Ventilation system operation state prediction method based on Markov chain

Through the Markov chain method, combined with historical maintenance data and simulation, a state transition probability matrix of ventilation system is established, which solves the complexity and prediction problems of ventilation system reliability assessment in the existing technology, and achieves rapid ventilation systems for large mines , accurate reliability assessment and future status prediction.

CN118939959APending Publication Date: 2024-11-12LIAONING TECHNICAL UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410914420.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-09
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

The existing ventilation system reliability assessment method fails to effectively consider the complexity of the mine ventilation network and the strong coupling between the systems, making it difficult to achieve fast and real-time reliability assessment and future state prediction, and the existing methods are mainly applicable to Small ventilation networks cannot meet the needs of large systems.

Method used

Using the Markov chain-based method, the average fault interval time and repair time are calculated by obtaining historical maintenance data, simulated state transition, established a state transition probability matrix, and predict the future state of the ventilation system to achieve the evaluation of the overall reliability of the system and predict.

Benefits of technology

It realizes rapid and simple reliability assessment and future status prediction of ventilation systems, can grasp the system change trends from a macro perspective, provide a basis for operation and regulation strategies, and is suitable for complex large mine ventilation systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118939959A_ABST
    Figure CN118939959A_ABST
Patent Text Reader

Abstract

The invention discloses a ventilation system operation state prediction method based on a Markov chain, and relates to the technical field of mine ventilation. According to the method, the whole ventilation system is taken as a research object, the reliability of the whole ventilation system is not determined by analyzing the reliability of each component unit of the ventilation system one by one, and the operation process of the ventilation system is regarded as a Markov process by starting with the operation state of the whole system; the evaluation problem of the reliability of the ventilation system is converted into a random mathematical problem; according to the method for evaluating the reliability of the ventilation system in the operating environment from the probability perspective, the change trend of the ventilation system can be mastered macroscopically, and a basis is provided for state forecasting and operation regulation and control strategy making of the ventilation system.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of mine ventilation, and in particular to a method for predicting the operating state of a ventilation system based on a Markov chain. Background Art

[0002] The ventilation system has the characteristics of dynamics, randomness and "no memory". Predicting the future state based on the current state of the ventilation system is of great significance to ensure the safe, reliable and stable operation of the ventilation system and failure warning. The ventilation system is mainly composed of three core elements: ventilation network, ventilation power facilities and ventilation structures. They are interrelated and interdependent, and together determine the operating efficiency of the entire ventilation system. In the process of studying the reliability of the ventilation system, it is usually necessary to first conduct in-depth research and analysis on these three components, and then explore the internal connection and mutual influence between them, and finally comprehensively evaluate the overall reliability level of the ventilation system.

[0003] The existing ventilation system reliability assessment methods have not taken into account that the mine ventilation network is a complex mixed system, and only regard each component as an independent individual that does not affect each other, which is inconsistent with the actual situation. Secondly, the factors affecting the reliability of the ventilation system are complex and diverse, and it is difficult to fully consider the impact of each factor on reliability. Finally, most reliability assessment methods are based on different evaluation indicators to divide the reliability level, which is cumbersome to calculate and is only applicable to small ventilation networks. In addition, when analyzing the reliability of the ventilation system, the focus is mostly on the current ventilation system reliability judgment, and it is impossible to achieve real-time and rapid assessment of the reliability of the ventilation system. Therefore, it is necessary to seek a method that takes the ventilation system as a whole as the research object, starts from the overall operating status of the system, and evaluates the reliability of the ventilation system under the operating environment from a probabilistic perspective, so as to grasp the changing trend of the ventilation system from a macro perspective and provide a basis for the formulation of ventilation system status forecast and operation control strategy. Summary of the invention

[0004] In view of the above problems, the present invention provides a ventilation system reliability assessment method based on Markov chain, which comprehensively considers the dynamics, randomness and strong coupling of the system, simply and quickly realizes the reliability assessment of the ventilation system, and can predict the state of the system for a period of time in the future.

[0005] The technical solution of the present invention is:

[0006] A ventilation system operating state prediction method based on Markov chain specifically comprises the following steps:

[0007] Step 1: Obtain historical maintenance data of the mine, and calculate the mean time between failures (MTBF) and mean time to repair (MTTR) of the mine ventilation system based on the historical maintenance data of the mine; the historical maintenance data of the mine includes the maintenance cycle and repair time of each facility in the mine ventilation system; the types of facilities include structures, tunnels and ventilation power facilities, and each type of facility includes several sub-type facilities;

[0008] Step 1.1: Calculate the average maintenance cycle and mean repair time for each sub-type of facilities in each type of facility;

[0009] Specifically: Set the type of a facility in the mine ventilation system to X, the subtype to x, and count the number of facilities of this subtype X c , calculate the average maintenance cycle of this subtype of facilities and mean time to repair The calculation formula is:

[0010]

[0011] Where X is the type, including structure S, tunnel R and ventilation power facility F; c is the subtype number, including the subtype number i in the structure * , the number of the subtype in the lane j * and the number of subtypes in ventilation power facilities k * ; is the maintenance period of the lth facility among the facilities of subtype x;

[0012]

[0013] In the formula, is the repair time of the lth facility among the facilities of subtype x;

[0014] Step 1.2: Based on the average maintenance cycle and average repair time of each sub-type facility, calculate the average working time between two adjacent faults and the average repair time of each fault of the entire mine ventilation system, that is, the mean time between failures MTBF and the mean time to repair MTTR. The calculation formula is:

[0015]

[0016] Where n S is the total number of subtypes in the structure, The i-th * The average maintenance cycle of sub-type facilities; n R is the total number of subtypes in the lane, is the jth * The average maintenance cycle of sub-type facilities; nF is the total number of subtypes in ventilation power facilities, For the kth ventilation power facility * The average maintenance cycle for each sub-type of facilities; The i-th * The average repair time for each subtype of facility, is the jth * The average repair time for each subtype of facility, For the kth ventilation power facility * The average repair time for each sub-type of facility;

[0017] Step 2: Calculate the failure rate and maintenance rate of the mine ventilation system based on the mean time between failures (MTBF) and mean time to repair (MTTR) of the mine ventilation system;

[0018] Specifically: Assume that the failure rate λ and maintenance rate μ of the mine ventilation system are both constants, and the calculation formula is as follows:

[0019]

[0020] In the formula, λ is the failure rate of the mine ventilation system, and μ is the maintenance rate of the mine ventilation system;

[0021] Step 3: Divide the operating status of the mine ventilation system to obtain several operating statuses;

[0022] Step 4: Simulate the operation state of the mine ventilation system according to the failure rate and maintenance rate of the mine ventilation system to obtain state transition samples and the number of state transition samples between various operation states; the state transition samples represent the operation state transition of the mine ventilation system at a time interval;

[0023] Specifically, the sequential Monte Carlo method is used to simulate the operation status of the mine ventilation system. The time interval Δt is set. The failure rate of the mine ventilation system within the Δt time interval is λ and the maintenance rate is μ. K simulations are performed, and each simulation is T Δt time interval;

[0024] Step 5: According to the state transition samples and the number of state transition samples, the distribution of the operating state of the mine ventilation system at each time section and the operating state transition of adjacent time sections are counted to obtain the Markov state transition probability matrix;

[0025] Specifically, the Markov chain of finite discrete states is represented by a directed graph, and the operating state of the mine ventilation system is represented by E = {e1, ..., e M}, set at time interval Δt ij The internal mine ventilation system is operated by iTransition to running state j The probability is P ij , then the state transition probability matrix P of the mine ventilation system can be expressed as:

[0026]

[0027] Wherein, M is the number of divided operating states;

[0028] And satisfy the constraints:

[0029]

[0030] The random process of the state change of the mine ventilation system is assumed to be a time-homogeneous Markov chain, that is:

[0031] p[Y(t+Δt ij )=e j |Y(t)=e i ]=p[Y(Δt ij )=e j |Y(0)=e i ]=p ij (Δt ij )

[0032] In the formula, Y(t+Δt ij ) is t+Δt ij The running state at that time, Y(t) is the current running state, and Y(Δt ij ) is Δt ij The running state at time , Y(0) is the initial state, and t is the current time;

[0033] If the time interval Δt ij Similarly, the state transition probability matrix of the mine ventilation system is simplified to:

[0034]

[0035] When the number of state transfer samples tends to infinity, the statistical value of the state transfer probability matrix is ​​infinitely close to the true value, so the state transfer probability matrix is ​​expressed as:

[0036]

[0037] Among them, n ij Indicates that the initial state is e i The transition state after the time interval Δt is e j The number of state transition samples; N i Indicates that the initial state is e i The number of state transition samples;

[0038] Step 6: Obtain the initial state of the mine ventilation system, use the state transition probability matrix of the mine ventilation system to obtain the future stable state probability of the mine ventilation system, and obtain the operation state prediction result of the mine ventilation system according to the stable state probability, so as to realize the current reliability evaluation and future state prediction of the ventilation system;

[0039] Specifically, the initial state of the ventilation system is set to φ(0). According to the definition of the state transition probability matrix P of the mine ventilation system, after Δt time, the operating state of the mine ventilation system at the next moment is:

[0040] φ(1)=φ(0)P

[0041] In the formula, φ(1) is the operating state of the mine ventilation system after Δt time from the initial state;

[0042] The operating status of the mine ventilation system after m Δt is:

[0043] φ(m)=φ(m-1)P=φ(0)P m

[0044] Where, φ(m) is the operating state of the mine ventilation system after m Δt; φ(m-1) is the operating state of the mine ventilation system after m-1 Δt;

[0045] When the number of time intervals m→∞, the operating state of the mine ventilation system gradually approaches a certain stable value, which is called the steady state probability matrix or long-term state probability matrix of the mine ventilation system;

[0046]

[0047] Where φ(∞) is the steady state probability matrix;

[0048] The element φ in the steady-state probability matrix i (∞) is the probability of a stable state, indicating that no matter what the initial state of the mine ventilation system is, after a long enough time, it will be in state e i The probability of i (∞), and:

[0049]

[0050] The operating state with the maximum probability of stable state is the prediction result of the mine ventilation system transfer operating state at the next moment, thereby realizing the evaluation of the ventilation system reliability.

[0051] Compared with the prior art, the present invention has the following advantages:

[0052] The present invention provides a method that takes the ventilation system as a whole as the research object, and no longer analyzes the reliability of each component of the ventilation system one by one to determine the reliability of the ventilation system as a whole. The method starts with the overall operating state of the system, regards the operating process of the ventilation system as a Markov process, and transforms the evaluation problem of the reliability of the ventilation system into a random mathematical problem; a method for evaluating the reliability of the ventilation system under the operating environment from a probabilistic perspective, so as to grasp the changing trend of the ventilation system from a macro perspective, and provide a basis for the formulation of ventilation system state forecasting and operation control strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a flow chart of a ventilation system operation state prediction method based on Markov chain in an example of the present invention;

[0054] Figure 2 It is a schematic diagram of the state transition of the ventilation system in an example of the present invention. DETAILED DESCRIPTION

[0055] The invention will be further described below with reference to the accompanying drawings and specific implementation examples.

[0056] A ventilation system operation status prediction method based on Markov chain, the process is as follows Figure 1 As shown, the specific steps include:

[0057] Step 1: Obtain historical maintenance data of the mine, and calculate the mean time between failures (MTBF) and mean time to repair (MTTR) of the mine ventilation system based on the historical maintenance data of the mine; the historical maintenance data of the mine includes the maintenance cycle and repair time of each facility in the mine ventilation system; the types of facilities include structures, tunnels and ventilation power facilities, and each type of facility includes several sub-type facilities;

[0058] Step 1.1: Calculate the average maintenance cycle and mean repair time for each sub-type of facilities in each type of facility;

[0059] Specifically: Assume that the type of a facility in the mine ventilation system is X (structure, tunnel or ventilation power facility), and the subtype is x, and count the number of facilities of this subtype X c , calculate the average maintenance cycle of this subtype of facilities and mean time to repair The calculation formula is:

[0060]

[0061] Where X is the type, including structure S, tunnel R and ventilation power facility F; c is the subtype number, including the subtype number i in the structure * , the number of the subtype in the lane j* and the number of subtypes in ventilation power facilities k * ; is the maintenance period of the lth facility among the facilities of subtype x;

[0062]

[0063] In the formula, is the repair time of the lth facility among the facilities of subtype x;

[0064] In this embodiment, the structures include permanent adjustable air doors, positive and negative air doors, permanent windbreak walls and other sub-types; the lanes include semicircular arch anchor spray return air lanes, semicircular arch U-shaped shed track lanes, rectangular trapezoidal shed belt lanes and other sub-types; the ventilation power facilities include ventilation power equipment, ventilation pipes and other sub-types;

[0065] Step 1.2: As shown in Table 1, based on the average maintenance cycle and average repair time of each sub-type facility, calculate the average working time between two adjacent failures and the average repair time of each failure of the entire mine ventilation system, that is, the mean time between failures (MTBF) and the mean time to repair (MTTR). The calculation formula is:

[0066]

[0067] Where n S is the total number of subtypes in the structure, The i-th * The average maintenance cycle of sub-type facilities; n R is the total number of subtypes in the lane, is the jth * The average maintenance cycle of sub-type facilities; n F is the total number of subtypes in ventilation power facilities, For the kth ventilation power facility * The average maintenance cycle for each sub-type of facilities; The i-th * The average repair time for each subtype of facility, is the jth * The average repair time for each subtype of facility, For the kth ventilation power facility * The average repair time for each sub-type of facility;

[0068] Table 1 Statistics of average maintenance cycle and average repair time of each sub-type of facilities in mine ventilation system

[0069]

[0070]

[0071] In the table: Indicates the total number of structures; Indicates the total number of lanes; Indicates the total number of ventilation power facilities;

[0072] Step 2: Calculate the failure rate and maintenance rate of the mine ventilation system based on the mean time between failures (MTBF) and mean time to repair (MTTR) of the mine ventilation system;

[0073] Specifically: due to the combined effects of factors such as the geographical location, nature of the mine, and environment of the mine, as well as different mining stages in the mine life cycle, the failure rate λ(t) of the mine ventilation system is also different; due to factors such as the type of fault, degree of fault, and technical level of maintenance technicians, the maintenance rate μ(t) is also different. Therefore, when studying the reliability of the mine ventilation system, the present invention sets the failure rate λ and maintenance rate μ of the mine ventilation system to be constants, that is, λ(t) = λ, μ(t) = μ. The calculation formula is as follows:

[0074]

[0075] In the formula, λ is the failure rate of the mine ventilation system, and μ is the maintenance rate of the mine ventilation system;

[0076] Step 3: Divide the operating status of the mine ventilation system to obtain several operating statuses;

[0077] Specifically: Figure 2 As shown in the figure, the mine ventilation system is divided into three operating states: e1, e2, and e3. Among them, e1 is the normal state, indicating that there is no fault in the mine ventilation system, each device is in good operating condition, and each parameter is within the specified range; e2 is the risk state, indicating that some constraints in the mine ventilation system cannot be met, but the mine ventilation system will not cause large-scale damage after the use of spare equipment or readjustment, such as damaged dampers, deformed tunnels, temporary blockage of transport vehicles, wear or damage of fan blades, etc. e3 is the fault state, indicating that a fault occurs in the mine ventilation system, causing a certain degree of casualties or property losses, and large-scale and long-term adjustments are required to restore the system to a safe and stable operating state, such as roof collapse, accidental opening of the fan and no recovery for a long time, complete damage to the damper, and serious blockage of the tunnel.

[0078] Step 4: Simulate the operation state of the mine ventilation system according to the failure rate and maintenance rate of the mine ventilation system to obtain state transition samples and the number of state transition samples between various operation states; the state transition samples represent the operation state transition of the mine ventilation system at a time interval;

[0079] Specifically, the sequential Monte Carlo method is used to simulate the operation status of the mine ventilation system. The time interval Δt is set to one day. The failure rate of the mine ventilation system within the Δt time interval is λ and the maintenance rate is μ. K simulations are performed, and each simulation is T Δt The number of samples transferred between various operating states is shown in Table 2.

[0080] Table 2 Number of state transition samples

[0081]

[0082] In the table, e i Indicates the running status, n ij Indicates that the initial state is e i The transition state after the time interval Δt is e j The number of state transition samples; N i Indicates that the initial state is e i The first row indicates that if the system is in the normal state e1 at this moment, the number of samples remaining in the normal state e1 after the time interval Δt is n 11 , n 12 samples are transferred to risk state e2, n 13 The second line indicates that if the system is in the risk state e2 at this moment, there will be n samples transferred to the fault state e3 after the time interval Δt. 22 samples remain in risk state e2; n 21 The number of samples is transferred to the normal state e1, n 23 The third line indicates that if the system is in fault state e3 at this moment, there will be n samples after time interval Δt. 33 The number of samples remains in the fault state e3, with n 31 The number of samples is transferred to the normal state e1, and there are n 32 The number of samples is transferred to risk state e2;

[0083] Step 5: According to the state transition samples and the number of state transition samples, the distribution of the operating state of the mine ventilation system at each time section and the operating state transition of adjacent time sections are counted to obtain the Markov state transition probability matrix;

[0084] Specifically, the Markov chain of finite discrete states is represented by a directed graph, and the three operating states of the mine ventilation system are represented by E = {e1, e2, e3}, which are set at the time interval Δt ij The internal mine ventilation system is operated by i Transition to running state j The probability is P ij , that is: the probability of the running state e1 changing to the running state e2 is P 12, the probability of the running state e1 changing to the running state e3 is P 13 ……, then the state transition probability matrix P of the mine ventilation system can be expressed as:

[0085]

[0086] And satisfy the constraints:

[0087]

[0088] The random process of the state change of the mine ventilation system is assumed to be a time-homogeneous Markov chain, that is:

[0089] p[Y(t+Δt ij )=e j |Y(t)=e i ]=p【Y(Δt ij )=e j |Y(0)=e i ]=p ij (Δt ij )

[0090] In the formula, Y(t+Δt ij ) is t+Δt ij The running state at that time, Y(t) is the current running state, and Y(Δt ij ) is Δt ij The running state at time , Y(0) is the initial state;

[0091] If the time interval Δt ij Similarly, the state transition probability matrix of the mine ventilation system can be simplified as:

[0092]

[0093] When the number of state transition samples is large enough, the state transition probability matrix of the mine ventilation system is calculated by the number of state transition samples n between each operating state. ij The proportion of the total number of samples is obtained. When the number of samples tends to infinity, the statistical value of the state transition probability matrix is ​​infinitely close to the true value, so the state transition probability matrix can be expressed as:

[0094]

[0095] Step 6: When the initial state of the mine ventilation system is known, the state transition probability matrix of the mine ventilation system is used to quickly analyze and obtain the future stable state probability of the mine ventilation system. According to the stable state probability, the operation state prediction result of the mine ventilation system is obtained to realize the current reliability evaluation and future state prediction of the ventilation system.

[0096] Specifically, the initial state of the ventilation system is set to φ(0). According to the definition of the state transition probability matrix P of the mine ventilation system, after Δt time, the operating state of the mine ventilation system at the next moment is:

[0097] φ(1)=φ(0)P

[0098] In the formula, φ(1) is the operating state of the mine ventilation system after Δt time from the initial state;

[0099] The operating status of the mine ventilation system after m Δt is:

[0100] φ(m)=φ(m-1)P=φ(0)P m

[0101] Where, φ(m) is the operating state of the mine ventilation system after m Δt; φ(m-1) is the operating state of the mine ventilation system after m-1 Δt;

[0102] When the number of time intervals m→∞, the operating state of the mine ventilation system gradually approaches a certain stable value, which is called the steady state probability matrix or long-term state probability matrix of the mine ventilation system;

[0103]

[0104] Where φ(∞) is the steady state probability matrix;

[0105] The element φ in the steady-state probability matrix i (∞) is the probability of a stable state, indicating that no matter what the initial state of the mine ventilation system is, after a long enough time, it will be in state e i The probability of i (∞), and:

[0106]

[0107] At this time, the calculated π1(∞), π2(∞), π3(∞) values ​​are the system steady state probability or long-term state probability distribution, which can be used to evaluate the current reliability of the ventilation system; the maximum value among π1(∞), π2(∞), π3(∞) is the prediction result of the system transferring to the next state at the next moment, thereby realizing the evaluation of the reliability of the ventilation system.

Claims

1. A ventilation system operating state prediction method based on Markov chain, characterized in that: The steps include: Step 1: Obtain historical maintenance data of the mine, and calculate the mean time between failures (MTBF) and mean time to repair (MTTR) of the mine ventilation system based on the historical maintenance data of the mine; the historical maintenance data of the mine includes the maintenance cycle and repair time of each facility in the mine ventilation system; the types of facilities include structures, tunnels and ventilation power facilities, and each type of facility includes several sub-type facilities; Step 2: Calculate the failure rate and maintenance rate of the mine ventilation system based on the mean time between failures (MTBF) and mean time to repair (MTTR) of the mine ventilation system; Step 3: Divide the operating status of the mine ventilation system to obtain several operating statuses; Step 4: Simulate the operation state of the mine ventilation system according to the failure rate and maintenance rate of the mine ventilation system to obtain state transition samples and the number of state transition samples between various operation states; the state transition samples represent the operation state transition of the mine ventilation system at a time interval; Step 5: According to the state transition samples and the number of state transition samples, the distribution of the operating state of the mine ventilation system at each time section and the operating state transition of adjacent time sections are counted to obtain the Markov state transition probability matrix; Step 6: Obtain the initial state of the mine ventilation system, use the state transition probability matrix of the mine ventilation system to obtain the future stable state probability of the mine ventilation system, and obtain the operating state prediction result of the mine ventilation system based on the stable state probability, so as to realize the current reliability evaluation and future state prediction of the ventilation system.

2. A ventilation system operating state prediction method based on Markov chain according to claim 1, characterized in that: Step 1 specifically includes: Step 1.1: Calculate the average maintenance cycle and mean repair time for each sub-type of facilities in each type of facility; Specifically: Set the type of a facility in the mine ventilation system to X, the subtype to x, and count the number of facilities of this subtype X c , calculate the average maintenance cycle of this subtype of facilities and mean time to repair The calculation formula is: Where X is the type, including structure S, tunnel R and ventilation power facility F; c is the subtype number, including the subtype number i in the structure * , the number of the subtype in the lane j * and the number of subtypes in ventilation power facilities k * ; is the maintenance period of the lth facility among the facilities of subtype x; In the formula, is the repair time of the lth facility among the facilities of subtype x; Step 1.2: Based on the average maintenance cycle and average repair time of each sub-type facility, calculate the average working time between two adjacent faults and the average repair time of each fault of the entire mine ventilation system, that is, the mean time between failures MTBF and the mean time to repair MTTR. The calculation formula is: Where n S is the total number of subtypes in the structure, The i-th * The average maintenance cycle of sub-type facilities; nR is the total number of sub-types in the lane, is the jth * The average maintenance cycle of sub-type facilities; nF is the total number of sub-types in ventilation power facilities, The kth ventilation power facility * The average maintenance cycle for each sub-type of facilities; The i-th * The average repair time for each subtype of facility, is the jth * The average repair time for each subtype of facility, The kth ventilation power facility * The average repair time for each subtype of facility.

3. A ventilation system operating state prediction method based on Markov chain according to claim 1, characterized in that: Step 2 is as follows: Assuming that the failure rate λ and maintenance rate μ of the mine ventilation system are both constants, the calculation formula is as follows: Where λ is the failure rate of the mine ventilation system, and μ is the maintenance rate of the mine ventilation system.

4. A ventilation system operating state prediction method based on Markov chain according to claim 1, characterized in that: Step 4 is as follows: the sequential Monte Carlo method is used to simulate the operation status of the mine ventilation system, and a time interval Δt is set. The failure rate of the mine ventilation system within the Δt time interval is λ and the maintenance rate is μ. K simulations are performed, and each simulation is T. Δt time interval.

5. The method for predicting the operating state of a ventilation system based on a Markov chain according to claim 1, characterized in that: Step 5 is as follows: the Markov chain of finite discrete states is represented by a directed graph, and the operating state of the mine ventilation system is represented by E = {e1, ..., e M }, set at time interval Δt ij The internal mine ventilation system is operated by i Transition to running state j The probability is P ij , then the state transition probability matrix P of the mine ventilation system can be expressed as: Wherein, M is the number of divided operating states; And satisfy the constraints: The random process of the state change of the mine ventilation system is assumed to be a time-homogeneous Markov chain, that is: p[Y(t+Δt ij )=e j |Y(t)=e i ]=p[Y(Δt ij )=e j |Y(0)=e i ]=p ij (Δt ij ) In the formula, Y(t+Δt ij ) is t+Δt ij The running state at that time, Y(t) is the current running state, and Y(Δt ij ) is Δt ij The running state at time , Y(0) is the initial state, and t is the current time; If the time interval Δt ij Similarly, the state transition probability matrix of the mine ventilation system is simplified to: When the number of state transfer samples tends to infinity, the statistical value of the state transfer probability matrix is ​​infinitely close to the true value, so the state transfer probability matrix is ​​expressed as: Among them, n ij Indicates that the initial state is e i The transition state after the time interval Δt is e j The number of state transition samples; N i Indicates that the initial state is e i The number of state transition samples.

6. A ventilation system operating state prediction method based on Markov chain according to claim 1, characterized in that: Step 6 is as follows: Specifically, the initial state of the ventilation system is set to φ(0). According to the definition of the state transition probability matrix P of the mine ventilation system, after Δt time, the operating state of the mine ventilation system at the next moment is: φ(1)=φ(0)P In the formula, φ(1) is the operating state of the mine ventilation system after Δt time from the initial state; The operating status of the mine ventilation system after m Δt is: φ(m)=φ(m-1)P=φ(0)P m Where, φ(m) is the operating state of the mine ventilation system after m Δt; φ(m-1) is the operating state of the mine ventilation system after m-1 Δt; When the number of time intervals m→∞, the operating state of the mine ventilation system gradually approaches a certain stable value, which is called the steady state probability matrix or long-term state probability matrix of the mine ventilation system; Where φ(∞) is the steady state probability matrix; The element φ in the steady-state probability matrix i (∞) is the probability of a stable state, indicating that no matter what the initial state of the mine ventilation system is, after a long enough time, it will be in state e i The probability of i (∞), and: The operating state with the maximum probability of stable state is the prediction result of the mine ventilation system transfer operating state at the next moment, thereby realizing the evaluation of the ventilation system reliability.