CN113569487B - BP neural network-based method for predicting step blasting throwing effect - Google Patents
BP neural network-based method for predicting step blasting throwing effect Download PDFInfo
- Publication number
- CN113569487B CN113569487B CN202110881576.3A CN202110881576A CN113569487B CN 113569487 B CN113569487 B CN 113569487B CN 202110881576 A CN202110881576 A CN 202110881576A CN 113569487 B CN113569487 B CN 113569487B
- Authority
- CN
- China
- Prior art keywords
- neural network
- throwing
- blasting
- training
- network model
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- 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
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Business, Economics & Management (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Data Mining & Analysis (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Geometry (AREA)
- General Business, Economics & Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a method for predicting a step blasting throwing effect based on a BP neural network, and belongs to the field of blasting engineering. The method comprises the following steps: and (3) designing a model test according to actual working conditions, carrying out tests under different groups, and recording test data. Applying a BP neural network model, taking the step height, the coal seam thickness, the goaf upper opening width, the minimum resistance line, the pitch, the row spacing, the inter-hole differential time and the coal seam slope angle as input parameters, taking the throwing rate, the pile loosening coefficient and the farthest throwing distance as output parameters, and importing test data into the neural network model for training; the neural network model after training can predict the step blasting throwing effect. The invention provides a method for predicting the step blasting throwing effect based on a BP neural network, which can accurately predict the step blasting throwing effect under specific working conditions and judge the feasibility of a blasting scheme. Provides reference for mine step blasting and improves the reliability of throwing blasting.
Description
Technical Field
The invention relates to a method for predicting a step blasting throwing effect based on a BP neural network, and belongs to the field of blasting engineering.
Background
In the open pit coal mine throwing blasting technology, measures such as large aperture, deep hole high step, large loading capacity, uncoupled loading capacity and the like are generally adopted, and partial rocks are thrown to a goaf by utilizing energy generated by explosive explosion without the need of moving equipment. This portion of rock thrown into the goaf is referred to as the effective throw, and its ratio to the total blasted rock mass is referred to as the effective throw rate, which is one of the important indicators for the effectiveness of the throw blast. On the other hand, the throwing blasting technology is only a part of production process links, and has to work together with other stripping equipment, the post-throwing blasting pile form is an important factor influencing the working efficiency of subsequent equipment, and the improvement of the effective throwing rate and the control of the throwing blasting pile form are key ways for saving the production cost and improving the working efficiency of the subsequent equipment. Obviously, the step throwing blasting parameters play an important role in the effective throwing rate and the form of the throwing blasting explosion stack, and the step throwing blasting parameters need to be optimally studied.
In the throwing blasting, the effective throwing rate is required to be determined according to the form of the blasting stack, so as to determine the reverse pile operation amount of the system; when the dragline is operated, the dragline stands on the flattened blasting pile, the blasting pile shape influences the design of the parameters of the working face of the inverted pile and the engineering quantity for constructing an expansion platform, so that the research on the morphological characteristics of the blasting pile is the most important basic work in the optimization design of the throwing blasting process, and a method for predicting the throwing effect after throwing blasting is urgently needed.
Disclosure of Invention
The invention aims to provide a method for predicting the step blasting throwing effect based on a BP neural network, wherein a neural network model after being tested to be qualified can be used for predicting the step blasting throwing condition under a preset working condition by utilizing the generalization capability of the neural network model, judging the feasibility of a blasting scheme and determining whether the throwing requirement is met or not; provides reference for mine step blasting and improves the reliability of throwing blasting.
The technical scheme adopted by the invention is as follows: a method for predicting a step blasting throwing effect based on a BP neural network comprises the following steps:
step one: designing a plurality of groups of test models according to actual working conditions, carrying out tests under different groups, and recording test data;
step two: building a BP neural network model: taking the step height, the coal seam thickness, the goaf upper opening width, the minimum resistance line, the hole pitch, the row pitch, the inter-hole differential time and the coal seam slope angle as input parameters, taking the throwing rate, the pile loosening coefficient and the farthest throwing distance as output parameters, and carrying out normalization treatment;
step three, determining structural parameters of the BP neural network: preprocessing the input parameters and the output parameters after normalization processing, and dividing the preprocessed input parameters and the preprocessed output parameters into a training sample set and a testing sample set; training the BP neural network model by using data in a training sample set, adjusting network parameters of the BP neural network model, and testing the trained BP neural network model by using data in a test data set after the training is successful so as to verify the correctness of the neural network model;
and fourthly, predicting the step blasting throwing condition under the specified working condition by using the trained BP neural network model, judging the feasibility of the blasting scheme, and determining whether the throwing requirement is met.
Specifically, in the first step, each group of test models comprises a throwing blasting step 1, a goaf 2, a reverse pile stacking body 3 and a background lattice plate 4, downward blast holes 1.2 are arranged on the throwing blasting step 1, the simulated coal bed 1.1 is located right below the throwing blasting step 1, and the background lattice plate 4 is erected behind the throwing blasting step 1, the goaf upper opening 2 and the reverse pile stacking body 3.
Specifically, in the first step, the following tests are performed in different groups: and (3) carrying out multiple groups of tests by adjusting one or more than two values of the minimum resistance line, the hole pitch, the row pitch, the differential time between main control row holes, the coal seam slope angle and the specific explosive consumption.
Specifically, the selected BP neural network type is a single hidden layer BP neural network, the BP neural network is trained by data in a training sample set, the data in the training sample set is reordered for each training, and the training is stopped when the error margin or the upper limit of the training times is reached.
Specifically, in the third step, when the trained BP neural network model is tested, the network output is obtained by using a simulation function to perform network model test, and then whether the error between the output and the true value meets the requirement is checked.
The invention has the beneficial effects that:
the program can train the original data of the BP neural network prediction model with the high step throwing blasting effect, dynamically displays the network training process and gives a training result.
2, the performance of the trained network can be tested, and a relative error curve graph representing the performance of the network is given.
3, when the blasting design parameters are given, the blasting effect can be predicted, and the predicted numerical result is displayed.
4, when the actually measured blasting effect is obtained, error analysis can be realized.
And 5, the network can be automatically updated, when a new blasting design and a corresponding trusted blasting effect measured value are obtained, parameters can be correspondingly added into an original model database, and as the use times of the program are increased, the sample database is increased, and the program prediction precision is gradually improved.
And 6, the operation is simple, and the prediction result is reliable.
Drawings
FIG. 1 is a schematic diagram of a general flow chart of an embodiment of the present invention;
FIG. 2 is a diagram of a test model;
FIG. 3 is a schematic diagram of a neural network model building process;
FIG. 4 is a graph of error reduction in the BP network training process;
FIG. 5 is a graph of true versus predicted error;
FIG. 6 is a view of a prediction of the morphology of the detonation stack;
in the figure: 1-throwing blasting steps, 2-goaf, 3-reverse pile stack bodies, 4-background grid plates, 1.1-simulated coal beds and 1.2 downward blast holes.
Detailed Description
The invention will be further described with reference to the drawings and the specific examples.
Example 1: as shown in fig. 1-6, a method for predicting a step blasting throwing effect based on a BP neural network comprises the following steps:
step one: designing a plurality of groups of test models according to actual working conditions (the model is obtained by calculating according to the field engineering geological conditions and parameters in proportion and is shown as a figure 2, the test models are built by concrete, model data are obtained by calculating according to the mine real data in proportion), in the embodiment, obtaining the space geometric parameters of a step surface and a goaf before casting blasting in the middle and the west (two west areas) of a black trench open pit coal mine according to scanner scanning data, carrying out tests under different groups according to the field engineering geological conditions and production requirements, and recording test data;
step two: building a BP neural network model (the implementation process is shown in figure 1): the specific implementation steps are as follows: taking the step height, the coal seam thickness, the goaf upper opening width, the minimum resistance line, the hole pitch, the row pitch, the inter-hole differential time and the coal seam slope angle as input parameters, taking the throwing rate, the pile loosening coefficient and the farthest throwing distance as output parameters, and carrying out normalization treatment;
step three, determining structural parameters of the BP neural network: preprocessing the input parameters and the output parameters after normalization processing, and dividing the preprocessed input parameters and the preprocessed output parameters into a training sample set and a testing sample set; training the BP neural network model by using data in a training sample set (shown in figures 3 and 4), adjusting network parameters of the BP neural network model, and testing the trained BP neural network model by using data in a test data set after the training is successful so as to verify the correctness of the neural network model;
and step four, predicting the step blasting throwing condition under the specified working condition by using the trained BP neural network model (and comparing the predicted step blasting throwing condition with the actual condition to obtain a relative error as shown in figure 5), judging the feasibility of the blasting scheme, and determining whether the throwing requirement is met. And obtaining a detonation stack morphology prediction figure 6.
The relative error of the neural network predictions is shown in fig. 5. As can be seen from the relative error of the predicted values in FIG. 5, the predicted values obtained by the BP network are basically consistent with the test values, and the relative error is within 0.3 percent.
Further, in the first step, each group of test models comprises a throwing blasting step 1, a goaf 2, a reverse pile stacking body 3 and a background grid plate 4, downward blast holes 1.2 are arranged on the throwing blasting step 1, the simulated coal bed 1.1 is located right below the throwing blasting step 1, and the background grid plate 4 is erected behind the throwing blasting step 1, the goaf 2 and the reverse pile stacking body 3.
Further, in the first step, the following tests are performed in different groups: and (3) carrying out multiple groups of tests by adjusting one or more than two values of the minimum resistance line, the hole pitch, the row pitch, the differential time between main control row holes, the coal seam slope angle and the specific explosive consumption.
Further, the selected BP neural network is a single hidden layer BP neural network, the BP neural network is trained by using data in a training sample set, the data in the training sample set is reordered for each training in order to obtain a good learning effect, and the training is stopped when the error margin or the training frequency reaches the upper limit (shown in fig. 4).
Further, in the third step, when the trained BP neural network model is tested, the network output is obtained by using a simulation function to perform network model test, and then whether the error between the output and the true value meets the requirement is checked.
According to the embodiment, the neural network can learn the internal inclusion rule well and make correct predictions on untested working conditions as long as the corresponding mining area space geometric parameters and the like are measured in advance and then are trained by test data by selecting proper neural network types.
The specific embodiments of the present invention have been described in detail with reference to the accompanying drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge of those skilled in the art without departing from the spirit of the present invention.
Claims (1)
1. A method for predicting a step blasting throwing effect based on a BP neural network is characterized by comprising the following steps of: the method comprises the following steps:
step one: designing a plurality of groups of test models according to actual working conditions, carrying out tests under different groups, and recording test data;
step two: building a BP neural network model: taking the step height, the coal seam thickness, the goaf upper mouth width, the minimum resistance line, the hole pitch, the row pitch, the inter-hole differential time and the coal seam slope angle as input parameters, taking the throwing rate, the pile loosening coefficient and the farthest throwing distance as output parameters, and carrying out normalization processing on the input parameters and the output parameters;
step three, determining structural parameters of the BP neural network: preprocessing the input parameters and the output parameters after normalization processing, and dividing the preprocessed input parameters and the preprocessed output parameters into a training sample set and a testing sample set; training the BP neural network model by using data in a training sample set, adjusting network parameters of the BP neural network model, and testing the trained BP neural network model by using data in a test data set after the training is successful so as to verify the correctness of the neural network model;
step four, predicting the step blasting throwing condition under the specified working condition by using the trained BP neural network model, judging the feasibility of the blasting scheme, and determining whether the throwing requirement is met;
in the first step, each group of test models comprises a throwing blasting step (1), a goaf (2), a reverse pile stacking body (3) and a background grid plate (4), downward blast holes (1.2) are arranged on the throwing blasting step (1), a simulated coal bed (1.1) is positioned right below the throwing blasting step (1), and the background grid plate (4) is erected behind the throwing blasting step (1), the goaf (2) and the reverse pile stacking body (3);
in the first step, the following tests are carried out in different groups: carrying out a plurality of groups of tests by adjusting one or more than two values of a minimum resistance line, a hole pitch, a row pitch, a main control row hole differential time, a coal seam slope angle and an explosive unit consumption;
the BP neural network type is a single hidden layer BP neural network, the BP neural network is trained by data in a training sample set, the data in the training sample set is reordered in each training, and the training is stopped when the error margin or the upper limit of the training times is reached.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110881576.3A CN113569487B (en) | 2021-08-02 | 2021-08-02 | BP neural network-based method for predicting step blasting throwing effect |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110881576.3A CN113569487B (en) | 2021-08-02 | 2021-08-02 | BP neural network-based method for predicting step blasting throwing effect |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113569487A CN113569487A (en) | 2021-10-29 |
| CN113569487B true CN113569487B (en) | 2023-08-08 |
Family
ID=78169935
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110881576.3A Active CN113569487B (en) | 2021-08-02 | 2021-08-02 | BP neural network-based method for predicting step blasting throwing effect |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113569487B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115046447A (en) * | 2022-07-19 | 2022-09-13 | 河南省公路工程局集团有限公司 | Multi-row differential roadbed deep hole blasting construction method |
| CN115630257B (en) * | 2022-12-19 | 2023-04-21 | 中南大学 | Blasting hopper volume prediction method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104074543A (en) * | 2014-06-16 | 2014-10-01 | 太原钢铁(集团)有限公司 | Method for processing large-scale underground goaf |
| CN109102109A (en) * | 2018-07-16 | 2018-12-28 | 四川大学 | A kind of piece of exploitation of stone explosion gradation prediction technique |
| 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 |
| CN112541392A (en) * | 2020-11-09 | 2021-03-23 | 北方爆破科技有限公司 | Open bench blasting prediction method based on deep neural network |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11416645B2 (en) * | 2019-02-05 | 2022-08-16 | Dyno Nobel Inc. | Systems for automated blast design planning and methods related thereto |
-
2021
- 2021-08-02 CN CN202110881576.3A patent/CN113569487B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104074543A (en) * | 2014-06-16 | 2014-10-01 | 太原钢铁(集团)有限公司 | Method for processing large-scale underground goaf |
| CN109102109A (en) * | 2018-07-16 | 2018-12-28 | 四川大学 | A kind of piece of exploitation of stone explosion gradation prediction technique |
| 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 |
| CN112541392A (en) * | 2020-11-09 | 2021-03-23 | 北方爆破科技有限公司 | Open bench blasting prediction method based on deep neural network |
Non-Patent Citations (1)
| Title |
|---|
| 郝全明等.BP 神经网络在岩层爆破参数优化中的应用.《煤炭技术》.2014,第33卷(第12期),第2节. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113569487A (en) | 2021-10-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113569487B (en) | BP neural network-based method for predicting step blasting throwing effect | |
| CN111860952B (en) | Method for rapidly optimizing key mining parameters of outburst coal seam | |
| CN107315862B (en) | Method for establishing open cut foundation pit engineering investigation and simulation parameter relationship | |
| CN104989456B (en) | Large-span underground engineering surrounding rock excavation stability monitoring and early warning method | |
| CN111814372B (en) | Blasting block control method combining numerical calculation and block screening | |
| CN106952003B (en) | Method for predicting surrounding rock damage type of underground cavern of stratified rock mass in high ground stress region | |
| CN109738940B (en) | Acoustic emission/microseismic event positioning method under condition of existing empty zone | |
| CN113255175A (en) | Layered soft rock blasting numerical simulation method | |
| CN109252855B (en) | Method and device for determining final cumulative yield of gas well | |
| CN111538071B (en) | Quantitative prediction method for displacement of steep dip stratified rock mass cavern group high side wall | |
| CN112364422A (en) | Shield construction earth surface deformation dynamic prediction method based on MIC-LSTM | |
| CN115660487A (en) | Tunnel smooth blasting comprehensive quality evaluation method | |
| CN115422630A (en) | Method for evaluating dynamic response and damage degree of concrete gravity dam structure under action of explosive load | |
| WO2024124758A1 (en) | Explosive-rock matching parameter optimization method based on residual blasting crater inversion | |
| CN116070907A (en) | Karst collapse susceptibility assessment method and system based on analytic hierarchy process | |
| CN112064617B (en) | Soil-stone mixture foundation quality detection method | |
| CN110119568B (en) | Method for evaluating stone-throwing effect influence factors of riprap bank protection | |
| Reddy | Modeling Load-Settlement Behavior Of Driven Piles Embedded In Cohesive Soils Using Artificial Neural Networks | |
| CN109992841B (en) | Blasting block space omnibearing partition coupling numerical simulation method | |
| CN105718668A (en) | Open pit mine cast blasting effect analysis method | |
| Liang et al. | Optimization of the process parameters of fully mechanized top-coal caving in thick-seam coal using bp neural networks. Sustainability. 2022; 14 (2): 1340 | |
| CN118095020B (en) | Construction method of porous flood diversion gate | |
| CN117150890A (en) | Deep phosphorite rock burst tendency evaluation method based on neural network | |
| CN113378423B (en) | Intelligent inverse analysis quick inversion method and system for urban rail protection MJS construction method piles | |
| CN112214730B (en) | Rock drillability explosiveness judging 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 |