CN103778469A - Blasting scheme selection method based on neural network optimization genetic algorithm - Google Patents
Blasting scheme selection method based on neural network optimization genetic algorithm Download PDFInfo
- Publication number
- CN103778469A CN103778469A CN201310023238.1A CN201310023238A CN103778469A CN 103778469 A CN103778469 A CN 103778469A CN 201310023238 A CN201310023238 A CN 201310023238A CN 103778469 A CN103778469 A CN 103778469A
- Authority
- CN
- China
- Prior art keywords
- blasting
- neural network
- genetic algorithm
- ann
- value
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
This invention discloses a blasting scheme selection method based on neural network optimization genetic algorithm and is characterized by using blasting impact factors and blasting hazard forms as an input value and an output value of the neural network to practice, and the practiced neural network is used as a fitness function for the genetic algorithm. The blasting impact factor include : blasthole (HL), spacing ((i)S(/i)), charge deepness ((i)B(i)), blocking deepness ((i)ST(/I)), specific charge ((i)PF(/i)), and hole drilling rate ((i)SD(/i)), and the blasting hazard forms include overbreak deepness ((i)B(/i)) and a distance of flying rocks ((i)FR(/i)). The genetic algorithm (GA) is used to find the best overbreak deepness ((i)B(/i)) and the distance of flying rocks ((i)FR(/i)) so as to optimize the blasting scheme parameters. The blasting scheme parameter optimization comprises data collection, fitness function construction based on genetic algorithm of ANN, blasting scheme parameter preference based on the genetic algorithm (GA) and determination of the final result of the blasting optimization scheme according to Pareto picture. The blasting scheme selection method can be widely applicable to the blasting scheme optimization selection during an exploitation of a strip mine.
Description
Technical field
the present invention relates to strip mining transformation blasting scheme is selected,
particularly relate to blasting scheme system of selection based on Neural Network Optimization genetic algorithm.
Background technology
The formulation of blasting scheme is the important content in mining activity.Parameter in scheme is selected affected by several factors.The blasting scheme that different mining areas is used is all variant, is mainly physical and mechanical property and the groundwater environment etc. of considering output, geologic condition, rock.
Definite requirement that should meet safety, technology and economic aspect of blasting parameter.No person will produce serious accident, and wherein the overbreak degree of depth (BB) and stone fling distance (FR) (hereinafter referred overbreak and slungshot) are one of the most common and dangerous accidents.Overbreak is the phenomenon that exceedes predetermined depth due to the length of shot that inappropriate parameter setting causes, and it will exceed prior prediction to the destruction of rock mass, causes the accident such as landslide and avalanche of rock; Slungshot (FR) is due to parameter improper rock after causing explosion energy to make the to destroy safe range of setting in advance that flies out equally, gives the man-machine accident damaging around.
Some researchs are done to overbreak with slungshot and correlative factor thereof both at home and abroad.But the effect of the overbreak and slungshot research based on experimental formula is also bad, and this is because experimental formula just simply meets experimental data, may not have theoretical foundation at all.
Use ANN using dark the borehole in blasting scheme (HL), spacing (S), shot depth (B), the obstruction degree of depth (ST), powder factor (PF) and boring rate (SD) as its input value, overbreak (BB) and slungshot (FR) are trained and predicted as output valve, and done network error analysis.Neural network (hereinafter referred ANN) after use training, as the fitness function of genetic algorithm (hereinafter referred GA), is optimized blasting parameter by GA, finally uses Pareto figure to obtain optimized parameter.
Summary of the invention
In strip mining transformation process, the problem that blasting scheme exists, the present invention proposes a kind of blasting scheme system of selection based on Neural Network Optimization genetic algorithm.
1. the data of choosing: collect data for training and verification model, borehole dark (
hL), spacing (
s), shot depth (
b), block the degree of depth (
sT), powder factor (
pF) and boring rate (
sD) as its input value, by overbreak (
bB) and slungshot (
fR) as output valve.Data (
hL,
s,
b,
sT,
pF,
sD,
bB,
fR) be altogether 100 groups, be divided at random training set (80%) and test set (20%).
2. the structure of the suitable function of the GA based on ANN: the main task of ANN training is to select to imply the number of plies, imply into interior neuron number, transport function and weight.This adaptive process is exactly network error while reaching minimum threshold values, determines the process of above-mentioned parameter.Network error normally by square error (
mSE) and the coefficient of determination (r
2) judge as evaluation index,
mSEas the formula (1)
In formula: O
iand T
ithe output valve of representative prediction and actual measurement respectively, N represents data logarithm.
In order to improve efficiency and the generalization of training, the input value of ANN and output valve should be standardized, and these values drop on according to corresponding algorithm
In [1,1].Formula (2) value of making of formatting use linear gauge drops in above-mentioned scope.
Predict with feed forward type neural network (FFNN).First to determine the implicit network number of plies and calculate neuronic quantity effective utilization of FFNN.If hidden layer has enough hidden neurons, using so the two-layer neural network of hyperbolic curve tangent S transport function and linear transmission function composition is comparatively suitable structure.Neuronic quantity is to avoid the key of over adaptation problem, if the performance of some ANN adaptation training data is identical, so the simplest ANN is best.According to the experience of current research, determine that the neuronal quantity in hidden layer can not be by accurately calculating, generally can only after training and prediction, could, to its adjustment to increase its adaptability, still can estimate it by some relevant parameters, as shown in table 1.
Neuronic estimate amount in single hidden layer of table 1 data
Note: n
i, n
0and n
trespectively input neuron quantity, output neuron quantity and training sample quantity, n in this example
i=2, n
0=1, n
t=35(50 × 70%, the 50th, minimum experiment number); K is noise figure, k=4; ;
the constant that crosses the border,
=1.25.
3. the blasting scheme optimization of parameters based on GA: the ANN after use training is as GA fitness function, to the optimizing process of blasting parameter as shown in Figure 1.
Input value and the scope thereof optimized are set, and initial population number equates with the data set quantity of ANN, and chromosomal length and genic value are by the automatic setting of MATLAB.Chromosomal evaluation is by training the ANN obtaining to complete as fitness function through upper step.If the chromosome after evaluating meets the hereditary condition that stops, determining that so the chromosome of final algebraically carries out Pareto map analysis, finally corresponding
bBwith
fRminimum chromosome is optimum solution.
Accompanying drawing explanation
The optimizing process of Fig. 1 GA to blasting parameter
The real figure in Fig. 2 iron ore field
The predicted value of Fig. 3 slungshot and measured value graph of a relation
The super sudden and violent predicted value of Fig. 4 and measured value graph of a relation
Fig. 5 GA optimum results Pareto contrasts caption: the data that use in figure are normalizing data.
Slungshot and overbreak and value caption in Fig. 6 Pareto figure: the data that use in figure are normalizing data.
Embodiment
For above-mentioned purpose of the present invention, feature and advantage are become apparent more, below in conjunction with the correlation theory using and embodiment, the present invention is further detailed explanation.
Anshan iron and steel plant group iron ore as shown in Figure 2, is positioned at the southeast, Anshan 12km, and iron ore-deposit is positioned at numerous mountains mountain range In Northwestern Margin, and landforms belong to undulating topography, and in existing mining area, peak is positioned at northeast, mining area, and height above sea level is 100.2m, and exploitation is to approximately-280m level now.Ore bodies exists in Anshan group metamorphic rocks, ore body lower wall and gneissic flower hilllock matter mixed rock are with F15 fault contact, upper dish is integrated and is contacted with green mud quartz-schist, the east and Qianshan Granite are with F1 fault contact, west end is with migmatitic granite with F14 fault contact, and 300～320 ° of orebody trends, are inclined to NE, 60～80 °, inclination angle, ore body and country rock occurrence are basically identical.
The data of collecting comprise parameter
hL,
s,
b,
sT,
pFwith
sDas its input value,
bBwith
fRas output, parameter correlation information is as shown in table 2.Data (
hL,
s,
b,
sT,
pF,
sD,
bB,
fR) be altogether 100 groups, be divided at random training set (80%) and test set (20%).
Table 2 input/output argument relevant information
The main task of ANN training is to select to imply the number of plies, imply into interior neuron number, transport function and weight.This adaptive process is exactly network error while reaching minimum threshold values, determines the process of above-mentioned parameter.Network error normally by square error (
mSE) and the coefficient of determination (r
2) judge as evaluation index,
mSEas the formula (1).
In order to improve efficiency and the generalization of training, the input value of ANN and output valve should be standardized, and these values drop in [1,1] according to corresponding algorithm.Formula (2) value of making of formatting use linear gauge drops in above-mentioned scope.
Predict with feed forward type neural network (FFNN).First to determine the implicit network number of plies and calculate neuronic quantity effective utilization of FFNN.If hidden layer has enough hidden neurons, using so the two-layer neural network of hyperbolic curve tangent S transport function and linear transmission function composition is comparatively suitable structure.Neuronic quantity is to avoid the key of over adaptation problem, if the performance of some ANN adaptation training data is identical, so the simplest ANN is best.According to the experience of current research, determine that the neuronal quantity in hidden layer can not be by accurately calculating, generally can only after training and prediction, could, to its adjustment to increase its adaptability, still can estimate it by some relevant parameters.Final ANN structure is 6 neurons of input layer, and hidden layer 1 comprises 12 neurons, and hidden layer 2 comprises 6 neurons, and output layer comprises 2 neurons.
Fig. 3 and Fig. 4 represent respectively the relation between slungshot and super sudden and violent measured value and predicted value.The estimated performance of ANN model that therefrom can find out structure is very high.In the time carrying out the member of next step model, this ANN model, using the fitness function as the GA to blasting parameters optimization, further participates in parameter optimization.
4 blasting scheme optimization of parameters based on GA
ANN after use training is as GA fitness function, to the optimizing process of blasting parameter as shown in Figure 1.
Input value and the scope thereof optimized are set, and initial population number equates with the data set quantity of ANN, and chromosomal length and genic value are by the automatic setting of MATLAB.Chromosomal evaluation is by training the ANN obtaining to complete as fitness function through upper step.If the chromosome after evaluating meets the hereditary condition that stops, determining that so the chromosome of final algebraically (this example was 90 generations) carries out Pareto map analysis, finally corresponding
bBwith
fRminimum chromosome is optimum solution.
Because fitness function comprises two output valves, should use so multiple objective function optimization.For this function, often use tournament mode to carry out the selection of coefficient.The two point method that interaction coefficent is is 90% by crossover probability is determined.It is variation probability that the coefficient of variation is used 1% of the general application of engineering problem.
Fig. 5 has shown the result of GA Blasting Parameters Optimization.Because the suitable function of target is two-dimentional output function, use Pareto figure to be described here.In the time that GA was converged in for 90 generation, reach hereditary stop condition, algorithm stops, and 24 chromosomes that at this moment remain are optimum solutions.First point in figure (first point of upper left) slungshot value minimum, super sudden and violent value maximum (value after normalization), last point (last point of bottom right) slungshot value maximum, super sudden and violent value is minimum.According to the character of Pareto figure, optimum chromosome should be two coordinates and minimum point.Fig. 6 show in Pareto figure each put two coordinates and.Minimum value is the 11st point (chromosome) as can be seen from Figure, minimum value and be 0.05801.According to above-mentioned 24 chromosomes, according to the blasting scheme parameter value of individual features and formula (2), optimize output respectively
fRwith
bBvalue (the 14th behavior optimal value, i.e. optimum blasting scheme) as shown in table 3.Borehole dark (
hL), spacing (
s), shot depth (
b), block the degree of depth (
sT), powder factor (
pF) and boring rate (
sD) as its input value, by overbreak (
bB) and slungshot (
fR) do
Table 3 explosion prioritization scheme net result
Consider at the same time that slungshot and overbreak two aspects reach optimum, genetic algorithm has generated 24 chromosomes that finally reach hereditary condition.Analysis by Pareto figure to two factors, finally selects Sub_clause 11 chromosome to require the superior in chromosome as meeting heredity.At this moment minimum value and be 0.05801 after slungshot and overbreak normalization, actual value is respectively stone fling distance 27.3386m, overbreak degree of depth 0.6999m.I.e. the 11st behavior optimum blasting scheme in table 2.
Claims (9)
1. the blasting scheme system of selection based on Neural Network Optimization genetic algorithm,
it is characterized in that, the ANN using, for feed forward type neural network, optimizes by GA
bBwith
fR, first collect data for training and verification model, borehole dark (
hL), spacing (
s), shot depth (
b), block the degree of depth (
sT), powder factor (
pF) and boring rate (
sD) as its input value, by overbreak (
bB) and slungshot (
fR) as output valve (hereinafter use be called for short), optimum ANN after training, as the fitness function of GA, is optimized blasting parameter scheme, finally for two-dimensional optimization problem, use Pareto figure to determine the blasting parameter scheme of safety economy,
the present invention blasting scheme optimization in opencast mining engineering is selected.
2.
neural network according to claim 1, is characterized in that, feed forward type neural network (FFNN), the two-layer neural network that uses hyperbolic curve tangent S transport function and linear transmission function to form.
3.
feed forward type neural network neural network according to claim 2, is characterized in that, in hidden layer, neuronic estimative figure measures
minimum value.
4.
in hidden layer according to claim 2, neuronic estimate amount, is characterized in that, the value of estimator: n
i, n
0and n
trespectively input neuron quantity, output neuron quantity and training sample quantity, n
i=2, n
0=1, n
t=35(50 × 70%, the 50th, minimum experiment number); K is noise figure, k=4; ;
the constant that crosses the border,
=1.25.
5.
neural network according to claim 1, is characterized in that, god should standardize network input value and output valve, and these values drop in [1,1] according to corresponding algorithm, and normalization formula is:
In formula: x
'the value after normalization, x
maxand x
minbe respectively the maximal value and the minimum value that are normalized in experimental data ordered series of numbers.
6.
neural network according to claim 1, is characterized in that, collect data for training and verification model, borehole dark (
hL), spacing (
s), shot depth (
b), block the degree of depth (
sT), powder factor (
pF) and boring rate (
sD) as its input value, by overbreak (
bB) and slungshot (
fR) as output valve, data (
hL,
s,
b,
sT,
pF,
sD,
bB,
fR) be altogether 100 groups, be divided at random training set (80%) and test set (20%).
7.
genetic algorithm according to claim 1, is characterized in that, the suitable function of GA be based on ANN structure, error be by square error (
mSE) and the coefficient of determination (r
2) judge as evaluation index.
8.
blasting scheme system of selection according to claim 1, is characterized in that, blasting scheme optimization of parameters based on GA, ANN after use training is as GA fitness function, input value and the scope thereof optimized are set, and initial population number equates with the data set quantity of ANN, and chromosomal length and genic value are by the automatic setting of MATLAB.
9.
chromosome according to claim 8, is characterized in that, chromosomal evaluation is by training the ANN obtaining to complete as fitness function through upper step, if the chromosome after evaluating meets the hereditary condition that stops, determining that so the chromosome of final algebraically carries out Pareto map analysis, finally corresponding
bBwith
fRminimum chromosome is optimum solution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310023238.1A CN103778469A (en) | 2013-01-23 | 2013-01-23 | Blasting scheme selection method based on neural network optimization genetic algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310023238.1A CN103778469A (en) | 2013-01-23 | 2013-01-23 | Blasting scheme selection method based on neural network optimization genetic algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103778469A true CN103778469A (en) | 2014-05-07 |
Family
ID=50570679
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310023238.1A Pending CN103778469A (en) | 2013-01-23 | 2013-01-23 | Blasting scheme selection method based on neural network optimization genetic algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103778469A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104390534A (en) * | 2014-10-11 | 2015-03-04 | 同济大学 | Tunnel smooth surface blasting quality control method |
| CN106290263A (en) * | 2015-05-19 | 2017-01-04 | 中国科学院沈阳自动化研究所 | A kind of LIBS calibration and quantitative analysis method based on genetic algorithm |
| CN107506831A (en) * | 2017-08-03 | 2017-12-22 | 中国矿业大学(北京) | Blasting parameters determination method and system |
| CN108399296A (en) * | 2018-02-27 | 2018-08-14 | 大连海事大学 | A kind of vibration velocity prediction technique for the foundation pit static blasting closing on structures |
| CN109102109A (en) * | 2018-07-16 | 2018-12-28 | 四川大学 | A kind of piece of exploitation of stone explosion gradation prediction technique |
| CN110457758A (en) * | 2019-07-16 | 2019-11-15 | 江西理工大学 | Prediction technique, device, system and the storage medium in Instability of Rock Body stage |
| CN111062113A (en) * | 2018-12-03 | 2020-04-24 | 湘潭大学 | Novel stope mining blasting parameter comprehensive optimization method under complex filling body condition |
| CN111259601A (en) * | 2020-01-16 | 2020-06-09 | 南华大学 | Blasting blockiness prediction method, device and medium based on random GA-BP neural network group |
| CN111950203A (en) * | 2020-08-13 | 2020-11-17 | 中核华辰建筑工程有限公司 | Blasting vibration speed prediction method based on adaptive neural fuzzy inference system |
| CN112800673A (en) * | 2021-01-27 | 2021-05-14 | 昆明理工大学 | Method for predicting blasting block degree based on SA-GA-BP algorithm |
| US11307327B2 (en) | 2018-10-26 | 2022-04-19 | Tata Consultancy Services Limited | Method and system for online monitoring and optimization of mining and mineral processing operations |
| CN118070670A (en) * | 2024-03-13 | 2024-05-24 | 北京交通大学 | Digital twinning-based intelligent optimization system for smooth blasting by tunnel drilling and blasting method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050267851A1 (en) * | 2004-05-25 | 2005-12-01 | Semiconductor Leading Edge Technologies, Inc. | Fitness function circuit, genetic algorithm machine, and fitness evaluation method |
| CN101191416A (en) * | 2006-11-27 | 2008-06-04 | 杨仁树 | Rock lane digging method |
| CN101436264A (en) * | 2008-11-28 | 2009-05-20 | 北京交通大学 | Mine downhole debris flow prediction method based on subsection fuzzy BP neural network |
-
2013
- 2013-01-23 CN CN201310023238.1A patent/CN103778469A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050267851A1 (en) * | 2004-05-25 | 2005-12-01 | Semiconductor Leading Edge Technologies, Inc. | Fitness function circuit, genetic algorithm machine, and fitness evaluation method |
| CN101191416A (en) * | 2006-11-27 | 2008-06-04 | 杨仁树 | Rock lane digging method |
| CN101436264A (en) * | 2008-11-28 | 2009-05-20 | 北京交通大学 | Mine downhole debris flow prediction method based on subsection fuzzy BP neural network |
Non-Patent Citations (5)
| Title |
|---|
| A.BAHRAMI等: "Prediction of rock fragmentation due to blasting using artificial neural network", 《ENGINEERING WITH COMPUTERS》 * |
| 曹跃: "厂房近区中深孔爆破技术的应用", 《爆破》 * |
| 许洪涛，卢文波: "《遗传算法在工程爆破参数优化中的应用》", 《中国工程科学》 * |
| 赵彬等: "《基于BP神经网络的爆破参数优选》", 《矿冶工程》 * |
| 雷松林: "《基于BP网络和遗传算法的岩爆预测研究》", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104390534A (en) * | 2014-10-11 | 2015-03-04 | 同济大学 | Tunnel smooth surface blasting quality control method |
| CN106290263A (en) * | 2015-05-19 | 2017-01-04 | 中国科学院沈阳自动化研究所 | A kind of LIBS calibration and quantitative analysis method based on genetic algorithm |
| CN106290263B (en) * | 2015-05-19 | 2018-08-24 | 中国科学院沈阳自动化研究所 | A kind of LIBS calibration and quantitative analysis methods based on genetic algorithm |
| CN107506831A (en) * | 2017-08-03 | 2017-12-22 | 中国矿业大学(北京) | Blasting parameters determination method and system |
| CN108399296A (en) * | 2018-02-27 | 2018-08-14 | 大连海事大学 | A kind of vibration velocity prediction technique for the foundation pit static blasting closing on structures |
| CN109102109A (en) * | 2018-07-16 | 2018-12-28 | 四川大学 | A kind of piece of exploitation of stone explosion gradation prediction technique |
| US11307327B2 (en) | 2018-10-26 | 2022-04-19 | Tata Consultancy Services Limited | Method and system for online monitoring and optimization of mining and mineral processing operations |
| CN111062113B (en) * | 2018-12-03 | 2023-08-25 | 湘潭大学 | Comprehensive optimization method for mining explosion parameters of stope under complex filling conditions |
| CN111062113A (en) * | 2018-12-03 | 2020-04-24 | 湘潭大学 | Novel stope mining blasting parameter comprehensive optimization method under complex filling body condition |
| CN110457758B (en) * | 2019-07-16 | 2023-03-24 | 江西理工大学 | Method, device and system for predicting unstable phase of rock mass and storage medium |
| CN110457758A (en) * | 2019-07-16 | 2019-11-15 | 江西理工大学 | Prediction technique, device, system and the storage medium in Instability of Rock Body stage |
| CN111259601A (en) * | 2020-01-16 | 2020-06-09 | 南华大学 | Blasting blockiness prediction method, device and medium based on random GA-BP neural network group |
| CN111950203A (en) * | 2020-08-13 | 2020-11-17 | 中核华辰建筑工程有限公司 | Blasting vibration speed prediction method based on adaptive neural fuzzy inference system |
| CN112800673A (en) * | 2021-01-27 | 2021-05-14 | 昆明理工大学 | Method for predicting blasting block degree based on SA-GA-BP algorithm |
| CN118070670A (en) * | 2024-03-13 | 2024-05-24 | 北京交通大学 | Digital twinning-based intelligent optimization system for smooth blasting by tunnel drilling and blasting method |
| CN118070670B (en) * | 2024-03-13 | 2024-09-13 | 北京交通大学 | Digital twinning-based intelligent optimization system for smooth blasting by tunnel drilling and blasting method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103778469A (en) | Blasting scheme selection method based on neural network optimization genetic algorithm | |
| CN109447837B (en) | Danger assessment method for rock burst in mining area | |
| CN103995947B (en) | Improved coal seam floor water inrush vulnerability evaluation method | |
| CN109711597A (en) | A kind of Copper-nickel Sulfide Ore Deposit metallogenic prognosis method based on stratified random forest model | |
| CN108399497B (en) | Strip mining goaf stability evaluation method | |
| El-Kadi et al. | Assessing sustainability of groundwater resources on Jeju Island, South Korea, under climate change, drought, and increased usage | |
| CN102272414A (en) | Method for optimizing well production in reservoirs having flow barriers | |
| CN107122860B (en) | Rock burst danger level prediction method based on grid search and extreme learning machine | |
| CN107676064A (en) | Water-drive reservoir water content prediction method and prediction device thereof | |
| CN103412104A (en) | Method used for evaluating frangibility of underground water | |
| CN104063608B (en) | Method of determining underground water controlled water level by utilizing risk evaluation | |
| CN103244187A (en) | Early-warning comprehensive index system of coal and gas outburst inoculation and development process | |
| CN104390534B (en) | A kind of tunnel photoface exploision method of quality control | |
| CN116050877A (en) | Metal mine aquifer water enrichment evaluation method and evaluation system | |
| CN115980826A (en) | Rock burst intensity prediction method based on weighted meta-heuristic combined model | |
| CN103628816A (en) | Seismic attribute guidance horizontal-well-steering method and device thereof | |
| CN107123058A (en) | A kind of Method for Slope Stability Analysis | |
| CN105089659B (en) | Conglomerate oil reservoir seepage unit identification method | |
| CN104866664B (en) | A kind of Impermeable underground wall curtain body construction difficulty Forecasting Methodology | |
| CN104636802A (en) | Blasting scheme multi-criteria selection method based on improved genetic algorithm | |
| CN117033940A (en) | Surface mine blasting vibration speed prediction method based on Pearson-CNN model | |
| CN104834934B (en) | A kind of nucleome capture method for being used to identify reservoir fluid | |
| CN108549994A (en) | A kind of evaluation method of coal mine empty water water damage always | |
| CN105718668A (en) | Open pit mine cast blasting effect analysis method | |
| CN110457831A (en) | A kind of covering suitability evaluation methods of groundwater source field |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20140507 |
|
| RJ01 | Rejection of invention patent application after publication |