based on entropy characteristics and neural network | |
CN115186917A (en) | Active early warning type risk management and control system and method | |
CN105844501A (en) | Consumption behavior risk control system and method | |
CN110889440A (en) | Rockburst grade prediction method and system based on principal component analysis and BP neural network | |
CN112149750A (en) | Water supply network pipe burst identification data driving method | |
Zhang et al. | Geological Type Recognition by Machine Learning on In‐Situ Data of EPB Tunnel Boring Machines | |
CN112668873A (en) | Mine safety situation analysis and prediction early warning method | |
CN115906663A (en) | Building safety evaluation model establishing method, evaluation method, server and system | |
CN116777085B (en) | Coal mine water damage prediction system based on data analysis and machine learning technology | |
CN114154820A (en) | Production bottleneck prediction method based on increment simple cycle unit and double attention | |
CN107844679B (en) | Sectional type blast furnace molten iron silicon content prediction method and device | |
CN115700363B (en) | Method and system for diagnosing faults of rolling bearing of coal mining machine, electronic equipment and storage medium | |
CN111125965B (en) | Emergency regulation and control equipment reliability assessment method and system under disaster causing factor condition | |
CN107480647B (en) | Method for detecting abnormal behaviors in real time based on inductive consistency abnormality detection | |
CN115310705A (en) | Method and device for determining gas emission quantity and computer readable storage medium | |
CN114169597B (en) | Quantitative evaluation method for overall process of rock burst control measure implementation effect | |
CN113685166B (en) | Drilling accident early warning method and system |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |