CN116933920A - Prediction and early warning method and system for underground mine debris flow - Google Patents
Prediction and early warning method and system for underground mine debris flow Download PDFInfo
- Publication number
- CN116933920A CN116933920A CN202310756606.7A CN202310756606A CN116933920A CN 116933920 A CN116933920 A CN 116933920A CN 202310756606 A CN202310756606 A CN 202310756606A CN 116933920 A CN116933920 A CN 116933920A
- Authority
- CN
- China
- Prior art keywords
- debris flow
- disaster
- risk
- causing factor
- obtaining
- 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
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000005206 flow analysis Methods 0.000 claims abstract description 28
- 239000011159 matrix material Substances 0.000 claims description 61
- 238000012360 testing method Methods 0.000 claims description 19
- 238000011156 evaluation Methods 0.000 claims description 16
- 238000012549 training Methods 0.000 claims description 14
- 238000010606 normalization Methods 0.000 claims description 13
- 230000008569 process Effects 0.000 claims description 12
- 238000004458 analytical method Methods 0.000 claims description 11
- 238000010276 construction Methods 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 8
- 238000013528 artificial neural network Methods 0.000 claims description 7
- 238000012163 sequencing technique Methods 0.000 claims description 7
- 230000007274 generation of a signal involved in cell-cell signaling Effects 0.000 claims description 3
- 238000007405 data analysis Methods 0.000 claims description 2
- 238000004519 manufacturing process Methods 0.000 abstract description 4
- 239000010410 layer Substances 0.000 description 63
- 239000011435 rock Substances 0.000 description 10
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 6
- 238000011161 development Methods 0.000 description 5
- 241001177040 Tillus Species 0.000 description 4
- 238000012795 verification Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000012937 correction Methods 0.000 description 3
- 229910052742 iron Inorganic materials 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 238000003062 neural network model Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 229910052500 inorganic mineral Inorganic materials 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 239000011707 mineral Substances 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 238000003723 Smelting Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000007596 consolidation process Methods 0.000 description 1
- 230000008094 contradictory effect Effects 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000012634 fragment Substances 0.000 description 1
- 238000012067 mathematical method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000012502 risk assessment Methods 0.000 description 1
- 239000002356 single layer Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000010257 thawing Methods 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
- G06Q50/265—Personal security, identity or safety
Abstract
The application relates to the technical field of mine production safety, and provides a method and a system for predicting and early warning underground debris flow of a mine. Comprising the following steps: obtaining disaster factors which cause the occurrence of underground mine debris flow; disaster-causing factor data are obtained, and classification and assignment are carried out on the disaster-causing factor data to obtain disaster-causing factor levels; constructing a debris flow analysis model based on the disaster factor level; constructing a debris flow prediction model; acquiring real-time acquisition data of disaster factors, and inputting the real-time acquisition data into the debris flow prediction model to acquire a risk prediction result; and grading the risk prediction result according to a preset risk grading rule to obtain a risk prediction level, and generating an early warning signal according to the risk prediction level. The method and the system solve the problem of low debris flow early warning accuracy caused by more influence factors of the debris flow in the mine in the prior art, and improve the prediction early warning method and the system of the debris flow in the mine.
Description
Technical Field
The application relates to the technical field of mine production safety, in particular to a prediction and early warning method and system for underground mine debris flow.
Background
Along with the continuous development of mineral resources, the economic rapid development is brought, and meanwhile, a plurality of safety problems caused by development are also caused, so that the safety of mine workers is seriously influenced, and meanwhile, the development and the utilization of the mineral resources are also hindered. The underground mine mined by adopting the natural caving method is characterized in that a goaf is formed by mining underground ore bodies, movement, dislocation and collapse of the earth surface are caused in the stoping process, along with the increase of mining depth and the continuous expansion of the earth surface collapse range, once flood season is encountered, rainfall in the ore area, surface runoff on the upstream, snow thawing water on high mountain, even landslide mountain bodies and the like are driven to enter the collapse area together with the cover layer (the iceland layer) of tens meters of the earth surface and weathered broken rock fragments, thus the underground debris flow accident is easily caused, and the safety production of the mine is greatly influenced. In China, cheng Chao iron ore, large-smelting iron ore, mershan iron ore and the like have different degrees of debris flow geological disasters. Therefore, the method is very important to fully research and analyze the generation mechanism and disaster factors of the underground debris flow and predict and early warn the mine debris flow.
At present, the research of the occurrence mechanism of the debris flow, the prediction and the early warning and the like is mainly concentrated on the landslide debris flow in the open air, and the research on the debris flow of the underground mine is less. Therefore, through fully researching distribution rules, chemical composition, physical and mechanical properties and the like of the highland mine tillite, on the basis of deeply analyzing the generation factor and disaster-causing mechanism of the tillite consolidation body mud-rock flow, a mud-rock flow prediction and early warning model needs to be established, risk assessment is carried out on the mud-rock flow prediction and early warning model, and guidance is provided for mud-rock flow prevention and control.
In conclusion, the method solves the problem of low accuracy of the mine underground debris flow early warning, and improves the efficiency and accuracy of the mine underground debris flow early warning.
Disclosure of Invention
Based on the above, it is necessary to provide a method and a system for predicting and early warning the underground mine debris flow, which are used for solving the problem of low accuracy of the underground mine debris flow early warning and improving the efficiency and accuracy of the underground mine debris flow early warning.
In a first aspect, an embodiment of the present application provides a method for predicting and early warning a mine underground debris flow, where the method includes: obtaining disaster factors which cause the occurrence of underground mine debris flow, wherein the disaster factors comprise the amount of the tillite, the rainfall and the subsidence area of the earth surface; obtaining disaster causing factor data, grading and assigning the disaster causing factor data to obtain N disaster causing factor levels, wherein the disaster causing factor levels comprise corresponding data grading results, and N is an integer greater than or equal to 1; constructing a debris flow analysis model based on the N disaster causing factor levels, and obtaining disaster causing factor weight, disaster causing factor risk degree and target area risk degree through the debris flow generation analysis model; constructing a debris flow prediction model, and embedding the disaster causing factor weight into the debris flow prediction model; acquiring real-time acquisition data of disaster factors, and inputting the real-time acquisition data into the debris flow prediction model to acquire a risk prediction result; and grading the risk prediction result according to a preset risk grading rule to obtain a risk prediction level, and generating an early warning signal according to the risk prediction level.
In a second aspect, the embodiment of the application also provides a system for predicting and early warning the underground debris flow of the mine, which comprises: the disaster-causing factor obtaining module is used for obtaining disaster-causing factors which cause the occurrence of underground mine debris flow, wherein the disaster-causing factors comprise the amount of tillite, the rainfall and the surface subsidence area; the disaster-causing factor grading module is used for grading and assigning the disaster-causing factor data based on the obtained disaster-causing factor data to obtain N disaster-causing factor levels, wherein the disaster-causing factor levels comprise corresponding data grading results, and N is an integer greater than or equal to 1; the debris flow analysis model construction module is used for constructing a debris flow analysis model based on the N disaster causing factor levels, and obtaining disaster causing factor weight, disaster causing factor risk and target area risk through the debris flow generation analysis model; the debris flow prediction model construction module is used for constructing a debris flow prediction model and embedding the disaster-causing factor weight into the debris flow prediction model; the risk prediction result obtaining module is used for obtaining disaster factor real-time acquisition data, inputting the real-time acquisition data into the debris flow prediction model and obtaining a risk prediction result; the early warning signal generation module is used for classifying the risk prediction result according to a preset risk classification rule to obtain a risk prediction level and generating an early warning signal according to the risk prediction level.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
collecting disaster-causing factors which cause underground debris flow of a mine, grading and assigning disaster-causing factor data to obtain N disaster-causing factor levels, constructing a debris flow analysis model and a debris flow prediction model based on the disaster-causing factor levels, acquiring real-time acquisition data of the disaster-causing factors, inputting the real-time acquisition data into the debris flow prediction model to obtain a risk prediction result, dividing the risk prediction result according to a preset risk grade rule, generating early warning signals according to the risk prediction level, analyzing early warning information of the debris flow from a plurality of influencing factors, and improving the early warning efficiency and accuracy of the underground debris flow of the mine.
The foregoing description is only an overview of the present application, and is intended to be implemented in accordance with the teachings of the present application in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present application more readily apparent.
Drawings
FIG. 1 is a schematic flow chart of a method for predicting and warning the flow of underground debris in a mine according to an embodiment;
fig. 2 is a schematic flow chart of a mud-rock flow prediction model constructed in a method for predicting and pre-warning the underground mud-rock flow of a mine in an embodiment;
FIG. 3 is a block diagram of a predictive early warning system for mine downhole debris flow in one embodiment;
reference numerals illustrate: the system comprises a disaster factor obtaining module 11, a disaster factor grading module 12, a debris flow analysis model construction module 13, a debris flow prediction model construction module 14, a risk prediction result obtaining module 15 and an early warning signal generating module 16.
Detailed Description
The application provides a method and a system for predicting and early warning of underground debris flow of a mine, which are used for solving the technical problem of low accuracy of debris flow early warning in the prior art.
Having introduced the basic principles of the present application, the technical solutions of the present application will now be clearly and fully described with reference to the accompanying drawings, it being apparent that the embodiments described are only some, but not all, embodiments of the present application, and it is to be understood that the present application is not limited to the exemplary embodiments described herein. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application. It should be further noted that, for convenience of description, only some, but not all of the drawings related to the present application are shown.
Example 1
As shown in fig. 1, the application provides a method for predicting and pre-warning underground mine debris flow, which comprises the following steps:
s100: obtaining disaster factors which cause the occurrence of underground mine debris flow, wherein the disaster factors comprise the amount of the tillite, the rainfall and the subsidence area of the earth surface;
s200: obtaining disaster causing factor data, grading and assigning the disaster causing factor data to obtain N disaster causing factor levels, wherein the disaster causing factor levels comprise corresponding data grading results, and N is an integer greater than or equal to 1;
specifically, the disaster-causing factors are factors causing the occurrence of the debris flow, wherein the disaster-causing factors specifically refer to the amount of the tillite, the rainfall and the surface subsidence area, and the disaster-causing factors causing the occurrence of the mine underground debris flow can be known from the statements, and include the amount of the tillite, the rainfall and the surface subsidence area; the disaster causing factor data is obtained, the disaster causing factor data can be searched for related debris flow events based on a big data technology for inquiry, the disaster causing factor data is obtained, the disaster causing factor data comprises the content of the tillus, the rainfall and the surface subsidence area when debris flow occurs, the disaster causing factor data is classified and assigned, wherein the rule of classification and assignment can be set by a person skilled in the art based on actual condition in a self-defining way; n disaster causing factor levels can be obtained according to the disaster causing factor data classification, wherein the disaster causing factor levels comprise corresponding disaster causing factor data classification results, for example, N can be set to be 5, five disaster causing factor levels (1, 2,3, 4 and 5) are considered, wherein the index influence degree is small and the underground debris flow risk is small by using 1, and the index influence degree is large by using 5. By acquiring the data of the disaster causing factors and grading and assigning the disaster causing factors, the data influencing the debris flow is obtained, and data support is provided for the establishment of a debris flow analysis model.
And obtaining disaster causing factor data for causing the occurrence of underground mine debris flow, wherein the disaster causing factor data comprise the amount of the tillus material, the rainfall and the surface subsidence area, and the thickness of the surface tillus material is adopted instead because the amount of the tillus material is difficult to measure specifically.
The method comprises the steps of grading indexes, establishing a hierarchical structure of a debris flow analysis model through an analytic hierarchy process AHP, taking underground debris flow risk evaluation as a target layer, taking the thickness of the moraine, the surface collapse area and the rainfall (three cumulative days) which cause underground debris flow to occur as criterion layers, and taking five risk levels as scheme layers, wherein the five risk levels comprise high risk, general high risk, low risk and low risk.
S300: constructing a debris flow analysis model based on the N disaster causing factor levels, and obtaining disaster causing factor weight, disaster causing factor risk degree and target area risk degree through the debris flow generation analysis model;
specifically, a debris flow analysis model is constructed based on the N disaster causing factors, wherein the debris flow analysis model is a decision method based on a hierarchical analysis method, wherein evaluation indexes are classified after data standardization treatment, and elements related to the debris flow analysis model are decomposed into layers of target criteria, schemes and the like, and qualitative and quantitative analysis is performed on the basis of the analysis methods; the data obtained through analysis of the debris flow analysis model constructed by the disaster causing factor level has disaster causing factor weight, disaster causing factor risk and target area risk, namely the disaster causing factor weight refers to the weight ratio of each disaster causing factor in the target area risk, the disaster causing factor risk is obtained through layered calculation by judging rectangles, the disaster causing factor risk can be evaluated through an expert system and refers to the importance of the disaster causing factor to the disaster, and the target area risk refers to the probability and scale degree of causing the occurrence of the debris flow in the target area, namely the risk degree of the occurrence of the debris flow in the target area. Data support is provided for calculating weights later.
Further, the method comprises the following steps:
s310: the debris flow analysis model comprises a target layer, a criterion layer and a scheme layer;
s320: taking underground debris flow risk evaluation as a target layer and taking the disaster causing factors as a criterion layer;
s330: five risk levels are taken as scheme layers, wherein the five risk levels comprise a large risk level, a general large risk level, a small risk level and a small risk level;
s340: and constructing an hierarchical structure of the debris flow analysis model based on the target layer, the criterion layer and the scheme layer.
Specifically, a complex objective decision problem, namely how to analyze the influence degree of the debris flow based on disaster factors, is taken as a system, the objective is decomposed into a plurality of objectives or criteria, and the relevant calculation is carried out through qualitative indexes to be taken as a system method for multi-scheme optimization decision basis. Taking underground debris flow risk evaluation as a target layer, taking the amount of tillite, the rainfall and the surface subsidence area which cause the occurrence of underground debris flow as criterion layers, and taking five risk levels as scheme layers, wherein the five risk levels comprise high risk, general risk, low risk and low risk, and establishing an underground debris flow evaluation hierarchical structure based on the target layer, the criterion layers and the scheme layers; and constructing the debris flow analysis model based on the analysis model.
Further, the method comprises the following steps:
s350: constructing a debris flow expert database, evaluating the disaster-causing factors through the debris flow expert database, and determining the relative important coefficients of the disaster-causing factors;
s360: obtaining a judgment matrix of the target layer and the criterion layer based on the relative importance coefficient;
s370: performing hierarchical order sorting according to the judgment matrix to obtain a hierarchical order sorting result;
further, the method comprises the following steps:
s371: normalizing each column of elements of the judgment matrix to obtain a normalized judgment matrix;
s372: adding the normalization judgment matrix according to the number of rows of the matrix to obtain a matrix characteristic vector;
s373: and carrying out normalization processing on the matrix feature vector to obtain the hierarchical single-ordering result.
Specifically, the average geometry calculation and normalization are performed on the judgment matrix to obtain each index weight, and the target layer weight vector is recorded as
W＝(i 1 ，i 2 ，i 3 ) T
Wherein i is 1 For evaluating the weight value of the tillite object, i 2 I is a rainfall weight value for the evaluation index 3 And (5) evaluating the index surface subsidence area weight value. Normalizing each column of elements of the judgment matrix according to a formula to obtain a normalized judgment matrix; adding the normalization judgment matrix according to the number of matrix rows to obtain a matrix characteristic vector, wherein w is obtained i I.e. the feature vector obtained, vector w i Normalization is also the hierarchical single ordering result of the judgment matrix, namely the weight coefficient.
S380: and carrying out consistency check on the judgment matrix, and taking the hierarchical single sequencing result as the disaster causing factor weight after the consistency check is passed.
Specifically, by consulting the related expert scholars and combining the knowledge of the occurrence and development characteristics of the underground debris flow in the evaluation area in recent years, an artificial intelligence and database combined debris flow expert database is constructed, wherein disaster factor data, area danger level data and the like are included, a large amount of information of underground debris flow of the mine is stored, and the debris flow expert database can be updated through continuous learning. Through the mud-rock flow expert database pairThe disaster factor is evaluated, the relative importance coefficient of the disaster factor is determined, the accuracy of the relative importance coefficient can be improved through the evaluation of an expert system, the relative importance coefficient is determined through the expert system, n factors of a criterion layer, such as the content of the fringed object, the rainfall and the surface subsidence area, are determined, the influence degree of the factors on the debris flow is required to be compared, the proportion of the factors in the layer relative to a certain criterion is determined, the comparison of the factors is difficult, so that the comparison between two indexes is carried out firstly, and the two elements have the same importance scale of 1 to a certain attribute; the two elements are compared, the former element is slightly more important than the latter element by a scale of 3, the obvious important scale is 5, the much more important scale is 7, and the extremely important scale is 9; the importance of the former element is compared with the latter element, and the importance of the latter element is between calibrated standards and is scaled to be 2, 4, 6 and 8; the two elements are inversely compared to scale 1 /a ij Obtaining a judgment matrix of the target layer and the criterion layer in the evaluation area; performing hierarchical sorting on the judgment matrixes by using a certain mathematical method, wherein the hierarchical single sorting refers to the relative weight of each factor of each judgment matrix aiming at the criterion of the judgment matrix, and essentially calculating a weight coefficient to obtain a hierarchical single sorting result; and carrying out consistency test on the judgment matrix result, checking whether the random consistency ratio CR is smaller than 0.1, and verifying whether the hierarchical ordering result is reasonable. Based on the method, data support is provided for early warning of improving the underground debris flow of the mine.
Further, the method comprises the following steps:
s381: obtaining a consistency index calculation formula:
wherein CI represents a consistency index, lambda max Representing the maximum characteristic value, wherein n represents the number of disaster causing factors in the judgment matrix;
s382: obtaining an average random consistency index RI according to the specific value of n;
s383: obtaining a consistency proportion formula:
wherein CR represents a consistency ratio, CI represents a consistency index, and RI represents an average random consistency index;
s384: obtaining a consistency ratio through the consistency ratio formula, and judging that the consistency test is passed when the consistency ratio is smaller than 0.1;
s385: when the consistency ratio is greater than or equal to 0.1, the consistency check of the judgment matrix is not passed, and the judgment matrix needs to be adjusted.
Specifically, the consistency check is performed on the result of the judgment matrix, and since there is a possibility that the values in the judgment matrix are contradictory, it is necessary to correct the result by the consistency check, firstly, the consistency index is calculated according to the consistency index formula, CI represents the consistency index, lambda max Representing the maximum characteristic value, n represents the number of disaster causing factors in the judging matrix, obtaining an average random consistency index RI according to the specific value of n, wherein the average random consistency index is obtained by repeatedly carrying out random judging matrix characteristic root calculation for more than 500 times and then taking an arithmetic average value, inquiring an RI table according to the requirements of technicians, and then calculating a consistency ratio according to a consistency ratio formula, wherein CR represents the consistency ratio, CI represents the consistency index, and RI represents the average random consistency index; after calculation by the formula, if the consistency ratio is smaller than 0.1, judging that the consistency test passes, and using the calculated weight in the hierarchical single sequencing; if the consistency ratio is greater than 0.1, judging that the consistency test is not passed, and adjusting the judgment matrix to adjust the consistency matrix so as to make each row of the consistency matrix have a multiple relation. The data is guaranteed for constructing the debris flow analysis model, and higher accuracy is provided for the accuracy of the debris flow analysis model.
The disaster factor is evaluated by an expert, and the target layer and the criterion are constructed for the thickness of the tillite, the subsidence area of the ground surface and the rainfallA judgment matrix of the layer.
The criterion layer B is calculated based on the feature vector W= [0.469,0.806,2.644] of the target layer C, and the maximum feature root lambda max= 3.061
According to the consistency check formulaThe consistency index CI is found to be 0.031.
Obtaining an average random uniformity index RI of 0.58 according to n=3, wherein the RI reference table is
| n | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 |
| RI | 0 | 0 | 0.58 | 0.89 | 1.12 | 1.26 | 1.36 | 1.41 | 1.46 | 1.49 | 1.53 | 1.54 |
According to the consistency proportion formulaThe consistency ratio CR is 0.053
The random consistency check cr=0.053 <0.1 for matrix C-B, which indicates that the C-B matrix is consistent, so the matrix eigenvector Wi is meaningful.
And (3) carrying out normalization treatment on the characteristic vector W= [0.469,0.806,2.644] to obtain weights of three disaster causing factors of the amount of the moraine, the surface subsidence area and the rainfall, wherein the weights are respectively 0.12, 0.21 and 0.67.
S400: constructing a debris flow prediction model, and embedding the disaster causing factor weight into the debris flow prediction model;
specifically, a debris flow prediction model is constructed through a machine learning neural network algorithm, the model comprises a BP neural network model of an input layer, an hidden layer and an output layer, the learning process of the model is embodied as forward propagation of signals and backward propagation of errors, and is specifically embodied as sample input which is processed through the hidden layer and transmitted to the output layer, if a result output by the output layer has a large difference from an expected result, the backward propagation errors are subjected to adjustment of weight values of all layers so as to adjust the output precision, and meanwhile, all evaluation index weights obtained through the hierarchical analysis method are used as initial weights of BP neural network input indexes. Training the mud-rock flow prediction model by using the collected and standardized data set, and obtaining the mud-rock flow prediction model when the output result of the model tends to be in a convergence state.
Further, the method comprises the following steps:
s410: based on a BP neural network, constructing a network structure of the debris flow prediction model, wherein the debris flow prediction model comprises an input layer, an hidden layer and an output layer;
s420: taking the disaster causing factor weight as the initial weight of each input index of the input layer of the debris flow prediction model;
s430: constructing a sample data set according to the disaster causing factor risk and the target area risk, and carrying out standardized processing on the sample data set to obtain a standard sample data set;
s440: training the debris flow prediction model through the standard sample data set, introducing momentum factors in the training process to conduct supervised learning, and obtaining the debris flow prediction model when the model output result tends to be in a convergence state.
Specifically, the process of constructing the debris flow prediction model is as follows: based on BP neural network in machine learning, constructing a BP neural network model comprising an input layer, an hidden layer and an output layer, namely a network structure of the debris flow prediction model, wherein the debris flow prediction model can form parameters such as weight, threshold and the like connected between simple units in the supervision training process, and the trained debris flow prediction model can carry out complex nonlinear logic operation according to input data to output debris flow prediction information obtained by prediction; the learning process of the model is embodied by forward propagation of signals and backward propagation of errors, and is embodied by processing sample input through an implicit layer and transmitting the sample input to an output layer, and if the output result of the output layer has a large difference from an expected result, the backward propagation errors are used for adjusting the weight value of each layer so as to adjust the output precision. Setting an implicit layer as a single-layer neural network model, so as to realize the separability of nonlinear data, and simultaneously adopting each evaluation index weight obtained by a analytic hierarchy process as each input index initial weight of a BP neural network input layer; collecting the calculated tillite object quantity risk, rainfall risk, surface subsidence area risk and evaluation area risk to construct a sample data set, and carrying out standardized processing on the sample data set according to the following rules:
obtaining a normalized sample dataset, wherein R i For the normalized risk level (i=1, 2,3;1 represents the amount of the moraine, 2 represents the amount of rainfall, 3 represents the area of subsidence of the ground), w i Assigning a weight value, F to the risk degree, F max A maximum risk value F given to the index min The minimum risk value assigned to the index.
Training the debris flow prediction model through the standardized data set, introducing a momentum factor in the training process to conduct supervised learning BP algorithm, wherein the algorithm adjusts the weight rule to control the correction quantity by introducing the momentum factor alpha (0-1), so as to play a role in reducing oscillation, and ensure that the correction direction always goes towards the convergence direction, and the correction formula is as follows:
where w (k) represents a single connection weight vector, and may represent a single connection weight coefficient,is a negative gradient at time k>For the secondary gradient at the moment k-1, eta is the learning rate (eta > 0), alpha is the momentum factor, and 0 is less than or equal to alpha r < 1. Model convergence generally means that the training and verification loss curve does not have large fluctuation, and the fluctuation can still be within a certain tolerance range along with the continuous increase of the training round number, which means that the loss function is reduced to the minimum, and the debris flow prediction model can be obtained.
Further, the method comprises the following steps:
s450: based on big data, acquiring historical occurrence data of underground debris flow;
s460: acquiring historical disaster causing factor risk and historical target area risk according to the historical occurrence data, and constructing a sample test set according to the historical disaster causing factor risk and the historical target area risk;
s470: testing the debris flow prediction model through the sample test set;
s480: and presetting model test indexes, and obtaining the debris flow prediction model when the accuracy of the output result of the debris flow prediction model meets the preset model test indexes.
Specifically, based on the information collection of big data, the historical occurrence data of the underground debris flow is collected, the data is analyzed, the historical sample data is obtained, the historical sample data comprise the collected and calculated historical tillite object quantity risk, historical rainfall risk, historical ground surface collapse area risk and historical evaluation area risk, and the historical sample data are further subjected to data marking and are divided according to a certain proportion to obtain a training set, a verification set and a test set; the method comprises the steps of inputting a plurality of sample data in a training set into a debris flow prediction model, performing supervision training on an output result of the debris flow prediction model by using sample parameters in a verification set, enabling the output scheme of the debris flow prediction model to be consistent with the risk of a sample evaluation area, performing accuracy testing on the debris flow prediction model by using a test set after the data in the training set are trained, respectively inputting the risk of a plurality of sample historical ice levels, the risk of historical rainfall and the risk of historical earth surface subsidence area into the model, obtaining a plurality of prediction schemes as actual schemes, taking a plurality of sample prediction schemes corresponding to the input data in the test set as expected outputs, calculating errors between the actual outputs and the expected outputs, performing gradient descent updating on control parameters, and simply speaking, taking the errors between the actual outputs and the expected outputs as loss functions, wherein the smaller the description errors are, so that the debris flow prediction model with the accuracy meeting preset conditions can be obtained, and the accuracy of the risk prediction result of the debris flow prediction model is improved.
And carrying out disaster factor risk assignment on mine data to be evaluated, wherein the assignment standard is as follows:
| level of | The risk is small | Less risk | Risk of general | The risk is large | High risk |
| Score of | 1 | 2 | 3 | 4 | 5 |
And (3) assigning a risk degree to disaster-causing factor data of a target area to be evaluated as follows:
after the target area disaster factor risk is standardized, substituting the target area disaster factor risk into the constructed debris flow prediction model based on the BP neural network, and predicting the obtained target area risk as follows:
after the risk level of the target area to be evaluated is obtained, an early warning signal can be generated according to the risk level, so that the safety production of the mine is ensured.
For verification of the prediction model, among the 8 evaluated area data, the area A, the area C and the area E adopt the data of the area where the debris flow disaster occurs, and the prediction grade given by the prediction model also shows that the risk grade of the three areas is IV grade (high risk) or more, which fully explains the accuracy and applicability of the built underground debris flow prediction early warning model.
S500: acquiring real-time acquisition data of disaster factors, and inputting the real-time acquisition data into the debris flow prediction model to acquire a risk prediction result;
s600: and grading the risk prediction result according to a preset risk grading rule to obtain a risk prediction level, and generating an early warning signal according to the risk prediction level.
Specifically, acquiring real-time acquisition data of disaster factors, namely the tillite content, the rainfall and the earth surface subsidence surface, and inputting the real-time acquisition data into the debris flow prediction model to obtain a risk prediction result; and classifying the risk prediction result into the preset risk classification rule to obtain a risk prediction level, generating an early warning signal of the underground mine debris flow according to the risk prediction level, analyzing and calculating early warning information of the debris flow through a plurality of influence factors, and improving the efficiency and accuracy of the prediction early warning of the underground mine debris flow.
Further, the method comprises the following steps:
s610: acquiring a plurality of historical target area dangers according to the historical occurrence data;
s620: and inputting the dangers of the historical target areas into a debris flow prediction and early warning system for data analysis to obtain the preset dangers grading rule, wherein the dangers grading rule comprises dangers and corresponding dangers thresholds.
Specifically, according to the occurrence condition of underground mine debris flow searched and inquired by the big data technology, acquiring historical occurrence data, wherein the risk degree of a historical target area refers to the degree of disaster occurrence of the historical target area, and the historical occurrence data are acquired by the same analytic hierarchy process; inputting the dangers of the historical target areas into a debris flow prediction and early warning system, and intelligently dividing the areas to divide the areas into preset dangers, wherein the dangers comprise dangers and corresponding dangers, and generating early warning signals, so that the prediction and early warning of the underground debris flow of the mine are realized.
Example two
Based on the same inventive concept as the method for predicting and pre-warning the underground mine debris flow in the foregoing embodiment, as shown in fig. 3, the application further provides a system for predicting and pre-warning the underground mine debris flow, wherein the system comprises:
the disaster-causing factor obtaining module 11 is used for obtaining disaster-causing factors which cause the occurrence of underground mine debris flow, wherein the disaster-causing factors comprise the amount of tillite, the rainfall and the subsidence area of the earth surface;
the disaster-causing factor grading module 12 is used for obtaining disaster-causing factor data, grading and assigning the disaster-causing factor data to obtain N disaster-causing factor levels, wherein the disaster-causing factor levels comprise corresponding data grading results, and N is an integer greater than or equal to 1;
the debris flow analysis model construction module 13 is used for constructing a debris flow analysis model according to the N disaster causing factor levels, and obtaining disaster causing factor weight, disaster causing factor risk and target area risk through the debris flow generation analysis model;
the debris flow prediction model construction module 14, wherein the debris flow prediction model construction module 14 is used for constructing a debris flow prediction model and embedding the disaster-causing factor weight into the debris flow prediction model;
the risk prediction result obtaining module 15 is used for obtaining disaster-causing factors, acquiring data in real time, inputting the acquired data in real time into the debris flow prediction model, and obtaining a risk prediction result;
the early warning signal generation module 16 is configured to rank the risk prediction result according to a preset risk ranking rule, obtain a risk prediction level, and generate an early warning signal according to the risk prediction level.
Further, the embodiment of the application comprises the following steps:
the debris flow analysis model module is used for the debris flow analysis model and comprises a target layer, a criterion layer and a scheme layer;
the underground debris flow analytic hierarchy process module is used for taking underground debris flow risk evaluation as a target layer and taking the disaster causing factors as a criterion layer;
a risk level module for taking five risk levels as scheme layers, wherein the five risk levels comprise high risk, general risk, low risk and low risk;
and the hierarchical structure module is used for constructing a hierarchical structure of the debris flow analysis model based on the target layer, the criterion layer and the scheme layer.
Further, the embodiment of the application further comprises:
the debris flow expert database construction module is used for constructing a debris flow expert database, evaluating the disaster factors through the debris flow expert database and determining the relative important coefficients of the disaster factors;
the judgment matrix acquisition module is used for acquiring judgment matrices of the target layer and the criterion layer based on the relative importance coefficients;
the hierarchical single sequencing result obtaining module is used for performing hierarchical single sequencing according to the judgment matrix to obtain a hierarchical single sequencing result;
and the disaster causing factor weight acquisition module is used for carrying out consistency check on the judgment matrix, and taking the hierarchical single sequencing result as the disaster causing factor weight after the consistency check is passed.
Further, the embodiment of the application further comprises:
the normalization judgment matrix obtaining module is used for carrying out normalization processing on each column of elements of the judgment matrix to obtain a normalization judgment matrix;
the matrix characteristic vector obtaining module is used for adding the normalization judgment matrix according to the number of matrix rows to obtain a matrix characteristic vector;
the hierarchical single-order result obtaining module is used for carrying out normalization processing on the matrix eigenvectors to obtain the hierarchical single-order result.
Further, the embodiment of the application further comprises:
the consistency index calculation formula obtaining module is used for obtaining consistency
The performance index calculation formula:
wherein CI represents a consistency index, lambda max Representing the maximum characteristic value, wherein n represents the number of disaster causing factors in the judgment matrix;
the average random consistency index RI obtaining module is used for obtaining the average random consistency index RI according to the specific value of n;
the consistency proportion formula obtaining module is used for obtaining consistency proportion formulas
The formula:
wherein CR represents a consistency ratio, CI represents a consistency index, and RI represents an average random consistency index;
the consistency check judging module is used for obtaining a consistency ratio through the consistency ratio formula, and judging that the consistency check passes when the consistency ratio is smaller than 0.1;
and the judging matrix adjusting module is used for judging that the consistency check is not passed when the consistency ratio is more than or equal to 0.1, and adjusting the judging matrix.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples illustrate only a few embodiments of the application, which are described in detail and are not to be construed as limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of protection of the present application is to be determined by the appended claims.
Claims (9)
1. The method for predicting and early warning the underground debris flow of the mine is characterized by comprising the following steps of:
obtaining disaster factors which cause the occurrence of underground mine debris flow, wherein the disaster factors comprise the amount of the tillite, the rainfall and the subsidence area of the earth surface;
obtaining disaster causing factor data, grading and assigning the disaster causing factor data to obtain N disaster causing factor levels, wherein the disaster causing factor levels comprise corresponding data grading results, and N is an integer greater than or equal to 1;
constructing a debris flow analysis model based on the N disaster causing factor levels, and obtaining disaster causing factor weight, disaster causing factor risk degree and target area risk degree through the debris flow generation analysis model;
constructing a debris flow prediction model, and embedding the disaster causing factor weight into the debris flow prediction model;
acquiring real-time acquisition data of disaster factors, and inputting the real-time acquisition data into the debris flow prediction model to acquire a risk prediction result;
and grading the risk prediction result according to a preset risk grading rule to obtain a risk prediction level, and generating an early warning signal according to the risk prediction level.
2. The method as recited in claim 1, further comprising:
the debris flow analysis model comprises a target layer, a criterion layer and a scheme layer;
taking underground debris flow risk evaluation as a target layer and taking the disaster causing factors as a criterion layer;
five risk levels are taken as scheme layers, wherein the five risk levels comprise high risk, general risk, low risk and low risk;
and constructing an hierarchical structure of the debris flow analysis model based on the target layer, the criterion layer and the scheme layer.
3. The method as recited in claim 2, further comprising:
constructing a debris flow expert database, evaluating the disaster-causing factors through the debris flow expert database, and determining the relative important coefficients of the disaster-causing factors;
obtaining a judgment matrix of the target layer and the criterion layer based on the relative importance coefficient;
performing hierarchical order sorting according to the judgment matrix to obtain a hierarchical order sorting result;
and carrying out consistency check on the judgment matrix, and taking the hierarchical single sequencing result as the disaster causing factor weight after the consistency check is passed.
4. The method of claim 3, wherein said hierarchically ordered list according to said decision matrix further comprises:
normalizing each column of elements of the judgment matrix to obtain a normalized judgment matrix;
adding the normalization judgment matrix according to the number of rows of the matrix to obtain a matrix characteristic vector;
and carrying out normalization processing on the matrix feature vector to obtain the hierarchical single-ordering result.
5. The method of claim 3, wherein said performing a consistency check on said decision matrix further comprises:
obtaining a consistency index calculation formula:
wherein CI represents a consistency index, lambda max Representing the maximum characteristic value, wherein n represents the number of disaster causing factors in the judgment matrix;
obtaining an average random consistency index RI according to the specific value of n;
obtaining a consistency proportion formula:
wherein CR represents a consistency ratio, CI represents a consistency index, and RI represents an average random consistency index;
obtaining a consistency ratio through the consistency ratio formula, and judging that the consistency test is passed when the consistency ratio is smaller than 0.1;
when the consistency ratio is greater than or equal to 0.1, the consistency check is judged to be failed, and the judgment matrix is required to be adjusted.
6. The method of claim 4, wherein the constructing a debris flow prediction model further comprises:
based on a BP neural network, constructing a network structure of the debris flow prediction model, wherein the debris flow prediction model comprises an input layer, an hidden layer and an output layer;
taking the disaster causing factor weight as the initial weight of each input index of the input layer of the debris flow prediction model;
constructing a sample data set according to the disaster causing factor risk and the target area risk, and carrying out standardized processing on the sample data set to obtain a standard sample data set;
training the debris flow prediction model through the standard sample data set, introducing momentum factors in the training process to conduct supervised learning, and obtaining the debris flow prediction model when the model output result tends to be in a convergence state.
7. The method as recited in claim 5, further comprising:
based on big data, acquiring historical occurrence data of underground debris flow;
acquiring historical disaster causing factor risk and historical target area risk according to the historical occurrence data, and constructing a sample test set according to the historical disaster causing factor risk and the historical target area risk;
testing the debris flow prediction model through the sample test set;
and presetting model test indexes, and obtaining the debris flow prediction model when the accuracy of the output result of the debris flow prediction model meets the preset model test indexes.
8. The method as recited in claim 6, further comprising:
acquiring a plurality of historical target area dangers according to the historical occurrence data;
and inputting the dangers of the historical target areas into a debris flow prediction and early warning system for data analysis to obtain the preset dangers grading rule, wherein the dangers grading rule comprises dangers and corresponding dangers thresholds.
9. A predictive early warning system for mine downhole debris flow, the system comprising:
the disaster-causing factor obtaining module is used for obtaining disaster-causing factors which cause the occurrence of underground mine debris flow, wherein the disaster-causing factors comprise the amount of tillite, the rainfall and the surface subsidence area;
the disaster-causing factor grading module is used for grading and assigning the disaster-causing factor data based on the obtained disaster-causing factor data to obtain N disaster-causing factor levels, wherein the disaster-causing factor levels comprise corresponding data grading results, and N is an integer greater than or equal to 1;
the debris flow analysis model construction module is used for constructing a debris flow analysis model based on the N disaster causing factor levels, and obtaining disaster causing factor weight, disaster causing factor risk and target area risk through the debris flow generation analysis model;
the debris flow prediction model construction module is used for constructing a debris flow prediction model and embedding the disaster-causing factor weight into the debris flow prediction model;
the risk prediction result obtaining module is used for obtaining disaster factor real-time acquisition data, inputting the real-time acquisition data into the debris flow prediction model and obtaining a risk prediction result;
the early warning signal generation module is used for classifying the risk prediction result according to a preset risk classification rule to obtain a risk prediction level and generating an early warning signal according to the risk prediction level.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310756606.7A CN116933920A (en) | 2023-06-26 | 2023-06-26 | Prediction and early warning method and system for underground mine debris flow |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310756606.7A CN116933920A (en) | 2023-06-26 | 2023-06-26 | Prediction and early warning method and system for underground mine debris flow |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116933920A true CN116933920A (en) | 2023-10-24 |
Family
ID=88381854
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310756606.7A Pending CN116933920A (en) | 2023-06-26 | 2023-06-26 | Prediction and early warning method and system for underground mine debris flow |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116933920A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117745466A (en) * | 2024-02-21 | 2024-03-22 | 中国有色金属工业昆明勘察设计研究院有限公司 | Tailing pond counting intelligent operation and maintenance system |
-
2023
- 2023-06-26 CN CN202310756606.7A patent/CN116933920A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117745466A (en) * | 2024-02-21 | 2024-03-22 | 中国有色金属工业昆明勘察设计研究院有限公司 | Tailing pond counting intelligent operation and maintenance system |
| CN117745466B (en) * | 2024-02-21 | 2024-04-26 | 中国有色金属工业昆明勘察设计研究院有限公司 | Tailing pond counting intelligent operation and maintenance system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Dodangeh et al. | Novel hybrid intelligence models for flood-susceptibility prediction: Meta optimization of the GMDH and SVR models with the genetic algorithm and harmony search | |
| CN109241627B (en) | Probability grading dynamic support method and device for automatically designing support scheme | |
| KR100982448B1 (en) | Ground subsidence prediction system and predicting method using the same | |
| CN115688404B (en) | Rainfall landslide early warning method based on SVM-RF model | |
| CN116108758B (en) | Landslide susceptibility evaluation method | |
| CN112949202A (en) | Bayesian network-based rockburst probability prediction method | |
| CN109934398A (en) | A kind of drill bursting construction tunnel gas danger classes prediction technique and device | |
| CN112966722A (en) | Regional landslide susceptibility prediction method based on semi-supervised random forest model | |
| CN113326660B (en) | Tunnel surrounding rock extrusion deformation prediction method based on GA-XGboost model | |
| CN111339478B (en) | Meteorological data quality assessment method based on improved fuzzy analytic hierarchy process | |
| CN116933920A (en) | Prediction and early warning method and system for underground mine debris flow | |
| CN108280289B (en) | Rock burst danger level prediction method based on local weighted C4.5 algorithm | |
| CN115130375A (en) | Rock burst intensity prediction method | |
| CN110633504A (en) | Prediction method for coal bed gas permeability | |
| CN117009735A (en) | High-strength forest fire occurrence probability calculation method combining BiLSTM and nuclear density estimation | |
| CN113723446A (en) | Ground disaster monitoring and early warning method and device, computer equipment and storage medium | |
| CN115980826A (en) | Rock burst intensity prediction method based on weighted meta-heuristic combined model | |
| KR20200052398A (en) | Method and apparatus for landslide susceptibility mapping using machine-learning architecture | |
| CN117035418A (en) | Tunnel construction comprehensive risk evaluation method and device based on multi-source data fusion | |
| CN112200356A (en) | Landslide prediction method, device, equipment and storage medium | |
| CN111144637A (en) | Regional power grid geological disaster forecasting model construction method based on machine learning | |
| Xue et al. | PREDICTION OF SLOPE STABILITY BASED ON GA-BP HYBRID ALGORITHM. | |
| CN116384627A (en) | Geological disaster evaluation method based on machine learning | |
| CN117196350A (en) | Mine geological environment characteristic monitoring and recovery treatment method and system | |
| CN116090696A (en) | Landslide geological disaster risk classification prediction method suitable for mountain railway line |
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 |