CN112183901A - Coal and gas outburst strength prediction method based on deep learning - Google Patents
Coal and gas outburst strength prediction method based on deep learning Download PDFInfo
- Publication number
- CN112183901A CN112183901A CN202011229641.6A CN202011229641A CN112183901A CN 112183901 A CN112183901 A CN 112183901A CN 202011229641 A CN202011229641 A CN 202011229641A CN 112183901 A CN112183901 A CN 112183901A
- Authority
- CN
- China
- Prior art keywords
- model
- coal
- gas outburst
- training
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to the technical field of gas exploitation, and particularly relates to a coal and gas outburst strength prediction method based on deep learning, which comprises the following steps: preparing data, namely selecting prediction indexes of coal and gas outburst, defining training data, and carrying out standardization processing on the data; step two: feature extraction, namely defining network or model composition according to a data set, mapping input to a target, and extracting geological index features; step three: configuring a learning process, selecting a loss function, an optimizer and an index to be monitored, and setting iteration times; step four: training the model, inputting a sample and calling a fit method of the model to iterate on training data, and training and optimizing the model; step five: and the model is verified, the coal and gas outburst samples are predicted on a verification set and are compared with an actual result, and the prediction precision of the model is determined.
Description
Technical Field
The invention relates to the technical field of gas exploitation, in particular to a coal and gas outburst strength prediction method based on deep learning.
Background
Coal and gas outburst and other coal and rock dynamic disaster phenomena and accidents such as coal and gas explosion caused by the coal and gas dynamic disaster phenomena severely restrict the production of mines and the improvement of economic benefits. In recent years, with the increasing of the mining intensity and depth of coal mines, the coal and gas outburst problem is increasingly remarkable, and the effective outburst intensity prediction method has great significance for outburst prevention and outburst elimination work. At present, the methods for predicting coal and gas outburst in China mainly comprise qualitative comparative analysis, comprehensive evaluation methods, electromagnetic radiation prediction outburst, microseismic technology prediction outburst, linear regression analysis methods and the like. Because the coal and gas outburst influence factors are complex and have strong nonlinear characteristics, and the method mainly takes a single factor as a main factor, the method has great limitation and cannot truly reflect the outburst characteristics and the outburst area.
The nonlinear dynamic mechanism of the salient phenomenon shows that nonlinear mapping relations which are difficult to describe by an explicit function exist between each basic factor influencing the salient and the salient danger. The deep learning is used as a branch field of machine learning, the essential characteristic of the deep learning is a layer-by-layer training mechanism, the deep learning has strong nonlinear mapping capability and high self-learning and self-adapting capability, and compared with a traditional coal and gas outburst prediction method and a shallow neural network prediction method, the deep learning has stronger expression capability and better mapping effect, so that the accuracy of outburst strength prediction can be improved.
Therefore, a coal and gas outburst strength prediction model is established by using a deep learning framework Keras on the basis of fully extracting geological factors influencing outbursts by taking a certain mining area as a research object and taking a comprehensive effect hypothesis of coal and gas outbursts as a theoretical guidance, and a coal and gas outburst prediction method based on deep learning is provided.
Disclosure of Invention
This section is for the purpose of summarizing some aspects of embodiments of the invention and to briefly introduce some preferred embodiments. In this section, as well as in the abstract and the title of the invention of this application, simplifications or omissions may be made to avoid obscuring the purpose of the section, the abstract and the title, and such simplifications or omissions are not intended to limit the scope of the invention.
Therefore, the invention aims to provide a coal and gas outburst strength prediction method based on deep learning, which has stronger expression capability and better mapping effect, thereby being capable of improving the accuracy of outburst strength prediction.
To solve the above technical problem, according to an aspect of the present invention, the present invention provides the following technical solutions:
a coal and gas outburst strength prediction method based on deep learning comprises the following steps:
the method comprises the following steps: preparing data, namely selecting prediction indexes of coal and gas outburst, defining training data, and carrying out standardization processing on the data;
step two: feature extraction, namely defining network or model composition according to a data set, mapping input to a target, and extracting geological index features;
step three: configuring a learning process, selecting a loss function, an optimizer and an index to be monitored, and setting iteration times;
step four: training the model, inputting a sample and calling a fit method of the model to iterate on training data, and training and optimizing the model;
step five: and (5) verifying the model, predicting the coal and gas outburst samples on the verification set, comparing the predicted results with actual results, and determining the prediction accuracy of the model.
As a preferable scheme of the deep learning-based coal and gas outburst strength prediction method of the present invention, wherein: the cross entropy loss function, categoratic _ cross control, is used as the loss function.
As a preferable scheme of the deep learning-based coal and gas outburst strength prediction method of the present invention, wherein: the RMSprop optimization algorithm is used as an optimizer of the model to accelerate the learning speed of the algorithm.
Compared with the prior art, the invention has the beneficial effects that: the method is characterized in that a coal and gas outburst strength prediction model is established by utilizing a deep learning framework Keras, a coal and gas outburst prediction method based on deep learning is provided, learning is performed from continuous layers (layers) in a reinforcing mode, complex features can be expressed in a gradual and layer-by-layer mode, and then the feature engineering is automated.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the present invention will be described in detail with reference to the accompanying drawings and detailed embodiments, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise. Wherein:
FIG. 1 is a flow chart of the steps of the present invention;
FIG. 2 is a flow chart of deep learning according to the present invention;
FIG. 3 is a diagram of a neural network architecture according to the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways than those specifically described herein, and it will be apparent to those of ordinary skill in the art that the present invention may be practiced without departing from the spirit and scope of the present invention, and therefore the present invention is not limited to the specific embodiments disclosed below.
Next, the present invention will be described in detail with reference to the drawings, wherein for convenience of illustration, the cross-sectional view of the device structure is not enlarged partially according to the general scale, and the drawings are only examples, which should not limit the scope of the present invention. In addition, the three-dimensional dimensions of length, width and depth should be included in the actual fabrication.
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
Example 1
30 outburst accident cases occurring in the Panjiang mine area are taken as samples for forecasting the outburst strength, and the forecasting of the outburst strength is realized mainly through the following links.
(1) Selecting indexes and carrying out quantitative processing;
according to the prominent comprehensive hypothesis theory, gas geology and gas prominent prediction are combined, 12 parameters are selected from three aspects of geological structure, coal body structure and gas as the influence factor indexes of the prominent accidents (table 1).
TABLE 1 highlighting impact factor indicators for accidents
The collection of the impact factor indexes of the protrusion intensity is performed for 30 cases of protrusion accidents, and the mathematical and dimensionless treatment is performed on each prediction index (table 2).
TABLE 2 training samples
(2) And (5) carrying out numerical processing on the label. Before a neural network reads a sample, vectorization processing is carried out on output, and the protrusion intensity is divided into three types according to the convention, wherein small protrusions with the protrusion coal amount of less than 100t are classified into one type and are set as (1,0,0), medium protrusions with the protrusion coal amount of 100-500 t are classified into one type and are set as (0,1,0), and large and extra-large protrusions with the protrusion coal amount of more than 500t are classified into one type and are set as (0,0, 1).
(3) And determining the network structure. The total number of evaluation indexes is 12, so that an input layer m for determining deep learning is 12 and corresponds to 12 geological factors influencing coal and gas outburst respectively, and an output layer n is 3 and corresponds to three types in the labels respectively. Due to the limited number of samples, the number of hidden layers is determined to be 2, the number of neurons is 32, the whole network has 2627 total parameters, and the specific structure is shown in fig. 2.
(4) The loss function (objective function) is mainly used to measure the error between the predicted value and the target value, and a cross-entropy loss function (cross-entropy _ cross-entropy) is adopted as the loss function [10] in consideration of the multi-classification characteristic of the model. And (3) adopting an RMSprop optimization algorithm [11] as an optimizer of the model to accelerate the learning speed of the algorithm.
(5) And training a prediction model.
The data of the training set are input into the model for training after being subjected to feature standardization processing, the precision of the model for the training set reaches 0.9091 after 26 iterations, the loss is reduced to 0.1515, and the prediction model can meet the precision requirement of coal and gas outburst prediction.
(6) Prediction results and instance validation
To verify the effectiveness of the prediction model, 7 outstanding accident cases are selected from the scene to test the model, and the quantitative result of the prediction parameters of the outstanding cases is shown in the table 3.
Table 3 verification examples
The model prediction results are shown in table 4.
TABLE 4 Back-judgment verification results
As can be seen from Table 4, 6 of the 7 prediction results are completely consistent with the actual results, only the sample No. 4 has the small-sized outburst error prediction as the medium-sized outburst, the model prediction accuracy rate is more than 85%, and the requirements of coal and gas outburst strength prediction can be met.
While the invention has been described above with reference to an embodiment, various modifications may be made and equivalents may be substituted for elements thereof without departing from the scope of the invention. In particular, the various features of the disclosed embodiments of the invention may be used in any combination, provided that no structural conflict exists, and the combinations are not exhaustively described in this specification merely for the sake of brevity and resource conservation. Therefore, it is intended that the invention not be limited to the particular embodiments disclosed, but that the invention will include all embodiments falling within the scope of the appended claims.
Claims (3)
1. A coal and gas outburst strength prediction method based on deep learning is characterized by comprising the following steps: the method comprises the following steps:
the method comprises the following steps: preparing data, namely selecting prediction indexes of coal and gas outburst, defining training data, and carrying out standardization processing on the data;
step two: feature extraction, namely defining network or model composition according to a data set, mapping input to a target, and extracting geological index features;
step three: configuring a learning process, selecting a loss function, an optimizer and an index to be monitored, and setting iteration times;
step four: training the model, inputting a sample and calling a fit method of the model to iterate on training data, and training and optimizing the model;
step five: and (5) verifying the model, predicting the coal and gas outburst samples on the verification set, comparing the predicted results with actual results, and determining the prediction accuracy of the model.
2. The method for predicting coal and gas outburst strength based on deep learning according to claim 1, wherein the method comprises the following steps: the cross entropy loss function, categoratic _ cross control, is used as the loss function.
3. The method for predicting coal and gas outburst strength based on deep learning according to claim 1, wherein the method comprises the following steps: the RMSprop optimization algorithm is used as an optimizer of the model to accelerate the learning speed of the algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011229641.6A CN112183901A (en) | 2020-11-06 | 2020-11-06 | Coal and gas outburst strength prediction method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011229641.6A CN112183901A (en) | 2020-11-06 | 2020-11-06 | Coal and gas outburst strength prediction method based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112183901A true CN112183901A (en) | 2021-01-05 |
Family
ID=73917845
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011229641.6A Pending CN112183901A (en) | 2020-11-06 | 2020-11-06 | Coal and gas outburst strength prediction method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112183901A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116467572A (en) * | 2023-04-27 | 2023-07-21 | 中国矿业大学(北京) | Coal-rock gas composite dynamic disaster prediction method based on deep learning |
| CN116975775A (en) * | 2023-06-29 | 2023-10-31 | 中信重工开诚智能装备有限公司 | Deep learning method for coal mine gas outburst prediction based on multi-mode data fusion |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107403234A (en) * | 2017-06-24 | 2017-11-28 | 中铁二院工程集团有限责任公司 | Tunnel Coal and Gas Outbursts Prediction method based on neutral net |
| CN111126658A (en) * | 2019-11-15 | 2020-05-08 | 太原理工大学 | Coal mine gas prediction method based on deep learning |
-
2020
- 2020-11-06 CN CN202011229641.6A patent/CN112183901A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107403234A (en) * | 2017-06-24 | 2017-11-28 | 中铁二院工程集团有限责任公司 | Tunnel Coal and Gas Outbursts Prediction method based on neutral net |
| CN111126658A (en) * | 2019-11-15 | 2020-05-08 | 太原理工大学 | Coal mine gas prediction method based on deep learning |
Non-Patent Citations (1)
| Title |
|---|
| 魏丹: "支持向量机多分类预测技术研究——桑树坪煤矿煤与瓦斯突出预测", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116467572A (en) * | 2023-04-27 | 2023-07-21 | 中国矿业大学(北京) | Coal-rock gas composite dynamic disaster prediction method based on deep learning |
| CN116467572B (en) * | 2023-04-27 | 2023-11-21 | 中国矿业大学(北京) | Coal-rock gas composite dynamic disaster prediction method based on deep learning |
| CN116975775A (en) * | 2023-06-29 | 2023-10-31 | 中信重工开诚智能装备有限公司 | Deep learning method for coal mine gas outburst prediction based on multi-mode data fusion |
| CN116975775B (en) * | 2023-06-29 | 2024-01-30 | 中信重工开诚智能装备有限公司 | Deep learning method for coal mine gas outburst prediction based on multi-mode data fusion |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111860952B (en) | Method for rapidly optimizing key mining parameters of outburst coal seam | |
| Lyu et al. | LSTM based encoder-decoder for short-term predictions of gas concentration using multi-sensor fusion | |
| CN112183901A (en) | Coal and gas outburst strength prediction method based on deep learning | |
| CN112308306A (en) | Multi-mode input coal and gas outburst risk prediction method | |
| CN113946950B (en) | Method for rapidly delineating target area of gold prospecting | |
| CN112529341B (en) | Drilling well leakage probability prediction method based on naive Bayesian algorithm | |
| CN115345378B (en) | Shale gas well yield evaluation method based on machine learning | |
| CN108952699B (en) | Intelligent stratum lithology identification method in complex geological drilling process | |
| CN114154484B (en) | Construction professional term library intelligent construction method based on mixed depth semantic mining | |
| CN110705179A (en) | Water guide crack belt height accurate prediction method and system based on fusion neural network | |
| CN109543352B (en) | Shale fracture toughness prediction method and device and electronic equipment | |
| CN112035345A (en) | Mixed depth defect prediction method based on code segment analysis | |
| CN116911432A (en) | Prediction method, storage medium and processor for horizontal well productivity | |
| CN111102006B (en) | Dynamic early warning intelligent decision method for outburst mine extraction and mining deployment imbalance | |
| CN116756679A (en) | Multi-source information fusion-based method for judging geological mode of down-the-hole mountain | |
| CN106126736A (en) | Software developer's personalized recommendation method that software-oriented safety bug repairs | |
| Qodirov et al. | Development of artificial neural network for predicting drill pipe sticking in real-time well drilling process | |
| CN114117654A (en) | Horizontal well perforation optimization design method and device based on machine learning | |
| CN111061935B (en) | Science and technology writing recommendation method based on self-attention mechanism | |
| CN111273352A (en) | Intelligent detection method and device for geological structure and electronic equipment | |
| Fan et al. | A Novel Ensemble Machine Learning Model for Oil Production Prediction with Two-Stage Data Preprocessing | |
| CN113887125A (en) | Complex simulation system operation effectiveness evaluation method | |
| Ma et al. | In-situ stresses prediction by using a CNN-BiLSTM-Attention hybrid neural network | |
| Jian et al. | Research on early-warning method and its application of complex system of circular economy for oil and gas exploitation | |
| Wang et al. | A LSTM-based method for simulation execution validity evaluation |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210105 |