CN111678683B - Method and device for predicting roof pressure of intelligent fully mechanized coal mining face of coal mine - Google Patents

Method and device for predicting roof pressure of intelligent fully mechanized coal mining face of coal mine Download PDF

Info

Publication number
CN111678683B
CN111678683B CN202010510637.0A CN202010510637A CN111678683B CN 111678683 B CN111678683 B CN 111678683B CN 202010510637 A CN202010510637 A CN 202010510637A CN 111678683 B CN111678683 B CN 111678683B
Authority
CN
China
Prior art keywords
pressure
working
resistance
support
safety valve
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
Application number
CN202010510637.0A
Other languages
Chinese (zh)
Other versions
CN111678683A (en
Inventor
程敬义
闫万梓
孙鑫
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China University of Mining and Technology CUMT
Original Assignee
China University of Mining and Technology CUMT
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by China University of Mining and Technology CUMT filed Critical China University of Mining and Technology CUMT
Priority to CN202010510637.0A priority Critical patent/CN111678683B/en
Publication of CN111678683A publication Critical patent/CN111678683A/en
Application granted granted Critical
Publication of CN111678683B publication Critical patent/CN111678683B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01MTESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M13/00Testing of machine parts
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N3/00Investigating strength properties of solid materials by application of mechanical stress
    • G01N3/02Details
    • G01N3/04Chucks
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N3/00Investigating strength properties of solid materials by application of mechanical stress
    • G01N3/32Investigating strength properties of solid materials by application of mechanical stress by applying repeated or pulsating forces
    • G01N3/36Investigating strength properties of solid materials by application of mechanical stress by applying repeated or pulsating forces generated by pneumatic or hydraulic means
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/24Classification techniques
    • G06F18/243Classification techniques relating to the number of classes
    • G06F18/24323Tree-organised classifiers

Landscapes

  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Pathology (AREA)
  • Immunology (AREA)
  • General Health & Medical Sciences (AREA)
  • Biochemistry (AREA)
  • Analytical Chemistry (AREA)
  • Health & Medical Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Evolutionary Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)

Abstract

The invention discloses a method and a device for predicting the pressure of a roof plate of an intelligent fully mechanized coal mining face of a coal mine, and belongs to the technical field of equipment of the intelligent fully mechanized coal mining face. The invention comprises the following steps: step S10: selecting multi-dimensional working cycle characteristic parameters of a coal mine fully mechanized mining face support; s20: training the training set according to a decision tree algorithm to generate a decision tree A; s30: the mine pressure appearance characteristic index of the previous working cycle is used as a training set and is combined with a decision tree algorithm to generate a decision tree B; s40: and generating a decision tree C according to the mine pressure appearance degrees of different grades in the current working cycle in the step S30 by combining a decision tree algorithm. The invention selects a plurality of characteristic parameters related to the pressure, grades the mine pressure display degree of a single bracket, grades the end point of the working face pressure prediction model, and the grades are used by the decision tree model.

Description

Method and device for predicting roof pressure of intelligent fully mechanized coal mining face of coal mine
Technical Field
The invention relates to a technology for predicting the pressure of a roof plate of an intelligent fully-mechanized coal mining face of a coal mine, and belongs to the technical field of equipment of the intelligent fully-mechanized coal mining face.
Background
In the prior art, the manufacturing and mining process level of the coal mine fully-mechanized mining equipment is continuously improved, however, casualty accidents caused by the stability problems of a roof and a support of a fully-mechanized mining face still do not happen timely, local roof fall, a pressing frame and the like are still main reasons for the casualty accidents of the fully-mechanized mining face, and the outage rate of the fully-mechanized mining face caused by the accidents is still high, so that a large amount of economic loss is caused. In addition, the automation and intelligence level of the mining equipment is improved, and the gradual realization of intelligent mining is an important trend of the development of the fully-mechanized coal mining technology. The intelligent fully-mechanized coal mining face is characterized in that the fully-mechanized coal mining face adopts complete fully-mechanized coal mining equipment with full and comprehensive sensing, self-learning, decision-making and automatic execution functions. The method is characterized in that support surrounding rock coupling adaptive control, support parameters such as initial support force and the like are adaptively adjusted, roof pressure advance prediction, roof fall/pressure frame accident advance early warning, support group self-organization coordination control and the like of the fully mechanized coal mining face are important problems for restricting the improvement of the intelligent mining level of the fully mechanized coal mining face, and the basis for solving the problems is to realize intelligent perception of the states of supports and a roof.
In addition, in recent years, casualty accidents caused by roof fall problems of fully-mechanized coal mining surfaces still do not happen at all times, outage rates caused by accidents such as local roof fall are still high, more and more fully-mechanized coal mining surfaces are equipped with electro-hydraulic control hydraulic supports at present, and mass monitoring data collected by upright post pressure sensors covering all supports of the fully-mechanized coal mining surfaces provide important opportunities for early warning of roof fall accidents of the fully-mechanized coal mining surfaces.
In order to accurately and reliably perform intelligent sensing on a working surface, stable and effective data acquisition is required as a support. However, the types of data collected by the sensors when the hydraulic support works are limited, and how to effectively analyze the limited data or reasonably combine the data to make the data become characteristic parameters capable of reflecting the relationships among the support, the support and surrounding rocks, the mine pressure law and the like is a significant research.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides the method and the device for predicting the roof pressure of the intelligent fully-mechanized coal mining face, which can accurately realize the accurate prediction of the large-range pressure of the fully-mechanized coal mining face, have high reliability, and can help to analyze the reason of the pressure prediction according to the prediction process of the decision tree model.
The invention is realized by the following technical scheme: in order to achieve the above object, a method for predicting the roof pressure of a fully mechanized coal mining face according to a first aspect of the present invention includes the following steps: s10: selecting multi-dimensional working cycle characteristic parameters of a coal mine fully-mechanized coal mining face support to predict the pressure of the working face, and generating a target training set according to the selected index characteristic parameters, wherein the characteristic parameters are inherent attributes of the fully-mechanized coal mining face and are related to the pressure of the working face; s20: training a training set according to a decision tree algorithm to generate a decision tree A, predicting the mine pressure appearance degree of the previous working cycle by the decision tree A, and obtaining the mine pressure appearance degrees of different levels of the previous working cycle, wherein the training set belongs to a target training set; s30: the mining pressure display characteristic index of the previous working cycle is used as a training set and is combined with a decision tree algorithm to generate a decision tree B, the mining pressure display degree of the current working cycle is predicted by the decision tree B, and mining pressure display degrees of different grades of the current working cycle are obtained, wherein the training set belongs to a target training set; s40: and generating a decision tree C according to the mine pressure display degrees of different levels in the current working cycle in the step S30 by combining a decision tree algorithm, predicting the pressure state of the whole working surface by the decision tree C, and classifying the pressure state of the working surface.
In addition, the method for predicting the coming pressure of the top plate of the fully mechanized mining face of the coal mine can also have the following technical characteristics:
preferably, when the multi-dimensional duty cycle characteristic parameters are selected in step S10, different parameters are selected with different degrees of sensitivity according to different production texture conditions.
Preferably, the mine pressure display degree is graded according to the working resistance change of the hydraulic support.
The invention also aims to provide a device for predicting the coming pressure of the top plate of the fully mechanized coal mining face.
According to a second aspect of the invention, a device for predicting the pressure of a roof panel of a fully mechanized coal mining face comprises: the system comprises a selecting unit, a judging unit and a judging unit, wherein the selecting unit selects multi-factor working cycle characteristic parameters of a coal mine fully-mechanized coal mining face support to predict the coming pressure of a working face and generates a target training set according to the selected index characteristic parameters, and the characteristic parameters are inherent attributes of the fully-mechanized coal mining face; the first training unit is used for training a training set according to a decision tree algorithm to generate a decision tree A, predicting the mine pressure appearance degree of the previous working cycle by the decision tree A and obtaining the mine pressure appearance degrees of different grades of the previous working cycle, wherein the training set belongs to a target training set; the second training unit is used for generating a decision tree B by taking the mine pressure appearance characteristic index of the previous working cycle as a training set and combining a decision tree algorithm, predicting the mine pressure appearance degree of the current working cycle by the decision tree B and obtaining the mine pressure appearance degrees of different grades of the current working cycle, wherein the training set belongs to a target training set; and a third training unit, which generates a decision tree C according to the mining pressure appearance degrees of different levels in the current working cycle in the step S30 and by combining a decision tree algorithm, predicts the pressure coming state of the whole working surface by the decision tree C, and classifies the pressure coming state of the working surface by the decision tree C.
Preferably, when the screening unit selects the characteristic parameters of the multi-dimensional working cycle, the screening unit selects different parameters with different degrees of sensitivity according to different production texture conditions.
Preferably, the mine pressure display degree is graded according to the working resistance change of the hydraulic support.
The invention has the beneficial effects that:
the method and the device for predicting the roof pressure of the intelligent fully mechanized coal mining face can accurately realize the accurate prediction of the large-range pressure of the fully mechanized coal mining face, have high reliability, help to analyze the reason of the pressure prediction according to the prediction process of the decision tree model, and realize the roof fall early warning of the fully mechanized coal mining face by combining with an artificial intelligence algorithm. The definition of the new parameters of the series of multi-dimensional working cycle characteristics provided by the invention overcomes the defect that the data acquisition quantity of the traditional hydraulic support is limited, namely, a few parameters of the traditional initial supporting force, the traditional final resistance, the traditional time weighted resistance and the like are expanded, the interaction relation between the support state and the support and the surrounding rock can be analyzed from more dimensions, and powerful data support is provided for analyzing and solving related problems.
Drawings
FIG. 1 is a flow chart of a method for predicting the pressure of a roof panel of a fully mechanized coal mining face according to the present invention;
FIG. 2 is a schematic diagram illustrating a resistance variation process of a hydraulic support on a top plate of a fully mechanized coal mining face according to the present invention;
FIG. 3 is a schematic diagram of a structure of a prediction decision tree A according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a structure of a prediction decision tree B according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a structure of a coming pressure prediction decision tree C according to an embodiment of the present invention;
FIG. 6 is a graph of stent resistance time series;
FIG. 7 is a graph of the safety valve open duty cycle support resistance timing;
FIG. 8 is a plan view of a working cycle of the degree of pressure development according to an embodiment of the present invention;
FIG. 9 is a diagram illustrating a process of evolution of an incoming pressure state according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the specification, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without any inventive step, are within the scope of the present invention.
Techniques, methods and apparatus known to those skilled in the art may not be discussed in detail but are intended to be part of the specification where appropriate.
According to a first aspect of the present invention, a method for predicting the roof pressure of a fully mechanized coal mining face as shown in fig. 1 includes the following steps:
s10: selecting multi-dimensional working cycle characteristic parameters of a coal mine fully-mechanized coal mining face support to predict the pressure of the working face, and generating a target training set according to the selected index characteristic parameters, wherein the characteristic parameters are inherent attributes of the fully-mechanized coal mining face and are related to the pressure of the working face;
when the multi-factor working cycle characteristic parameters are selected as judgment indexes for the working face pressure prediction, different production geological conditions have different degrees of sensitivity to different parameters, and the parameters selected in the prediction are different, namely when the multi-factor working cycle characteristic parameters are selected, the different degrees of sensitivity to different parameters are selected according to different production texture conditions. The characteristic parameters of the multi-dimensional working cycle of the coal mine fully mechanized coal mining face support in the invention can comprise:
a. resistance increase rate R in initial pressurization stagei
b. Resistance increase rate R in relatively stable bearing stages
c. Drag rate R before moving rackf
d. The ratio of the time-weighted operating resistance to the nominal operating resistance, i.e. RTY=PTWAP/Py
e. Number of yields (number of opening of safety valve) Y
f. Duty cycle time Tc
g. Last cycle mine pressure display degree Q
h. The working cycle predicts the ore pressure display degree P
The method can also comprise parameters such as net end resistance, the accumulated resistance reduction amount of the opened safety valve, the accumulated plunger descending amount of the opened safety valve, initial supporting force, time-weighted working resistance, end resistance, the opening times of the safety valve and the like.
The invention also proposes the following parameters:
aiming at the technical problems of single research parameter, unstable parameter characterization effect and the like in the existing support pressure data analysis, the invention defines the new multi-dimensional work cycle characteristic parameters aiming at the support monitoring data, and the parameters convert the limited data of the support into the new characteristic parameters capable of reflecting the relationship among the support, the support and the surrounding rock, thereby providing reasonable and effective novel parameter support for the follow-up work of intelligent perception of the support and the surrounding rock.
Referring to FIG. 6, a stent resistance time sequence plot, t 1-initial boost time; t 2-relatively stable bearing time; t 3-bearing time before moving the rack; m1-initial support pressure increase amount; m2 — amount of pressure increase in relatively stable load-bearing phase; m3-load stage pressure increase before racking.
Therefore, the following parameters are provided for the support resistance time sequence curve when the safety valve is not opened, and the following parameters are provided for the support resistance time sequence curve when the safety valve is not opened:
[1] initial holding force correction value SLC
Defining as the initial force correction value a phenomenon or parameter having the following characteristics or properties: after the hydraulic support of the fully mechanized mining face reaches the initial supporting force, the pressure can be rapidly reduced in a short time, the reduction range is 0.5-5 MPa, the whole pressure reduction process is 0.5-3 min, then the pressure is slowly increased or kept relatively stable until a new relative balance pressure is reached, and the pressure at the moment is used as the initial supporting force correction value provided for the top plate by the support;
[2] the resistance increasing rate RIIB LR (initial loading period) in the initial pressurizing stage is the resistance increasing rate of the bracket in a rapid resistance increasing stage (which depends on the resistance increasing characteristic and geological conditions of the bracket and is generally 5-30 min) in a short time after initial support, and is an important index for measuring the intensity degree of roof plate movement;
defining a phenomenon or parameter having the following characteristics or properties as the initial boost phase rate of increase:
in the initial pressurization stage, the time spent by the m-number stent is defined as the initial pressurization time t1mTime t1mThe change of the initial support pressure of the inner support is defined as the increase M of the initial support pressure1mAnd then defining the pressure resistance increasing rate of the No. m bracket in the stage as follows:
Figure GDA0003581144160000051
t1the time length depends on the resistance-increasing characteristic of the support and geological conditions, generally ranges from 5min to 30min, and in addition, if the initial supporting force needs to be corrected, the initial supporting force correction value is needed to be carried out at the time tSLCCalculating the resistance increasing rate in the initial pressurization stage;
[3] relative stable bearing stage resistance increasing rate RIRSL
The support bearing stage between the initial pressurizing stage and the pre-moving pressurizing stage is called a relatively stable bearing stage.
Defining a phenomenon or parameter having the following characteristics or properties as a relatively stable bearer phase resistance increase rate:
in the relatively stable bearing stage, the time spent by the bracket I is defined as the relatively stable bearing time t2lTime t2lThe pressure change of the inner support is defined as the pressure increment M in the relatively stable bearing stage2lAnd then defining the pressure resistance increasing rate of the No. l bracket in the stage as follows:
Figure GDA0003581144160000061
[4] resistance increasing rate RIPSMS in boosting stage before moving frame
Under the influence of coal cutting of a coal mining machine or descending and forward movement of an adjacent support, part of top plate load originally acting on a coal wall or the adjacent support is instantaneously transferred to the support, so that the working resistance of the support is rapidly increased in a short time (generally 1-20 min before the support is moved), and the resistance increasing rate of the stage is called as the resistance increasing rate of a pre-pressurizing stage of the support movement;
defining a phenomenon or parameter having the following characteristics or properties as the rate of increase in resistance during the pre-racking boost phase: in the stage of pre-moving pressurization, the time spent by the bracket P is defined as the time t of the pre-moving pressurization stage3pTime t3pThe pressure change of the inner support is defined as the pressure increment M in the bearing stage before moving the support3pDefining the pressure resistance increasing rate of the No. p bracket in the stage as follows:
Figure GDA0003581144160000062
[5] initial pressure boost stage plunger retraction rate RIIBRS
The shrinkage rate of the plunger in the initial pressurization stage is the shrinkage rate of the plunger in a short time (which is generally 5-30 min depending on the resistance increasing characteristic and geological conditions of the support) after the initial support;
defining a phenomenon or parameter having the following characteristics or properties as the rate of plunger recession during the initial pressurization phase: in the initial pressurization stage, the time spent by the m-number stent is defined as the initial pressurization time t1mTime t1mThe lower shrinkage value of the inner plunger is defined as the lower shrinkage L of the plunger in the initial pressurization stage1mDefining the shrinkage rate of the m-number bracket plunger at the stage as follows:
Figure GDA0003581144160000063
[6] rate of collapse RIRSLRS of the plunger during relatively stable loading stage
The support bearing stage between the initial pressurizing stage and the pre-moving pressurizing stage is called a relatively stable bearing stage;
defining a phenomenon or parameter having the following characteristics or properties as the rate of collapse of the plunger during the relatively stable load-bearing phase:
in the relatively stable bearing stage, the time spent by the bracket I is defined as the relatively stable bearing time t2lTime t2lThe lower shrinkage value of the inner plunger is defined as the lower shrinkage L of the plunger in the relatively stable bearing stage2lThen, defining the shrinkage rate of the number l bracket plunger at the stage as follows:
Figure GDA0003581144160000071
[7] RiPSSRS (RiPSSRS) shrinkage rate of movable column in pressurization stage before moving rack
Under the influence of coal cutting of a coal mining machine or descending and forward movement of adjacent supports, part of top plate load originally acting on a coal wall or the adjacent supports is instantaneously transferred to the supports, so that the working resistance of the supports is rapidly increased in a short time (generally 1-20 min before the supports are moved), and the bearing stage is called as a pre-support moving pressurization stage; because the load borne by the adjacent brackets is transferred to the brackets, the pressure of the upright posts of the brackets is increased, and the upright posts of the brackets are contracted due to the elastic compression of the emulsion in the oil cylinders;
defining a phenomenon or parameter having the following characteristics or properties as the plunger retraction rate during the pre-racking loading stage:
in the stage of bearing before moving the rack, the time spent by the P number of racks is defined as the bearing time t before moving the rack3pTime t3pInternal plunger downward shrinkage value is fixedDefining the lower shrinkage L of the plunger in the front bearing stage of moving rack3pThen, defining the shrinkage rate of the number p bracket plunger at the stage as follows:
Figure GDA0003581144160000072
in addition, in the case of long-time production stoppage or strong pressure coming on the working surface, the pressure of the bracket may reach or exceed the rated working resistance, and in order to prevent the bracket upright post from being damaged by the high-pressure emulsion, the safety valve must be opened to overflow the high-pressure emulsion so as to reduce the pressure in the upright post hydraulic cylinder, and then the safety valve is closed, wherein the whole opening and closing process is called a yield cycle (or the safety valve is opened once). In the mine pressure observation of the fully mechanized mining face, the opening times (yield times) of the safety valve are important indexes reflecting the activity intensity of the top plate, and for a hydraulic support with reasonable bearing capacity, the yield times of the top plate in the pressure-incoming period are generally larger than that in the non-pressure-incoming period. At present, the extraction of safety valve features is not comprehensive, and the following parameters are defined to describe specific changes in the safety valve opening process by referring to a safety valve opening work cycle support resistance time sequence chart in fig. 7:
[1]relief valve opening pressure Pvs
In order to prevent the stand column of the bracket from being damaged by the high-pressure emulsion, when the pressure of the stand column of the bracket reaches or exceeds the set rated working resistance, the safety valve must be opened to enable the high-pressure emulsion to overflow so as to reduce the pressure in the hydraulic cylinder of the stand column, and then the safety valve is closed;
the support resistance when the safety valve is opened is defined as the opening pressure of the safety valve;
this value has the following characteristics: after the working resistance of the hydraulic support reaches maximum values A1, A2 and A3.. the working resistance of the hydraulic support begins to fall greatly, the working resistance values of the points A1, A2 and A3.. are the opening pressure of the safety valve;
[2]safety valve closing pressure Pvc
Defining the resistance of the bracket when the safety valve is closed again after being opened as the closing pressure of the safety valve;
this value has the following characteristics: when the working resistance of the hydraulic support reaches maximum values A1, A2 and A3.. An, if the working resistance of the hydraulic support starts to be reduced and is reduced to resistance minimum values B1, B2 and B3.. Bn and stops, the working resistance value of the minimum value is the opening pressure of the safety valve;
[3]opening duration T of the safety valveTd
Defining the duration t of the safety valve from opening to closingSThe duration of the opening of the safety valve.
This value has the following characteristics: when the working resistance of the hydraulic support reaches maximum points A1, A2 and A3.. An, if the working resistance of the hydraulic support starts to reduce and stops at minimum resistance points B1, B2 and B3.. Bn, the period t is in the processSThe value of the opening duration of the safety valve is obtained;
[4]restart interval time t of safety valvers
Defining the time t which is elapsed in the process of the safety valve adjacent opening and the support resistance rising from the closing to the opening of the safety valversA restart interval also called a safety valve;
this value has the following characteristics: after the working resistance of the hydraulic support reaches the maximum value point An, the time value t of the working resistance of the support during rising is within the distance of the maximum value point An +1 of the resistance of the support next timers
[5] Resistance reduction amount during opening of safety valve
The pressure reduction quantity of the hydraulic support during the period from the opening to the closing of the safety valve is defined as the resistance reduction quantity during the opening period of the safety valve, and the numerical value has the following characteristics: when the working resistance of the hydraulic support reaches the maximum value An, the minimum value delta p of the support resistance is kept in a descending state during the period of the maximum value An +1 of the support resistance next timea
[6] Resistance increase of safety valve in closed state
Defining the increasing resistance quantity of the working resistance of the hydraulic support from the n-th time of closing the safety valve to the n +1 times of opening the safety valve in the continuous opening process of the safety valve, wherein the value has the following characteristics: when the working resistance of the hydraulic support is from a minimum value point Bn to An maximum value point An +1, supportingCradle resistance increase Δ p during a rise state of rack pressureb
[7] Downward shrinkage of plunger during opening of safety valve
The plunger descending quantity of the upright post during the resistance reducing period of the hydraulic support during the period from the nth opening to the nth closing of the safety valve is defined as the plunger descending quantity during the opening period of the safety valve, and the numerical value has the following characteristics:
the support is from the hydraulic support working resistance maximum value An to the vertical column shrinkage value delta a between minimum value Bn;
[8] plunger downward shrinkage under closed state of safety valve
Defining the plunger descending quantity of the safety valve from the pressure minimum value Bn closed at the nth time to the pressure maximum value An +1 opened at the (n + 1) th time as the plunger descending quantity in the closed state of the safety valve in the continuous opening process of the safety valve, wherein the numerical value has the following characteristics:
and when the working resistance of the hydraulic support reaches a minimum value point Bn and is away from a next maximum value point An +1 of the support resistance, the support pressure is at a support column retraction value delta b in a rising state.
S20: training a training set according to a decision tree algorithm to generate a decision tree A, predicting the mine pressure appearance degree of the previous working cycle by the decision tree A, and obtaining the mine pressure appearance degrees of different levels of the previous working cycle, wherein the training set belongs to a target training set;
the mining pressure display degree can be divided into different grades, in the specific embodiment of the invention, the classification into A-E grades is taken as an example, which means that the working resistance of the bracket is gradually increased from too low to very high, namely, the mining pressure display degree is graded according to the change of the working resistance of the hydraulic bracket. In addition, a class F rack pressure exposure level is established for the currently ongoing work cycle, which is substantially similar to E.
S30: taking the mine pressure manifestation characteristic index of the previous working cycle as a training set and combining a decision tree algorithm to generate a decision tree B, predicting the mine pressure manifestation degree of the current working cycle by the decision tree B, and obtaining the mine pressure manifestation degrees of different grades of the current working cycle, wherein the training set belongs to a target training set;
s40: and generating a decision tree C according to the mine pressure display degrees of different levels in the current working cycle in the step S30 by combining a decision tree algorithm, predicting the pressure state of the whole working surface by the decision tree C, and classifying the pressure state of the working surface.
The working face pressure-bearing predicted end point is a final predicted result, and the result can be divided into different categories, grades or degrees. In the embodiment of the invention, the pressure state of the working surface is divided into the following five types:
class i, the working face is in a future pressure state. The mine pressure of all the brackets on the working face is below grade D, or only a few brackets (not more than a%) reach above grade D.
And II, the possibility of the working face pressing in the future exists, the mine pressure display degree of the working face is that the D-grade bracket reaches a percent of the total number of the brackets, but the D-grade bracket does not exceed b percent.
Class III, a small number of brackets on the working surface are in an incoming pressure state, and the probability that the working surface is in the coming pressure state is increased; the working face mine pressure display degree of the bracket at the grade E or above reaches a percent of the total number of the brackets, but is less than c percent.
And IV, the working surface has more brackets in the pressure state, so that the possibility of coming pressure in a large range exists. The bracket with the working face mine pressure display degree above E grade is more than or equal to c%
And class V, the working face is in a strong pressure state. The mining pressure display degree of the bracket with c% or more of the working surface reaches grade F.
Specifically, according to the data partitioning basis, in order to predict the mine pressure appearance degree of each bracket and the whole working surface, 3 decision tree models are established in the present example, the decision tree a is used for evaluating the mine pressure appearance level of the previous working cycle (working cycle a in fig. 2) of the bracket, the decision tree B is used for predicting the mine pressure appearance level of the current working cycle (working cycle B in fig. 2) of the bracket, and the decision tree C is used for evaluating and predicting the whole mine pressure state of the working surface. The working surface pressure can be accurately predicted through three cycles.
More specifically, decision tree a: judging the mine pressure display degree of the completed circulation; as shown in fig. 3, the decision tree a is used to determine the degree of pressure development of the completed work cycle. The model comprehensively considers the working cycle characteristic parameters such as time-weighted working resistance, the resistance increasing rate in the initial pressurization stage, the yield times, the resistance increasing rate in the relatively stable bearing stage and the like. Decision tree a has 10 leaf nodes in total, resulting in 10 leaf nodes and A, B, C, D, E degrees of five-level mining impression. In the practical application process, the optimal decision tree structure and the value of the characteristic parameter of the decision point can be obtained by analyzing the specific production geological condition and the mine pressure display characteristic adjustment of the fully mechanized coal mining face.
Decision tree B: predicting the mine pressure display degree of the current working cycle; the decision tree a evaluation result is obtained by analyzing the measured value of the previous working cycle (cycle a in fig. 2), which is equivalent to summarizing the previous roof activity characteristic, and if the roof activity characteristic at the next moment is to be predicted, the final mine pressure appearance degree of the currently-performed working cycle must be judged. As shown in fig. 2, if the current time is M, that is, the current working cycle is in the initial pressurization phase or just enters the relatively stable load-bearing phase, the only available characteristic parameter of the working cycle B is the initial supporting force PsAnd the resistance increase rate R in the initial pressurization stagei. The fully mechanized mining work is a gradually developing process under the condition that the roof pressure is generally applied, the current working cycle and the previous working cycle generally keep similar mine pressure display degree, therefore, the mine pressure display characteristic of the previous working cycle is an important index for predicting the mine pressure display level of the current cycle, and the decision tree B comprehensively considers the initial support force P of the current cyclesAnd the resistance increase rate R in the initial pressurization stageiThe mine pressure display level C 'of the previous working cycle, the weighted working resistance TWAD' of the previous cycle time and the resistance increasing rate R before moving the rack of the previous cyclen' as an index of the apparent level of the mine pressure in the cycle. And because the mine pressure display degree of the roof is aggravated by the fact that the advancing speed of the working face is too slow, the mine pressure display degree of the next working cycle is possibly aggravated by the working cycle with too long duration, and the cycle time T of the previous working cycle is shortenedc' is also an important index for predicting the current cyclic variation trend. As shown in FIG. 4, the decision tree B is used to determine the current taskThe degree of pressure development of the circulation.
Decision tree C: predicting the integral pressure coming state of the working face; the pressure of the area with the working face not prone to is often not steps, the area pressed first is pressed 1 or more working cycles earlier than other areas, therefore, the asynchronism of the pressure of the working face can be utilized, and the large-scale pressure of the working face can be accurately forecasted by identifying a small number of supports in potential pressure states or pressure states in advance before the large-scale pressure is pressed. The face pressure prediction decision tree model is shown in fig. 5. According to the classification of the pressure state, when the pressure state of the working face is in the class II or III, the possibility that the pressure of the working face is about to be increased in the subsequent extraction process can be considered.
The method comprises the steps of firstly selecting a plurality of characteristic parameters related to the coming pressure, then grading the mine pressure display degree of a single support, and secondly grading the end point of a working face coming pressure prediction model, wherein the grades are used by a decision tree model.
The invention is verified on a working face, and the fact proves that the prediction model is accurate and effective, and can predict the coming pressure of the fully mechanized mining face.
The invention also aims to provide a device for predicting the coming pressure of the top plate of the fully mechanized coal mining face.
According to a second aspect of the invention, a device for predicting the coming pressure of a roof plate of a fully mechanized coal mining face comprises: the system comprises a selecting unit, a judging unit and a judging unit, wherein the selecting unit selects multi-factor working cycle characteristic parameters of a coal mine fully-mechanized coal mining face support to predict the coming pressure of a working face and generates a target training set according to the selected index characteristic parameters, and the characteristic parameters are inherent attributes of the fully-mechanized coal mining face; the first training unit is used for training a training set according to a decision tree algorithm to generate a decision tree A, predicting the mine pressure appearance degree of the previous working cycle by the decision tree A and obtaining the mine pressure appearance degrees of different grades of the previous working cycle, wherein the training set belongs to a target training set; the second training unit is used for generating a decision tree B by taking the mine pressure appearance characteristic index of the previous working cycle as a training set and combining a decision tree algorithm, predicting the mine pressure appearance degree of the current working cycle by the decision tree B and obtaining the mine pressure appearance degrees of different grades of the current working cycle, wherein the training set belongs to a target training set; and a third training unit, which generates a decision tree C according to the mining pressure appearance degrees of different levels in the current working cycle in the step S30 and by combining a decision tree algorithm, predicts the pressure coming state of the whole working surface by the decision tree C, and classifies the pressure coming state of the working surface by the decision tree C.
Preferably, when the screening unit selects the characteristic parameters of the multi-dimensional working cycle, the screening unit selects different parameters with different degrees of sensitivity according to different production texture conditions.
Preferably, the mine pressure display degree is graded according to the working resistance change of the hydraulic support.
As an embodiment of the present invention, referring to fig. 1-7, and particularly to fig. 8-9, in order to verify the effectiveness of the above top plate pressure prediction model, the pressure data of the support before and after one cycle of pressure applied to a certain working surface is selected for verification. The mining height of the working face is 6.1m, the coal seam roof is sequentially siltstone (average thickness is 4.1m), fine sandstone (average thickness is 6.8m), siltstone (average thickness is 12m), sandy mudstone (average thickness is 3m) and the like from bottom to top, and the periodic pressure of the working face is obvious. A working face is provided with a ZY18000/32/70D type electrohydraulic control hydraulic support 146, and the initial support force and the rated working resistance of the support are respectively 25.2MPa and 45.8 MPa. For the present example, in the coming pressure prediction decision tree model (R), the ratio R of the time-weighted working resistance and the rated working resistance of different decision nodesTYTaking 0.5, 0.7 and 0.8 respectively, and taking different decision nodes to obtain the resistance increasing rate R in the initial pressurization stageiRespectively taking 0.3MPa/min and 0.5MPa/min, and determining the resistance increasing rate R of the node in the relatively stable bearing stagesTaking 0MPa/min, and respectively taking 1 time and 2 times of opening times Y of the safety valve; the selection of relevant parameters in the coming pressure prediction decision tree model II and the model III is consistent with the model introduction example.
FIG. 8 is a plan view of the working cycle of the mining pressure appearance degree grades at different times and different positions calculated according to the model of the coming pressure prediction decision tree (I) and the model (II). In the figure, the mine pressure appearance degree grades A to F are respectively replaced by numbers 1 to 6, and interpolation smoothing processing is carried out on the mine pressure appearance degree grades of the working surfaces at adjacent time and adjacent positions, so that the numerical value is not an integer in some cases. Fig. 8 shows the overall mine pressure development grade of the bracket in a period of time (namely the working cycle time of the working cycle), namely the result of prediction by the coming pressure prediction decision tree model when the working cycle enters a relatively stable bearing stage.
Fig. 9 is a process of evolution of incoming state categories of different time node working planes, in this example, the incoming state evolution can be divided into the following stages:
(1) 4.6h to 13.8h, after the top plate is pressed for the last time, the resistance of the support gradually tends to be relaxed, the resistance increase of the support after the support is initially supported is not obvious, the ore pressure display degree of the working surface only with a small number of end heads of the support reaches grade D, the whole working surface is in a future pressure state (class I), the propelling speed of the working surface is high in the period, and the coal is cut for 9 hours. Fig. 9 (a) shows the pressure distribution and the pressure state of the entire face at the 8 th stage.
(2) 13.8h to 17.8h, with the next working cycle, resistance increase of part of the support in the middle of the working face after resistance pre-support is obvious, the resistance increase rate in the initial pressurization stage reaches more than 0.3MPa/min, the ore pressure display grade of the working face exceeding 5% reaches grade D, but the support does not reach grade E or the grade above, so that the pressure coming state of the working face at the stage is class II, namely the possibility of coming pressure of the working face exists, and pressure forecast can be sent out. Fig. 9 (b) shows the pressure distribution and the pressure state of the entire face at 15h at this stage.
(3) And (6) from 17.8h to 21.5h, the resistance increase of most of the brackets in the middle of the working face after primary support is obvious, the ore pressure display degree of more than 40% of the brackets reaches grade D, only a few brackets reach grade E and above, the incoming pressure probability of the working face is further increased, and the integral incoming pressure state of the working face reaches grade III. Fig. 9 (c) shows the pressure distribution and the pressure state of the entire face at 18.5h at this stage.
(4) After 21.5h, the working face enters the next coal mining cycle, resistance increase is rapid after a large number of supports in the middle of the working face are initially supported, the resistance increase rate reaches over 0.5MPa/min in the initial pressurization stage, the integral pressure state of the working face reaches class V, and 22h, large-range pressure is applied to the middle of the working face, a safety valve is frequently opened, gangue leakage between racks and coal wall burst are serious, and severe noise is accompanied with goaf.
According to the analysis of the above example, when the pressure of the whole working surface reaches II class and III class respectively about 4.2h and 8.2h before the large-range pressure of the middle part of the working surface is reached, the impending pressure of the top plate of the working surface is accurately predicted. Therefore, example verification shows that the roof pressure prediction model based on the multi-dimensional working cycle characteristic parameters and the decision tree theory, which is proposed above, is effective and accurate and can be used for predicting and forecasting the roof pressure of the fully mechanized mining face.
According to the coal mine fully-mechanized coal mining face roof pressure prediction device provided by the embodiment of the invention, the accurate prediction of the large-range pressure of the coal mine fully-mechanized coal mining face can be accurately realized, the reliability is high, and the reason of the pressure prediction can be helped to be analyzed according to the decision tree model prediction process.
The above description is only exemplary of the present invention and should not be taken as limiting the invention, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (7)

1. A method for predicting the pressure of a roof plate of an intelligent fully mechanized coal mining face of a coal mine is characterized by comprising the following steps: the method comprises the following steps:
s10: selecting multi-dimensional working cycle characteristic parameters of a coal mine fully-mechanized coal mining face support to predict the coming pressure of a working face, and generating a target training set according to the multi-dimensional working cycle characteristic parameters;
s20: training a training set according to a decision tree algorithm to generate a decision tree A, predicting the mine pressure appearance degree of the previous working cycle by the decision tree A, and obtaining the mine pressure appearance degrees of different levels of the previous working cycle, wherein the training set belongs to a target training set;
s30: taking the multi-dimensional working cycle characteristic parameters of the previous working cycle as a training set and combining a decision tree algorithm to generate a decision tree B, predicting the mine pressure manifestation degree of the current working cycle by the decision tree B, and obtaining the mine pressure manifestation degrees of different grades of the current working cycle, wherein the training set belongs to a target training set;
s40: generating a decision tree C according to the mine pressure display degrees of different levels of the current working cycle in the step S30 and by combining a decision tree algorithm, predicting the pressure state of the whole working surface by the decision tree C, and classifying the pressure state of the working surface;
in step S10, the multi-dimensional duty cycle characteristic parameters include the following parameters:
the following parameters are provided according to a support resistance time sequence curve when the safety valve is not opened:
[1] initial holding force correction value SLC
After the hydraulic support of the fully mechanized mining face reaches the initial support force, the pressure can rapidly drop within 0.5-3 min, the drop amplitude is 0.5-5 MPa, then the pressure slowly increases or keeps relatively stable until reaching a new relative balance pressure, and the pressure at the moment is used as the initial support force correction value SLC provided for the top plate by the support;
[2] initial boost phase resistance increase rate RIIBLR
The resistance increasing rate in the initial pressurizing stage is the resistance increasing rate in the rapid resistance increasing stage within 5-30 min after the initial support of the support, and is an important index for measuring the intensity of the top plate movement; in the initial pressurization stage, the time spent by the m-number stent is defined as the initial pressurization time t1mTime t1mThe change of the initial support pressure of the inner support is defined as the increase M of the initial support pressure1mAnd then defining the pressure resistance increasing rate of the No. m bracket in the stage as follows:
Figure 736889DEST_PATH_IMAGE001
wherein, t1The time length depends on the resistance-increasing characteristics of the support and the geological conditions, and if the initial supporting force needs to be correctedShould be corrected from the initial setting force at time tSLCCalculating the resistance increasing rate RIIBLR in the initial pressurization stage;
[3] resistance increasing rate RIRSL in relatively stable bearing stage
A support bearing stage between the initial pressurizing stage and the pre-moving pressurizing stage is called a relatively stable bearing stage;
in the relatively stable bearing stage, the time spent by the bracket I is defined as the relatively stable bearing time t2lTime t2lThe pressure change of the inner support is defined as the pressure increment M in the relatively stable bearing stage2lAnd then defining the pressure resistance increasing rate of the No. l bracket in the stage as follows:
Figure 919609DEST_PATH_IMAGE002
[4] resistance increasing rate RIPSMS in boosting stage before moving frame
Under the influence of coal cutting of a coal mining machine or descending and forward movement of adjacent supports, part of top plate load originally acting on a coal wall or the adjacent supports is instantaneously transferred to a working face support, so that the working resistance of the working face support is rapidly increased within 1-20 min, and the resistance increasing rate of the stage is called as the resistance increasing rate of a pre-support moving pressurization stage;
in the stage of pre-moving pressurization, the time spent by the bracket P is defined as the time t of the pre-moving pressurization stage3pTime t3pThe pressure change of the inner support is defined as the pressure increment M in the bearing stage before moving the support3pDefining the pressure resistance increasing rate of the No. p bracket in the stage as follows:
Figure 840291DEST_PATH_IMAGE003
[5] initial boost stage plunger pull-down rate RIIBRS
The plunger descending rate in the initial pressurizing stage is the plunger descending rate within 5-30 min after the initial support; in the initial pressurization stage, the time spent by the m-number stent is defined as the initial pressurization time t1mTime t1mInternal movable columnThe downward shrinkage value is defined as the downward shrinkage L of the plunger in the initial pressurizing stage1mDefining the shrinkage rate of the m-number bracket plunger at the stage as follows:
Figure 766659DEST_PATH_IMAGE004
[6] live column shrinkage rate RIRSLRS in relatively stable load-bearing stage
In the relatively stable bearing stage, the time spent by the bracket I is defined as the relatively stable bearing time t2lTime t2lThe lower shrinkage value of the inner plunger is defined as the lower shrinkage L of the plunger in the relatively stable bearing stage2lThen, defining the shrinkage rate of the number l bracket plunger at the stage as follows:
Figure 514121DEST_PATH_IMAGE005
[7] RiPSSRS (RiPSSRS) shrinkage rate of movable column in pressurization stage before moving rack
Under the influence of coal cutting of a coal mining machine or descending and forward movement of adjacent supports, part of top plate load originally acting on a coal wall or the adjacent supports is instantaneously transferred to a working face support, so that the working resistance of the working face support is rapidly increased within 1-20 min, and the stage is called as a pre-support-moving pressurization stage; because the load borne by the adjacent supports is transferred to the working face support, the pressure of the upright post of the working face support is increased, and the upright post of the working face support is contracted due to the elastic compression of the emulsion in the oil cylinder;
in the stage of bearing before moving the rack, the time spent by the P number of racks is defined as the bearing time t before moving the rack3pTime t3pThe lower shrinkage value of the inner plunger is defined as the lower shrinkage L of the plunger in the bearing stage before moving the rack3pThen, defining the shrinkage rate of the number p bracket plunger at the stage as follows:
Figure 867742DEST_PATH_IMAGE006
when the safety valve of the bracket is opened, the following parameters are extracted from the monitoring data of the resistance and the displacement of the bracket:
[1]relief valve opening pressure Pvs
In order to prevent the stand column of the bracket from being damaged by the high-pressure emulsion, when the pressure of the stand column of the bracket reaches or exceeds the set rated working resistance, the safety valve must be opened to enable the high-pressure emulsion to overflow so as to reduce the pressure in the hydraulic cylinder of the stand column, and then the safety valve is closed; defining the resistance of the bracket when the safety valve is opened as the opening pressure P of the safety valvevs
When the working resistance of the hydraulic support reaches maximum values A1, A2 and A3.. An, if the working resistance of the hydraulic support begins to drop greatly, the working resistance values of the points A1, A2 and A3.. An are the opening pressure of the safety valve;
[2]safety valve closing pressure Pvc
Defining the resistance of the bracket when the safety valve is closed again after being opened as the closing pressure of the safety valve;
when the working resistance of the hydraulic support reaches maximum values A1, A2 and A3.. An, if the working resistance of the hydraulic support starts to be reduced and is reduced to minimum resistance values B1, B2 and B3.. Bn to stop, the working resistance value of the minimum value is the closing pressure P of the safety valvevc
[3]Opening duration T of the safety valveTd
Defining the duration T of the safety valve from opening to closingTdThe safety valve opening duration;
when the working resistance of the hydraulic support reaches maximum points A1, A2 and A3.. An, if the working resistance of the hydraulic support starts to reduce and is reduced to minimum resistance points B1, B2 and B3.. Bn, the hydraulic support stops, during the period TTdThe value of the opening duration of the safety valve is obtained;
[4]restart interval time t of safety valvers
Defining the time t which is elapsed in the process of the safety valve adjacent opening and the support resistance rising from the closing to the opening of the safety valversAlso called the safety valve restart interval time;
when the working resistance of the hydraulic support reaches the maximum value point An, the time value of the working resistance of the support during rising is within the distance of the maximum value point An +1 of the resistance of the support next timeIs trs
[5] Resistance reduction amount during opening of safety valve
Defining the pressure reduction quantity of the hydraulic support during the period from opening to closing of the safety valve as the resistance reduction quantity during the opening period of the safety valve, and after the working resistance of the hydraulic support reaches a maximum value point An and is away from the next maximum value point An +1 period of the support resistance, reducing the support resistance by a reduction value delta p during the period that the support working resistance is in a reduction statea
[6] Resistance increase of safety valve in closed state
Defining the resistance increasing amount of the hydraulic support working resistance from the n-th time of the safety valve closing to the n +1 times of the safety valve opening in the continuous opening process of the safety valve, and when the hydraulic support working resistance is from a minimum value point Bn to An maximum value point An +1, the support resistance increasing value delta p of the support pressure in the rising stateb
[7] Downward shrinkage of plunger during opening of safety valve
Defining the plunger descending amount of the upright post during the resistance reducing period of the hydraulic support from the nth opening period to the nth closing period of the safety valve as the plunger descending amount during the opening period of the safety valve, and defining the upright post descending value delta a between the maximum value An and the minimum value Bn of the working resistance of the hydraulic support by the support;
[8] plunger downward shrinkage under closed state of safety valve
In the continuous opening process of the safety valve, the plunger descending amount of the safety valve from the nth closed pressure minimum value Bn to the nth +1 opened pressure maximum value An +1 is defined as the plunger descending amount in the closed state of the safety valve, and after the working resistance of the hydraulic support reaches the minimum value point Bn, the support upright descending amount delta b of the support pressure in the ascending state is defined in the period of being away from the next support resistance maximum value point An + 1.
2. The method for predicting the coming pressure of the roof plate of the intelligent fully mechanized coal mining face of the coal mine according to claim 1, wherein: when the multi-dimensional duty cycle characteristic parameters are selected in step S10, the parameters are selected to have different degrees of sensitivity according to different production geological conditions.
3. The method for predicting the coming pressure of the roof plate of the intelligent fully mechanized coal mining face of the coal mine according to claim 1, wherein: and grading the mine pressure display degree according to the working resistance change of the hydraulic support.
4. The method for predicting the coming pressure of the roof plate of the intelligent fully mechanized coal mining face of the coal mine according to claim 1, wherein: the mine pressure display degree is graded according to the working resistance change of the hydraulic support, and is divided into A-F grades, which shows that the working resistance of the support is gradually increased from too low to very high;
the pressure-coming state of the working face is divided into the following five types:
class I, the working face is in a future pressure state; the mine pressure display degree of all the brackets on the working surface is below grade D, or only no more than a% of the brackets reach above grade D;
II, the possibility of the working face to be pressed in the future exists; the number of the brackets with the working face mine pressure display degree in the D level reaches a% of the total number of the brackets, but the number of the brackets with the working face mine pressure display degree in the D level is not more than b% of the total number of the brackets;
class III, a small number of brackets on the working surface are in an incoming pressure state; the probability of the working face coming to press is increased; the bracket with the working face mine pressure display degree of grade E or above reaches a percent of the total number of the brackets, but is less than c percent;
IV, more brackets on the working surface are in an incoming pressure state; there is a potential for a wide range of incoming pressures; the bracket with the working face mine pressure display degree above grade E is more than or equal to c%;
v, the working surface is in a strong pressure state; the mining pressure display degree of the bracket with c% or more of the working surface reaches grade F.
5. A coal mine fully mechanized coal mining face roof pressure prediction device adopting the coal mine intelligent fully mechanized coal mining face roof pressure prediction method of claim 1, comprising:
the system comprises a selecting unit, a predicting unit and a training unit, wherein the selecting unit selects multi-dimensional working cycle characteristic parameters of a coal mine fully-mechanized coal mining face support to predict the coming pressure of a working face and generates a target training set according to the multi-dimensional working cycle characteristic parameters, and the multi-dimensional working cycle characteristic parameters are inherent attributes of the fully-mechanized coal mining face;
the first training unit is used for training a training set according to a decision tree algorithm to generate a decision tree A, predicting the mine pressure appearance degree of the previous working cycle by the decision tree A and obtaining the mine pressure appearance degrees of different grades of the previous working cycle, wherein the training set belongs to a target training set;
the second training unit is used for generating a decision tree B by taking the multi-dimensional working cycle characteristic parameters of the previous working cycle as a training set and combining a decision tree algorithm, predicting the mine pressure appearance degree of the current working cycle by the decision tree B and obtaining the mine pressure appearance degrees of different grades of the current working cycle, wherein the training set belongs to a target training set;
and the third training unit is used for generating a decision tree C according to the mine pressure display degrees of different grades in the current working cycle and by combining a decision tree algorithm, predicting the pressure state of the whole working surface by the decision tree C and classifying the pressure state of the working surface by the decision tree C.
6. The coal mine fully mechanized mining face roof pressure prediction device of claim 5, wherein: the system also comprises a screening unit, wherein when the screening unit selects the characteristic parameters of the multi-dimensional working cycle, the screening unit selects different degrees of sensitivity of different parameters according to different production geological conditions.
7. The coal mine fully mechanized mining face roof pressure prediction device of claim 5, wherein: and grading the mine pressure display degree according to the working resistance change of the hydraulic support.
CN202010510637.0A 2020-06-08 2020-06-08 Method and device for predicting roof pressure of intelligent fully mechanized coal mining face of coal mine Active CN111678683B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010510637.0A CN111678683B (en) 2020-06-08 2020-06-08 Method and device for predicting roof pressure of intelligent fully mechanized coal mining face of coal mine

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010510637.0A CN111678683B (en) 2020-06-08 2020-06-08 Method and device for predicting roof pressure of intelligent fully mechanized coal mining face of coal mine

Publications (2)

Publication Number Publication Date
CN111678683A CN111678683A (en) 2020-09-18
CN111678683B true CN111678683B (en) 2022-05-13

Family

ID=72435506

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010510637.0A Active CN111678683B (en) 2020-06-08 2020-06-08 Method and device for predicting roof pressure of intelligent fully mechanized coal mining face of coal mine

Country Status (1)

Country Link
CN (1) CN111678683B (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112149255A (en) * 2020-09-30 2020-12-29 中煤能源研究院有限责任公司 Coal face hydraulic support working resistance prediction method based on LSTM neural network
CN112380249B (en) * 2020-11-24 2023-05-30 中国矿业大学(北京) Coal face period pressure analysis method and device and electronic equipment
CN112798042A (en) * 2020-12-30 2021-05-14 中国矿业大学 Intelligent diagnosis method for working state and supporting quality of hydraulic support
CN113779789B (en) * 2021-09-06 2022-06-21 重庆大学 Intelligent working face overlying rock stress sensing method based on support dynamic pressure difference index
CN114991844B (en) * 2022-08-03 2022-10-25 太原向明智控科技有限公司 Intelligent mining pressure data acquisition and statistics method for fully mechanized coal mining face of coal mine
CN117743988B (en) * 2024-02-20 2024-04-19 太原理工大学 Instant prediction method for pressure-bearing state of hydraulic support after initial support

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105092147A (en) * 2015-07-27 2015-11-25 中国矿业大学 Experiment teaching system and method for rock strata movement simulation and top board pressure monitoring
CN110889087A (en) * 2019-10-29 2020-03-17 青岛本末岩控技术有限公司 Mine pressure big data real-time prediction system and method based on fully mechanized coal mining face
CN111144508A (en) * 2019-12-30 2020-05-12 中国矿业大学(北京) Automatic control system and control method for coal mine auxiliary shaft rail transportation

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP7099163B2 (en) * 2018-08-10 2022-07-12 株式会社リコー Learning device and learning method

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105092147A (en) * 2015-07-27 2015-11-25 中国矿业大学 Experiment teaching system and method for rock strata movement simulation and top board pressure monitoring
CN110889087A (en) * 2019-10-29 2020-03-17 青岛本末岩控技术有限公司 Mine pressure big data real-time prediction system and method based on fully mechanized coal mining face
CN111144508A (en) * 2019-12-30 2020-05-12 中国矿业大学(北京) Automatic control system and control method for coal mine auxiliary shaft rail transportation

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
Shield-Roof Interaction in Longwall Panels: Insights from Field Data and Their Application to Ground Control;Jingyi Cheng et al.;《ADVANCES IN CIVIL ENGINEERING》;20181231;第2018卷;第1-18页 *
煤矿采场智能岩层控制原理及方法;李化敏 等;《煤炭学报》;20190131;第44卷(第1期);第127-140页 *

Also Published As

Publication number Publication date
CN111678683A (en) 2020-09-18

Similar Documents

Publication Publication Date Title
CN111678683B (en) Method and device for predicting roof pressure of intelligent fully mechanized coal mining face of coal mine
CN107977749B (en) Prediction analysis method for oil field development index
CN107559045B (en) Underground coal mine Face Pressure Early-Warning System and method
CN112798042A (en) Intelligent diagnosis method for working state and supporting quality of hydraulic support
CN112529275A (en) Wind power climbing event prediction method based on feature extraction and deep learning
CN112302720B (en) Method and system for judging initial supporting force and cycle end resistance of working surface hydraulic support
CN110929384A (en) Mine pressure big data real-time analysis system and method based on fully mechanized coal mining face
CN104794550A (en) WT-KPCA-SVR coupling model based gas emission quantity prediction method
CN110847968A (en) Periodic pressure-bearing dividing method based on maximum working resistance curve of support cycle
CN107122860A (en) Bump danger classes Forecasting Methodology based on grid search and extreme learning machine
CN109902111A (en) A kind of earth and rockfill dam security information method for digging
Cheng et al. Shield‐Roof Interaction in Longwall Panels: Insights from Field Data and Their Application to Ground Control
CN111911214A (en) Method for monitoring working state of safety valve of hydraulic support
CN116108339A (en) Actual measurement data outlier detection and correction method for tunnel boring machine
CN110552685A (en) Method for calculating working fluid level of oil well by utilizing ground indicator diagram in wax precipitation well
CN116241306A (en) Bracket common various faults judging method based on electro-hydraulic control system
CN116861237A (en) Identification method and device for excavator working state, excavator and processor
CN114429253A (en) Method and system for intelligently prompting adjacent well risk in drilling process
CN116777085A (en) Coal mine water damage prediction system based on data analysis and machine learning technology
CN110991363A (en) Method for extracting CO emission characteristics of coal mine safety monitoring system under different coal mining processes
CN113027532B (en) Method for mining pressure information based on fully-mechanized mining support working resistance data
CN112610263B (en) Fully mechanized coal mining face-based follow machine control method and device and electronic equipment
CN114151133A (en) Grading early warning method for incoming pressure of top plate of fully mechanized mining face
Chen et al. A Deep Regression Method for Gas Well Liquid Loading Prediction
CN115729910A (en) Implementation method and system for analyzing drilling speed of drilling machine in real time

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