CN116680545B - Coal mine well exit personnel prediction method based on Markov random field - Google Patents
Coal mine well exit personnel prediction method based on Markov random field Download PDFInfo
- Publication number
- CN116680545B CN116680545B CN202310961610.7A CN202310961610A CN116680545B CN 116680545 B CN116680545 B CN 116680545B CN 202310961610 A CN202310961610 A CN 202310961610A CN 116680545 B CN116680545 B CN 116680545B
- Authority
- CN
- China
- Prior art keywords
- data acquisition
- well
- workers
- time
- probability
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000003245 coal Substances 0.000 title claims abstract description 47
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000004080 punching Methods 0.000 claims abstract description 28
- 238000009826 distribution Methods 0.000 claims description 34
- 238000005520 cutting process Methods 0.000 claims description 8
- 238000012216 screening Methods 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 description 7
- 230000008901 benefit Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000003190 augmentative effect Effects 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000007599 discharging Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/29—Graphical models, e.g. Bayesian networks
- G06F18/295—Markov models or related models, e.g. semi-Markov models; Markov random fields; Networks embedding Markov models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
- G06F18/15—Statistical pre-processing, e.g. techniques for normalisation or restoring missing data
-
- 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
- G06Q50/02—Agriculture; Fishing; Mining
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C1/00—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people
- G07C1/10—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people together with the recording, indicating or registering of other data, e.g. of signs of identity
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
The application discloses a coal mine well-exiting personnel prediction method based on a Markov random field, which relates to the technical field of coal mine management and can improve the number of predicted well-exiting personnel and the prediction accuracy of well-exiting time to a certain extent. The method comprises the following steps: establishing a plurality of data acquisition devices, and acquiring card punching data of workers; the punching data comprise punching time, worker position information and numbers of corresponding data acquisition devices; establishing an undirected graph by taking each data acquisition device as a node, and obtaining a well peak period by estimating probability density of the undirected graph; establishing a prediction model based on a Markov random field, wherein the prediction model is configured to be used for representing the movement rule of workers among different data acquisition devices and correcting the prediction model; based on the collected punching card data of the workers, outputting the number of people about to go out of the well at the same moment and the probability of going out of the well at a certain moment of each worker through the corrected prediction model.
Description
Technical Field
The application relates to the technical field of coal mine management, in particular to a coal mine well outlet personnel prediction method based on a Markov random field.
Background
Predicting coal mine personnel is helpful for improving the production efficiency and economic benefit of the coal mine. For mine enterprises, knowing the number and time of the personnel entering and exiting the mine is very important, and can help mine enterprise management personnel to better know and master the conditions of personnel in the mine, so that more scientific and reasonable safety management strategies and emergency plans are formulated, and the personal safety of miners is ensured, so that the efficiency and benefit of mine production can be improved.
The method for predicting the number of the underground personnel in the coal mine disclosed in the prior art is to build a position prediction model based on a Kalman filtering principle, train the position prediction model by utilizing collected historical data of punching cards of underground personnel, update the state prediction of the personnel position by the position prediction model, and predict the future position and corresponding time according to the current personnel position.
The Kalman filtering principle-based position prediction model disclosed in the prior art needs to be established by assuming that data to be processed obeys linear Gaussian distribution, and in the actual production working process of a coal mine, the situation that acquired data do not obey the linear Gaussian distribution exists, so that larger errors exist in the number of logging personnel and the logging time of the acquired logging personnel.
Disclosure of Invention
The embodiment of the application provides the coal mine well logging personnel prediction method based on the Markov random field, which can improve the number of predicted well logging personnel and the prediction accuracy of well logging time to a certain extent.
The embodiment of the application provides a coal mine well logging personnel prediction method based on a Markov random field, which comprises the following steps:
establishing a plurality of data acquisition devices, and acquiring card punching data of workers; the punching data comprise punching time, worker position information and numbers of corresponding data acquisition devices;
establishing an undirected graph by taking each data acquisition device as a node, and obtaining a well peak period by estimating probability density of the undirected graph;
establishing a prediction model based on a Markov random field, wherein the prediction model is configured to be used for representing the movement rule of workers among different data acquisition devices and correcting the prediction model;
based on the collected punching card data of the workers, outputting the number of people about to go out of the well at the same moment and the probability of going out of the well at a certain moment of each worker through the corrected prediction model.
In some embodiments, the establishing an undirected graph with each data acquisition device as a node includes:
acquiring the distance between two data acquisition devices and the average time of all workers moving between the two data acquisition devices at the same moment;
outputting the average speed of all workers moving between the two data acquisition devices at the moment;
and taking the average speed of all workers moving at the moment as the weight of the edge, and taking the data acquisition device as a node to establish an undirected graph for the card punching data at each moment.
In some embodiments, the obtaining the well rush hour by estimating the probability density of the undirected graph includes:
defining a high peak value of the data acquisition device, wherein the high peak value of the data acquisition device is the inverse of the arithmetic square root of the average value of the distances between any two data acquisition devices;
calculating a high peak value at each moment, and drawing a probability density estimation graph;
the moment when the output probability density is the largest is taken as the peak period of the well.
In some embodiments, the time at which the output probability density is greatest is included as a peak out time:
the method comprises the steps of obtaining the number of shifts of workers and the number of well-out peak periods each day, wherein the number of well-out peak periods is the same as the number of shifts of workers each day;
and screening the number corresponding to the moment with the maximum probability density according to the number of the well-out peak periods.
In some embodiments, the building a predictive model based on a markov random field comprises:
establishing an undirected graph model for representing the movement rule of coal miners among different data acquisition devices;
each node in the undirected graph model is represented as a data acquisition device at a corresponding position, the two-dimensional array is used for representing the positions of all workers at corresponding data acquisition devices at different times, and the prior information is the distance between different data acquisition devices and the distance from each data acquisition device to a mine outlet.
In some embodiments, the modifying the predictive model includes:
constructing a potential function for representing the movement probability of each worker between different data acquisition devices;
acquiring a first probability distribution representing the probability that all workers are located at each data acquisition device at each moment in time;
giving the current position of a data acquisition device of a worker, and acquiring a second probability distribution for predicting the ore-drawing time of the corresponding worker;
and carrying out maximum posterior probability estimation on the established prediction model based on the Markov random field, and outputting a minimum cutting set for judging the ore-drawing time of the worker.
In some embodiments, the constructing a potential function for representing a probability of each worker moving between different data acquisition devices comprises:
with natural constantConstructing potential function for the underlying exponential function, expressed as follows
wherein ,indicate->Person and->A distance between the data acquisition devices; /> and />Indicating that the worker is +.>Numbering the data acquisition device at the moment; />Indicate->The distance from the data acquisition device to the data acquisition device closest to the mine outlet; />、/>Are all positive constants; />Representing a set of peak times for workers to go out of the well.
In some embodiments, the acquiring a first probability distribution representing a probability that all workers are located at each data acquisition device at each time instant comprises:
the first probability distribution is expressed as a product of a potential function expressed as:
wherein ,expressed as normalized constant, ++> and />Representing different workers, < > about>Represented as different time periods.
In some embodiments, the obtaining a second probability distribution for predicting the ore-drawing time of the corresponding worker includes:
obtaining a second probability distribution through a Bayesian formula, wherein the second probability distribution comprises:
wherein ,indicating the ore removal time of a given worker; />Indicate->The personal worker is at->The position of the data acquisition device where the time periods are located; />Expressed as +.>The personal worker is at->Data acquisition device for each time period>Conditional probability of (2); />Indicating the ore presentation time +.>Is a priori probability of (c).
In some embodiments, the maximum a posteriori probability estimation of the established Markov random field-based predictive model comprises:
adding a source point and a sink point to the undirected graph model, connecting the source point with nodes corresponding to all the data acquisition devices associated with the well outlet time, and connecting the sink point with nodes corresponding to all the data acquisition devices associated with the ore outlet time; the source point is expressed as a well entering moment; the sink is expressed as the ore drawing time;
converting the undirected graph model added with the source points and the sink points into a directed graph;
solving the maximum flow minimum cut of the obtained directed graph;
acquiring a minimum cut set for representing nodes which contain all nodes associated with well-out time, wherein the well-out time is represented as a time period corresponding to a node closest to a sink in the minimum cut set;
when a worker punches a card at the data acquisition device corresponding to the current node and belongs to the minimum cutting set, outputting the situation that the worker is about to go out of the well, and outputting the corresponding predicted well-out moment.
One or more technical solutions provided in the embodiments of the present application at least have the following technical effects or advantages:
according to the embodiment of the application, the data with nonlinear and non-Gaussian distribution is established based on the Markov random field, so that the rule that the data acquired by actual acquisition does not accord with the linear distribution is more met, and the probability that the difference between the prediction result of the original data output acquired by Kalman processing and the actual situation is larger can be reduced; meanwhile, the prediction model established based on the Markov random field principle effectively aims at the condition of real-time dynamic change of acquired data, so that the purposes of improving the number of predicted well personnel and the prediction accuracy of well production time are achieved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments of the present application will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a coal mine well logging personnel prediction method based on a Markov random field provided by an embodiment of the application;
FIG. 2 is a schematic diagram of a coal mine well personnel prediction method based on a Markov random field for representing movement of workers between different data acquisition devices according to an embodiment of the present application;
FIG. 3 is a detailed flow chart of a coal mine well logging personnel prediction method based on a Markov random field, which is provided by an embodiment of the application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the application. 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.
In the description of the embodiments of the present application, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," "outer," and the like indicate or are based on the orientation or positional relationship shown in the drawings, merely to facilitate description of the embodiments of the present application and simplify description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the present application. The terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. Furthermore, the terms "mounted," "connected," "coupled," and "connected" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the embodiments of the present application will be understood by those of ordinary skill in the art according to specific circumstances.
Referring to fig. 1, fig. 2 and fig. 3, fig. 1 is a flowchart of a coal mine well personnel prediction method based on a markov random field according to an embodiment of the present application; FIG. 2 is a schematic diagram of a coal mine well personnel prediction method based on a Markov random field for representing movement of workers between different data acquisition devices according to an embodiment of the present application; FIG. 3 is a detailed flow chart of a coal mine well logging personnel prediction method based on a Markov random field, which is provided by an embodiment of the application.
In some embodiments, the application provides a coal mine well logging personnel prediction method based on a Markov random field, which comprises the following steps: step 110, step 120, step 130 and step 140.
Step 110, a plurality of data acquisition devices are established to acquire the card punching data of workers; the punching data comprise punching time, worker position information and numbers of corresponding data acquisition devices.
The data acquisition device is a base station adopting an ultra-low power consumption Bluetooth protocol, the plurality of data acquisition devices are distributed at different positions of the coal mine based on the geographical position and the working surface density interval of the coal mine, and the data acquisition device is used for transmitting Bluetooth signals so that a user terminal carried by a worker can receive the Bluetooth signals and transmit card punching data to the data acquisition device.
And (3) automatically punching cards when workers carry the attendance cards to different areas, and recording the numbers and the punching time of corresponding punching base stations, wherein the acquisition period of punching data is 30 days.
And 120, establishing an undirected graph by taking each data acquisition device as a node, and obtaining the well peak time by estimating the probability density of the undirected graph.
And 130, establishing a prediction model based on the Markov random field, wherein the prediction model is configured to be used for representing the movement rule of workers among different data acquisition devices, and correcting the prediction model.
And 140, outputting the number of people about to go out of the well at the same moment and the probability of going out of the well at a certain moment of each worker through the corrected prediction model based on the collected punching card data of the workers.
By establishing nonlinear and non-Gaussian distribution data based on a Markov random field, the method is more in line with the rule that the data acquired by actual acquisition is not in line with the linear distribution, and the probability that the difference between the prediction result of the original data output acquired by Kalman processing and the actual situation is larger can be reduced; meanwhile, the prediction model established based on the Markov random field principle effectively aims at the condition of real-time dynamic change of acquired data, so that the purposes of improving the number of predicted well personnel and the prediction accuracy of well production time are achieved.
In some embodiments, when executing the establishment of an undirected graph with each data acquisition device as a node, the method for predicting the coal mine personnel provided by the application comprises the following steps: acquiring the distance between two data acquisition devices and the average time of all workers moving between the two data acquisition devices at the same moment; outputting the average speed of all workers moving between the two data acquisition devices at the moment; and taking the average speed of all workers moving at the moment as the weight of the edge, and taking the data acquisition device as a node to establish an undirected graph for the card punching data at each moment.
Because the nodes of the undirected graph have the inherent attribute that the distance from a data acquisition device to an outlet is kept constant, the undirected graph is built on the basis of the card punching data of the same hour of each day within 30 days of an acquisition period; the weight value of the undirected graph is the average moving speed of all workers in the hour in 30 days in the two data acquisition devices, the average moving speed of all workers in the two data acquisition devices is obtained by dividing the distance between the two data acquisition devices by the average moving time of all workers in the hour in the 30 days in the two data acquisition devices, so that visual statistics on the punching card data is realized to a certain extent, and the efficiency of subsequent data analysis can be improved.
In some embodiments, the method for predicting coal mine personnel provided by the application comprises the following steps of: defining a high peak value of the undirected graph corresponding to the data acquisition device, wherein the high peak value of the data acquisition device is the inverse of the arithmetic square root of the average value of the distances between any two data acquisition devices; calculating a high peak value at each moment, and drawing a probability density estimation graph; the moment when the output probability density is the largest is taken as the peak period of the well.
The data analysis is carried out on the undirected graph obtained by the data acquisition device, the probability density estimation graph of the undirected graph established according to the position of each data acquisition device is obtained, and the probability density estimation graph is established, so that the well peak time is conveniently estimated based on the probability density estimation graph, and the convenience of analyzing the well outlet time of subsequent staff is improved.
In some embodiments, when the moment when the output probability density is maximum is executed as a well-out peak period, the method for predicting the coal mine well-out personnel provided by the application comprises the following steps: the method comprises the steps of obtaining the number of shifts of workers and the number of well-out peak periods each day, wherein the number of well-out peak periods is the same as the number of shifts of workers each day; and screening the number corresponding to the moment with the maximum probability density according to the number of the well-out peak periods. When the shift number of the daily workers is N which is larger than 1, taking N moments with the highest probability as the daily workers to log outPeak period collection of (2)。
Because the shift number of the workers in the coal mine is dynamically changed in different periods, in order to conveniently predict the well outlet time of the workers in the coal mine, a plurality of times with the largest well outlet probability of the workers in the coal mine are obtained through statistics to be used as peak time sets of the workers in the coal mine, when the well outlet time corresponding to each data acquisition device is located in the obtained peak time sets, the probability of the workers in the coal mine is larger, and meanwhile, when the workers are located in the data acquisition devices close to the outlet position, the probability of the corresponding workers leaving the coal mine from the outlet is also larger than that of the workers in the coal mine located in the data acquisition devices at other positions at the same time.
In some embodiments, in performing a markov random field-based prediction model, the method for predicting coal mine personnel comprises the following steps: establishing an undirected graph model for representing the movement rule of coal miners among different data acquisition devices; each node in the undirected graph model is represented as a data acquisition device at a corresponding position, and the two-dimensional array is used for representing the positions of all workers at corresponding data acquisition devices at different times, and the prior information is the distance between different data acquisition devices and the distance from each data acquisition device to a mine outlet.
The undirected graph model is built based on the movement rule of the coal mine workers among different data acquisition devices, and the distance between the different data acquisition devices and the distance between each data acquisition device and a mine outlet are represented by the undirected graph model, so that the movement rule of the different workers among the different data acquisition devices is conveniently analyzed, and the well outlet time of the coal mine workers is obtained accurately.
In some embodiments, when the prediction model is corrected, the method for predicting the coal mine well personnel provided by the application comprises the following steps: constructing a potential function for representing the movement probability of each worker between different data acquisition devices; acquiring a first probability distribution representing the probability that all workers are located at each data acquisition device at each moment in time; giving the current position of a data acquisition device of a worker, and acquiring a second probability distribution for predicting the ore-drawing time of the corresponding worker; and carrying out maximum posterior probability estimation on the established prediction model based on the Markov random field, and outputting a minimum cutting set for judging the ore-drawing time of the worker.
Illustratively, in constructing a potential function representing the probability of each worker moving between different data acquisition devices, the method for predicting coal mine personnel provided by the application comprises the following steps: the potential function is constructed by constructing a polynomial function, gaussian function, exponential function, or the like.
In some embodiments, constructing a potential function for representing a probability of each worker moving between different data acquisition devices includes:
with natural constantConstructing potential function for the underlying exponential function, expressed as follows
wherein ,indicate->Person and->A distance between the data acquisition devices; /> and />Indicating that the worker is +.>Numbering the data acquisition device at the moment; />Indicate->The distance from the data acquisition device to the data acquisition device closest to the mine outlet; />、/>Are all positive constants; />Representing a set of peak times for workers to go out of the well.
By constructing the potential function for representing the probability of each worker moving between different data acquisition devices, the method for constructing the potential function is based on historical punching card data and priori knowledge, and the purpose of effectively describing the preference and constraint between the worker movement rules can be achieved.
In some embodiments, obtaining a first probability distribution representative of a probability that all workers are located at each data acquisition device at each time instant comprises: the first probability distribution is expressed as a product of potential functions expressed as:
wherein ,expressed as normalized constant, ++> and />Representing differentWorker(s)>Represented as different time periods.
The probability of the data acquisition devices at different positions of all workers in all time periods is represented by setting the first probability distribution as the joint probability distribution, so that the probability of the data acquisition devices at different positions of the workers is conveniently obtained at the well-out moment of the corresponding workers.
In some embodiments, obtaining the second probability distribution for predicting the ore-drawing time of the corresponding worker includes:
the second probability distribution is obtained through a Bayesian formula, and the second probability distribution provided by the application comprises the following steps:
wherein ,indicating the ore removal time of a given worker; />Indicate->The personal worker is at->The position of the data acquisition device where the time periods are located; />Expressed as +.>The personal worker is at->Data acquisition device for each time period>Conditional probability of (2); />Indicating the ore presentation time +.>Is a priori probability of (c).
The second probability distribution is obtained based on a Bayesian formula and used as a conditional probability distribution for giving the position of a data acquisition device where a certain worker is currently located, so that the ore-drawing moment with the maximum probability is predicted, and the ore-drawing moment with the maximum probability is obtained based on the position where the corresponding worker is currently located and is used as the output well-drawing moment.
In some embodiments, the maximum a posteriori probability estimation of the established Markov random field-based predictive model comprises: adding a source point and a sink point to the undirected graph model, connecting the source point with nodes corresponding to all the data acquisition devices associated with the well outlet time, and connecting the sink point with nodes corresponding to all the data acquisition devices associated with the ore outlet time; the source point is expressed as a well entering moment; the sink is expressed as the ore drawing time; converting the undirected graph model added with the source points and the sink points into a directed graph; solving the maximum flow minimum cut of the obtained directed graph; acquiring a minimum cut set for representing nodes which contain all nodes associated with well-out time, wherein the well-out time is represented as a time period corresponding to a node closest to a sink in the minimum cut set; when a worker punches a card at the data acquisition device corresponding to the current node and belongs to the minimum cutting set, outputting the situation that the worker is about to go out of the well, and outputting the corresponding predicted well-out moment.
Illustratively, each undirected edge of the undirected graph is replaced by two opposite directed edges, and the same capacity and potential function values are maintained, so that a directed graph of the active sink is obtained; solving a maximum flow minimum cut of the obtained directed graph of the active sink by using a Ford-Fulkerson algorithm, wherein the algorithm is based on a greedy or dynamic programming idea, continuously searching an augmented path or an augmented tree to increase the flow or reduce a cut set, and the minimum cut set is a set containing all nodes associated with the well-out moment; the ore-out time is the time period corresponding to the node closest to the sink.
When a worker punches a card at the current base station position, if the worker belongs to the minimum cutting set, the worker predicts that the well is about to be discharged, and outputs predicted well time; otherwise, the corresponding worker is predicted to continue the operation.
By connecting the source point with all nodes associated with the moment of well production, and giving a large capacity; the sink is also connected with all nodes associated with the ore-drawing moment, and a large capacity is given; therefore, an undirected graph of the active sink is constructed, and the aim of effectively outputting the optimal ore-discharging time can be achieved by solving the maximum ore-discharging time value in the conditional probability distribution and based on the maximum posterior probability estimation and the maximum flow minimum cutting problem.
According to the embodiment of the application, the data with nonlinear and non-Gaussian distribution is established based on the Markov random field, so that the rule that the data acquired by actual acquisition does not accord with the linear distribution is more met, and the probability that the difference between the prediction result of the original data output acquired by Kalman processing and the actual situation is larger can be reduced; meanwhile, the prediction model established based on the Markov random field principle effectively aims at the condition of real-time dynamic change of acquired data, so that the purposes of improving the number of predicted well personnel and the prediction accuracy of well production time are achieved.
In this specification, each embodiment is described in a progressive manner, and the same or similar parts of each embodiment are referred to each other, and each embodiment is mainly described as a difference from other embodiments.
The above embodiments are only for illustrating the technical solution of the present application, and not for limiting the present application; although the application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced with equivalents; such modifications and substitutions do not depart from the spirit of the application.
Claims (8)
1. A coal mine well logging personnel prediction method based on a markov random field, the method comprising:
establishing a plurality of data acquisition devices, and acquiring card punching data of workers; the punching data comprise punching time, worker position information and numbers of corresponding data acquisition devices;
establishing an undirected graph by taking each data acquisition device as a node, and obtaining a well peak period by estimating probability density of the undirected graph;
establishing a prediction model based on a Markov random field, wherein the prediction model is configured to be used for representing the movement rule of workers among different data acquisition devices and correcting the prediction model;
based on the collected punching card data of the workers, outputting the number of people about to go out of the well of the workers at the same moment and the probability of going out of the well of each worker at a certain moment through the corrected prediction model;
the establishing of the predictive model based on the Markov random field comprises the following steps: establishing an undirected graph model for representing the movement rule of coal miners among different data acquisition devices; each node in the undirected graph model is represented as a data acquisition device at a corresponding position, and the two-dimensional array is used for representing the positions of all workers at corresponding data acquisition devices at different times, and the prior information is the distance between different data acquisition devices and the distance from each data acquisition device to a mine outlet;
the modifying the prediction model includes: constructing a potential function for representing the movement probability of each worker between different data acquisition devices; acquiring a first probability distribution representing the probability that all workers are located at each data acquisition device at each moment in time; giving the current position of a data acquisition device of a worker, and acquiring a second probability distribution for predicting the ore-drawing time of the corresponding worker; and carrying out maximum posterior probability estimation on the established prediction model based on the Markov random field, and outputting a minimum cutting set for judging the ore-drawing time of the worker.
2. The method for predicting coal mine personnel based on markov random fields of claim 1, wherein the establishing an undirected graph with each data acquisition device as a node comprises:
acquiring the distance between two data acquisition devices and the average time of all workers moving between the two data acquisition devices at the same moment;
outputting the average speed of all workers moving between the two data acquisition devices at the moment;
and taking the average speed of all workers moving at the moment as the weight of the edge, and taking the data acquisition device as a node to establish an undirected graph for the card punching data at each moment.
3. The method for predicting coal mine personnel based on markov random fields of claim 2, wherein the obtaining the peak period of the well by estimating the probability density of the undirected graph comprises:
defining a high peak value of the data acquisition device, wherein the high peak value of the data acquisition device is the inverse of the arithmetic square root of the average value of the distances between any two data acquisition devices;
calculating a high peak value at each moment, and drawing a probability density estimation graph;
the moment when the output probability density is the largest is taken as the peak period of the well.
4. A markov random field based coal mine well personnel prediction method according to claim 3 wherein the time at which the output probability density is greatest is comprised of:
the method comprises the steps of obtaining the number of shifts of workers and the number of well-out peak periods each day, wherein the number of well-out peak periods is the same as the number of shifts of workers each day;
and screening the number corresponding to the moment with the maximum probability density according to the number of the well-out peak periods.
5. The method of coal mine personnel prediction based on markov random fields of claim 1, wherein the constructing a potential function for representing the probability of each worker moving between different data acquisition devices comprises:
with natural constantConstructing potential function for the underlying exponential function, expressed as follows
wherein ,indicate->Person and->A distance between the data acquisition devices; /> and />Indicating that the worker is +.>Numbering the data acquisition device at the moment; />Indicate->The distance from the data acquisition device to the data acquisition device closest to the mine outlet; />、/>Are all positive constants; />Representing a set of peak times for workers to go out of the well.
6. The method of coal mine personnel prediction based on markov random fields of claim 5, wherein the obtaining a first probability distribution representative of a probability that all workers are at each data acquisition device at each time instant comprises:
the first probability distribution is expressed as a product of a potential function expressed as:
wherein ,expressed as normalized constant, ++> and />Representing different workers, < > about>Represented as different time periods.
7. The method of coal mine personnel prediction based on markov random fields of claim 6, wherein the obtaining a second probability distribution for predicting the time of the mine of the corresponding worker comprises:
obtaining a second probability distribution through a Bayesian formula, wherein the second probability distribution comprises:
wherein ,indicating the ore removal time of a given worker; />Indicate->The personal worker is at->The position of the data acquisition device where the time periods are located; />Expressed as +.>The personal worker is at->Data acquisition device for each time period>Conditional probability of (2); />Indicating the ore presentation time +.>Is a priori probability of (c).
8. The method for predicting coal mine personnel based on markov random fields as set forth in claim 1, wherein the estimating the maximum posterior probability of the established prediction model based on markov random fields comprises:
adding a source point and a sink point to the undirected graph model, connecting the source point with nodes corresponding to all the data acquisition devices associated with the well outlet time, and connecting the sink point with nodes corresponding to all the data acquisition devices associated with the ore outlet time; the source point is expressed as a well entering moment; the sink is expressed as the ore drawing time;
converting the undirected graph model added with the source points and the sink points into a directed graph;
solving the maximum flow minimum cut of the obtained directed graph;
acquiring a minimum cut set for representing nodes which contain all nodes associated with well-out time, wherein the well-out time is represented as a time period corresponding to a node closest to a sink in the minimum cut set;
when a worker punches a card at the data acquisition device corresponding to the current node and belongs to the minimum cutting set, outputting the situation that the worker is about to go out of the well, and outputting the corresponding predicted well-out moment.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310961610.7A CN116680545B (en) | 2023-08-02 | 2023-08-02 | Coal mine well exit personnel prediction method based on Markov random field |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310961610.7A CN116680545B (en) | 2023-08-02 | 2023-08-02 | Coal mine well exit personnel prediction method based on Markov random field |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116680545A CN116680545A (en) | 2023-09-01 |
CN116680545B true CN116680545B (en) | 2023-10-20 |
Family
ID=87785857
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310961610.7A Active CN116680545B (en) | 2023-08-02 | 2023-08-02 | Coal mine well exit personnel prediction method based on Markov random field |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116680545B (en) |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104899664A (en) * | 2015-06-17 | 2015-09-09 | 西南石油大学 | Drilling well risk prediction method based on Markov chain and Bayesian network |
CN105825297A (en) * | 2016-03-11 | 2016-08-03 | 山东大学 | Markov-model-based position prediction method |
WO2019128184A1 (en) * | 2017-12-26 | 2019-07-04 | 中煤科工集团重庆研究院有限公司 | Calculation method and testing system for cumulative dust exposure to respirable dust of mine workers |
CN112257745A (en) * | 2020-09-11 | 2021-01-22 | 煤炭科学技术研究院有限公司 | Hidden Markov-based method and device for predicting health degree of underground coal mine system |
RU2759071C1 (en) * | 2021-02-12 | 2021-11-09 | Акционерное общество "СУЭК-Кузбасс" | Automated mine emergency forecasting system and method for automated mine emergency forecasting |
WO2022007754A1 (en) * | 2020-07-10 | 2022-01-13 | 炬星科技(深圳)有限公司 | Worker position estimation method and device, and storage medium |
CN114973320A (en) * | 2022-05-17 | 2022-08-30 | 中国矿业大学 | Underground coal mine personnel detection method based on depth information |
US11500117B1 (en) * | 2021-09-13 | 2022-11-15 | Institute Of Geology And Geophysics, Chinese Academy Of Sciences | Method and system for evaluating filling characteristics of deep paleokarst reservoir through well-to-seismic integration |
CN116029331A (en) * | 2023-01-10 | 2023-04-28 | 西安外事学院 | Big data crowd sensing collaborative prediction method for safety state of coal mine working face |
CN116524540A (en) * | 2023-05-04 | 2023-08-01 | 中煤科工集团重庆研究院有限公司 | Coal mine underground people counting method based on bipartite graph |
-
2023
- 2023-08-02 CN CN202310961610.7A patent/CN116680545B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104899664A (en) * | 2015-06-17 | 2015-09-09 | 西南石油大学 | Drilling well risk prediction method based on Markov chain and Bayesian network |
CN105825297A (en) * | 2016-03-11 | 2016-08-03 | 山东大学 | Markov-model-based position prediction method |
WO2019128184A1 (en) * | 2017-12-26 | 2019-07-04 | 中煤科工集团重庆研究院有限公司 | Calculation method and testing system for cumulative dust exposure to respirable dust of mine workers |
WO2022007754A1 (en) * | 2020-07-10 | 2022-01-13 | 炬星科技(深圳)有限公司 | Worker position estimation method and device, and storage medium |
CN112257745A (en) * | 2020-09-11 | 2021-01-22 | 煤炭科学技术研究院有限公司 | Hidden Markov-based method and device for predicting health degree of underground coal mine system |
RU2759071C1 (en) * | 2021-02-12 | 2021-11-09 | Акционерное общество "СУЭК-Кузбасс" | Automated mine emergency forecasting system and method for automated mine emergency forecasting |
US11500117B1 (en) * | 2021-09-13 | 2022-11-15 | Institute Of Geology And Geophysics, Chinese Academy Of Sciences | Method and system for evaluating filling characteristics of deep paleokarst reservoir through well-to-seismic integration |
CN114973320A (en) * | 2022-05-17 | 2022-08-30 | 中国矿业大学 | Underground coal mine personnel detection method based on depth information |
CN116029331A (en) * | 2023-01-10 | 2023-04-28 | 西安外事学院 | Big data crowd sensing collaborative prediction method for safety state of coal mine working face |
CN116524540A (en) * | 2023-05-04 | 2023-08-01 | 中煤科工集团重庆研究院有限公司 | Coal mine underground people counting method based on bipartite graph |
Non-Patent Citations (3)
Title |
---|
基于无偏灰色马尔科夫模型的煤矿事故死亡人数预测;李红霞;车丹丹;李琰;;煤矿安全(第01期);全文 * |
基于马尔可夫链的公交停靠站乘客数预测;刘哲华;;黑龙江交通科技(第11期);全文 * |
基于马尔科夫链的煤矿灾变预测系统的建模研究;郭秋敏;徐博;;工矿自动化(第07期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN116680545A (en) | 2023-09-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20220092418A1 (en) | Training method for air quality prediction model, prediction method and apparatus, device, program, and medium | |
CN102316496A (en) | Data merging method based on Kalman filtering in wireless sensor network | |
Peng et al. | Hierarchical edge computing: A novel multi-source multi-dimensional data anomaly detection scheme for industrial Internet of Things | |
CN103854072A (en) | Processing method and system for path selection | |
CN104469833A (en) | Heterogeneous network operation and maintenance management method based on user perception | |
CN104850653A (en) | Short-term tourist traffic and trend prediction system based on streaming data extraction | |
CN109242446A (en) | A kind of city intelligent energy panorama control platform | |
Xia et al. | Traffic prediction based on ensemble machine learning strategies with bagging and lightgbm | |
CN110621026A (en) | Base station flow multi-time prediction method | |
CN112001573B (en) | Production line management method, device, system and storage medium | |
US20200099336A1 (en) | Techniques for forecasting solar power generation | |
CN116866095B (en) | Industrial router with touch panel and standby control method thereof | |
CN116680545B (en) | Coal mine well exit personnel prediction method based on Markov random field | |
CN101763087B (en) | Industrial process dynamic optimization system and method based on nonlinear conjugate gradient method | |
CN101894356A (en) | A kind of open intelligent earth system architecture and implementation method | |
CN117375249B (en) | Intelligent analysis and positioning system and method for lightning protection accident of power distribution network | |
Zhu et al. | Ensemble Methodology: Innovations in Credit Default Prediction Using LightGBM, XGBoost, and LocalEnsemble | |
Abdulzahra MSc et al. | Energy conservation approach of wireless sensor networks for IoT applications | |
Yu et al. | Integration of wireless sensor network and IoT for smart environment monitoring system | |
CN116307291A (en) | Distributed photovoltaic power generation prediction method and prediction terminal based on wavelet decomposition | |
Samaranayake et al. | Learning the dependency structure of highway networks for traffic forecast | |
Qu et al. | Temporal-spatial collaborative prediction for lte-r communication quality based on deep learning | |
CN114019831A (en) | Water resource monitoring Internet of things platform | |
CN104850657B (en) | A kind of rate addition method of holographic situational map | |
CN102065449A (en) | Method for predicting mobile communication telephone traffic based on clustered LS-SVM (Least Squares-Support Vector Machine) |
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 |