Intelligent management and control system for multi-parameter fusion of coal mine bunker

Abstract

The present application relates to the technical field of data processing, and more particularly to a kind of intelligent management and control system of coal mine coal bunker multi-parameter fusion, the system includes processor and memory, processor executes the computer program of memory, to realize the following steps: in the process of monitoring coal wall state using D-S evidence theory, obtain each evidence source for monitoring coal wall state, according to the coupling relationship between material level and material weight, and the volume of material attached to the wall and the wall vibration signal, the quality function in the process of iterative fusion of each evidence source is optimized, and the optimized quality function is obtained, according to the optimized quality function, each evidence source is iteratively fused, and the fusion probability of coal wall in coal mine coal bunker at current time is obtained;According to the fusion probability of coal wall at current time and each historical time, the intelligent management and control of coal mine coal bunker is realized, and the accuracy of intelligent management and control of coal mine coal bunker is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to an intelligent management and control system for a coal mine bunker based on multi-parameter fusion. BACKGROUND

[0002] In the production process of a coal mine, the coal bunker of the coal mine undertakes important links such as temporary storage, buffering and adjustment of the amount of raw coal. The safe and stable operation of the coal bunker of the coal mine directly affects the continuity of the mining equipment and the efficiency of the entire transportation system. Among them, coal wall hanging is a common problem in the coal bunker of the coal mine, which has a multi-faceted impact on the safe and stable operation of the coal bunker of the coal mine. For example, when wall hanging is severe, it may form a "rat hole" or arch, causing uneven discharge or even flow interruption, which requires manual intervention to dredge and affects continuous production. The wall-hung material may suddenly fall off in large areas, causing impact on the structure of the bunker, and even causing equipment damage or personnel injury. Long-term adhesion of coal may cause spontaneous combustion due to oxidation and heating. Therefore, accurate identification and early warning of the state of coal wall hanging is the key to ensuring stable operation of production.

[0003] Currently, the industry usually deploys sensors such as level sensors, weight sensors, vibration sensors and industrial cameras at various positions of the coal bunker of the coal mine, and uses D-S evidence theory to fuse the information of various sensors to obtain the probability of coal wall hanging, and then realizes intelligent management and control of the coal bunker of the coal mine.

[0004] However, in actual industrial applications, phenomena such as "suspended material", "puffy material" and "blockage" caused by softening, caking, uneven particle size and moisture changes of coal may cause contradictory data to be collected by various sensors. For example, the coal position detected by the level sensor is very high, showing a "full bunker" state, but the weight detected by the weight sensor is decreasing, showing an "empty bunker" state. At this time, the states of the bunker shown by the level sensor and the weight sensor are seriously inconsistent. These contradictory data can greatly increase the conflict coefficient in the D-S evidence theory, resulting in a highly concentrated quality function in a physically unreasonable proposition, or a highly dispersed distribution that cannot be interpreted, and thus cannot distinguish between real coal wall hanging and empty bunker, affecting the accuracy of intelligent management and control of the coal bunker of the coal mine.

[0005] Therefore, how to use D-S evidence theory to improve the accuracy of identifying the state of coal wall hanging, and thus improve the accuracy of intelligent management and control of the coal bunker of the coal mine, has become a problem to be solved. SUMMARY

[0006] Therefore, the embodiments of the present application provide an intelligent management and control system for a coal mine bunker based on multi-parameter fusion to solve the problem of improving the accuracy of identifying the state of coal wall hanging using D-S evidence theory, and thus improving the accuracy of intelligent management and control of the coal bunker of the coal mine.

[0007] The embodiment of the application provides a kind of intelligent management and control system of coal mine coal bunker multi-parameter fusion, including memory, processor and the computer program stored in memory and running on processor, characterized in that, when processor executes computer program, the following steps are realized:

[0008] In the process of monitoring the coal wall hanging state of the coal mine coal bunker at the current time by using D-S evidence theory, each evidence source for monitoring the coal wall hanging state in the coal mine coal bunker is obtained, the evidence source includes the material level in the coal mine coal bunker, the material weight, the volume of material adhesion of wall and the wall vibration signal;

[0009] According to the monitoring data of each evidence source at the current time, the coupling relationship between the material level and the material weight, and the coupling relationship between the volume of material adhesion of wall and the wall vibration signal are analyzed, and the wall hanging characteristic index of the coal mine coal bunker at the current time is obtained;

[0010] In the process of iteratively fusing each evidence source by using Dempster combination rule, the wall hanging characteristic index is used to optimize all mass functions at each iteration fusion, and all optimized mass functions at each iteration fusion are obtained, and according to all optimized mass functions at each iteration fusion, each evidence source is iteratively fused, and the fusion probability of coal wall hanging in the coal mine coal bunker at the current time is obtained;

[0011] The fusion probability of coal wall hanging in the coal mine coal bunker at a preset number of historical times before the current time is obtained, and the coal wall hanging state of the coal mine coal bunker at the current time is intelligently controlled according to the fusion probability of coal wall hanging in the coal mine coal bunker at the current time and each historical time.

[0012] Preferably, according to the monitoring data of each evidence source at the current time, the coupling relationship between the material level and the material weight, and the coupling relationship between the volume of material adhesion of wall and the wall vibration signal are analyzed, and the wall hanging characteristic index of the coal mine coal bunker at the current time is obtained, including:

[0013] The monitoring data of each evidence source at the current time is linearly normalized to obtain the normalized value of each evidence source at the current time;

[0014] According to the normalized value of the material level and the material weight at the current time, the first coupling characteristic value between the material level and the material weight is obtained;

[0015] According to the normalized value of the volume of material adhesion of wall and the wall vibration signal at the current time, the second coupling characteristic value between the volume of material adhesion of wall and the wall vibration signal is obtained;

[0016] The first coupling characteristic value between the material level and the material weight is obtained according to a normalized value of the material level and the material weight at the current moment.

[0017] Preferably, the first coupling characteristic value between the material level and the material weight is obtained according to a normalized value of the material level and the material weight at the current moment.

[0018] The first coupling characteristic value between the material level and the material weight is obtained by linearly normalizing a product of an inverse of a sum between a normalized value of the material weight at the current moment and a preset constant and a normalized value of the material level at the current moment.

[0019] Preferably, the second coupling characteristic value between the wall material adhesion volume and the wall vibration signal is obtained according to a normalized value of the wall material adhesion volume and the wall vibration signal at the current moment.

[0020] The second coupling characteristic value between the wall material adhesion volume and the wall vibration signal is obtained by calculating a product between a difference between a constant 1 and a normalized value of the wall vibration signal at the current moment and a normalized value of the wall material adhesion volume at the current moment.

[0021] Preferably, the all optimized quality functions at each iteration fusion are obtained by optimizing all quality functions at each iteration fusion using the wall characteristic index.

[0022] For any quality function at any iteration fusion, an initial conflict coefficient in the any quality function is obtained, denoted as an initial conflict coefficient, a product between a preset strength adjustment factor and the wall characteristic index is calculated, a constant 1 is subtracted from the product to obtain an adjustment coefficient, and a product between the initial conflict coefficient and the adjustment coefficient is calculated to obtain a final conflict coefficient.

[0023] The initial conflict coefficient in the any quality function is replaced by the final conflict coefficient to obtain an optimized quality function of the any quality function.

[0024] Preferably, the coal wall hanging state of the coal mine bunker at the current moment is intelligently controlled according to a fusion probability of coal wall hanging in the coal mine bunker at each historical moment and the current moment.

[0025] A time interval between each historical moment and the current moment is obtained, a preset attenuation factor is taken as a base, and each time interval is taken as an index to obtain an attenuation weight of the fusion probability of coal wall hanging in the coal mine bunker at each historical moment corresponding to each time interval, and the preset attenuation factor is less than a constant 1.

[0026] The constant 1 is taken as the attenuation weight of the fusion probability of the coal wall sticking in the coal mine bunker at the current moment, the weighted average of all fusion probabilities is calculated according to the attenuation weights of all fusion probabilities, and the final probability of the coal wall sticking in the coal mine bunker at the current moment is obtained.

[0027] According to the final probability of the coal wall sticking in the coal mine bunker at the current moment, the intelligent management and control of the coal wall sticking state of the coal mine bunker at the current moment is carried out.

[0028] Preferably, the intelligent management and control of the coal wall sticking state of the coal mine bunker at the current moment according to the final probability of the coal wall sticking in the coal mine bunker at the current moment comprises:

[0029] The preset wall sticking probability threshold corresponding to different preset wall sticking control strategies is obtained, the final probability of the coal wall sticking in the coal mine bunker at the current moment is compared with each preset wall sticking probability threshold, and the preset wall sticking control strategy of the coal mine bunker at the current moment is obtained.

[0030] Compared with the prior art, the embodiment of the present application has the following beneficial effects:

[0031] The present application obtains the wall sticking characteristic index for quantifying the possibility of wall sticking by analyzing the coupling relationship between the material level and the material weight, and the coupling relationship between the material adhesion volume of the bin wall and the bin wall vibration signal. In the process of iterative fusion of each evidence source by using the Dempster combination rule, the wall sticking characteristic index is used to optimize all quality functions at each iteration fusion, effectively reducing the decision-making error caused by the defects of the combination rule in the traditional D-S evidence theory in the high conflict scene, obtaining all optimized quality functions with stronger anti-interference ability at each iteration fusion, and then obtaining the fusion probability of the coal wall sticking in the coal mine bunker at the current moment, improving the accuracy of identifying the state of the coal wall sticking by using the D-S evidence theory. At the same time, the fusion probability of the coal wall sticking in the coal mine bunker at the current moment and each historical moment is obtained to combine the time sequence weighted smoothing, suppress the instantaneous interference, obtain a reliable conclusion, and finally realize the accurate identification and robust early warning of the state of the silo wall sticking. BRIEF DESCRIPTION OF DRAWINGS

[0032] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiments or prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0033] Figure 1It is a flow chart of an intelligent management and control method of a coal mine coal bunker multi-parameter fusion provided by an embodiment of the present application. DETAILED DESCRIPTION

[0034] The embodiments of the present disclosure are described in detail below, examples of which are shown in the accompanying drawings. The embodiments described below by referring to the accompanying drawings are exemplary and are intended to explain the present disclosure, and cannot be understood as a limitation of the present disclosure.

[0035] It should be noted that the terms "first", "second", and the like in the specification of the present disclosure and the above drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The implementation described in the following exemplary embodiments does not represent all implementations consistent with the present disclosure. Rather, they are only examples of devices and methods consistent with some aspects of the present disclosure.

[0036] In order to illustrate the technical solutions of the present application, specific embodiments are described below.

[0037] The specific scenario to which the present application is directed is that: the industry usually uses D-S evidence theory to fuse the information of various sensors to obtain the probability of coal hanging wall, and then realizes intelligent management and control of the coal mine coal bunker. However, when each sensor collects contradictory data, the conflict coefficient in the D-S evidence theory will be greatly increased, thereby causing the final quality function to be highly concentrated on a physically unreasonable proposition, or to be extremely dispersed and unable to be interpreted, and thus unable to distinguish between real coal hanging wall and empty bunker, thereby affecting the accuracy of intelligent management and control of the coal mine coal bunker. Therefore, the present application analyzes the coupling relationship between different sensors to obtain a hanging wall characteristic index, which is used to optimize the quality function, so as to greatly improve the accuracy and stability of identifying coal hanging wall using D-S evidence theory, and ultimately realize intelligent management and control of the coal mine coal bunker.

[0038] The present application provides an intelligent management and control system of a coal mine coal bunker multi-parameter fusion, comprising a processor and a memory, the processor executes the computer program stored in the memory to realize an intelligent management and control method of a coal mine coal bunker multi-parameter fusion, as shown in Figure 1 The method comprises the following steps:

[0039] Step S101, in the process of monitoring the coal hanging wall state of the coal mine coal bunker at the current time using D-S evidence theory, obtaining each evidence source for monitoring the coal hanging wall state in the coal mine coal bunker, the evidence source comprising the material level in the coal mine coal bunker, the material weight, the volume of the material adhering to the wall, and the wall vibration signal.

[0040] In the process of monitoring the coal wall-hanging state of the coal bunker at the current time by using the D-S evidence theory, first, an identification framework is constructed The identification framework can be set as a basic two-nominal proposition, that is, {wall-hanging occurs, normal (no wall-hanging)} or a three-nominal proposition, for example, {wall-hanging occurs, normal (no wall-hanging), other faults (non-wall-hanging abnormalities, such as in-out material blockage, mechanical failure, etc.)}. In the embodiment of the present application, {wall-hanging occurs, normal (no wall-hanging)} is taken as the identification framework , which is not limited here, and the implementer can set it according to the specific scene. The construction of the identification framework is a prior art, which will not be described here.

[0041] Then, evidence sources are set. When monitoring the coal wall-hanging, it is usually necessary to measure the material height in the coal bunker, the material weight, the volume of material adhesion on the bunker wall and the vibration signal of the bunker wall. Therefore, a radar material level meter or a laser range finder and other material level measurement sensors are arranged in the coal bunker to collect the material level in the coal bunker; a load cell is arranged in the coal bunker to collect the material weight in the coal bunker; a sound wave / ultrasonic sensor and other vibration sensors are arranged in the coal bunker to collect the vibration signal of the bunker wall of the coal bunker; an infrared thermal imager or an industrial endoscope and other image sensors are arranged in the coal bunker to obtain the volume of material adhesion on the bunker wall by image analysis on the collected images. The material level, the material weight, the volume of material adhesion on the bunker wall and the vibration signal of the bunker wall are taken as the evidence sources when monitoring the coal wall-hanging state of the coal bunker at the current time by using the D-S evidence theory, which are used to perform basic probability assignment (BPA) on all subsets (propositions) in the identification framework to obtain the belief of each evidence source to each subset (proposition) in the identification framework . In the identification framework , the subsets are {wall-hanging occurs}, {normal (no wall-hanging)} and {wall-hanging occurs, normal (no wall-hanging)}. For the convenience of subsequent description, {wall-hanging occurs} is denoted as A, {normal (no wall-hanging)} is denoted as B, and {wall-hanging occurs, normal (no wall-hanging)} is denoted as C. The material level is denoted as , the material weight is denoted as , the volume of material adhesion on the bunker wall is denoted as , and the vibration signal of the bunker wall is denoted as . The belief of the material level to A is represented as , the belief of the material level to B is represented as , and the belief of the material level to C is represented as . The belief of the material weight to A is represented as , the belief of the material weight to B is represented as The confidence of the material weight to C is expressed as By analogy, the basic probability assignment (BPA) is not described here.

[0042] In actual industrial applications, the phenomena of "suspended material", "puffy material", "blockage" and the like caused by softening, caking, uneven particle size, moisture change and the like of coal will cause contradictory data to be collected by various sensors. For example, the coal position detected by a level measurement sensor is very high, showing a "full bin" state, but the weight detected by a weighing sensor is decreasing, showing an "empty bin" state. At this time, the bin states displayed by the measurement sensor and the weighing sensor are seriously inconsistent. These contradictory data will greatly increase the conflict coefficient in the D-S evidence theory, which may cause the final quality sum to be less than 1, output invalid values, or cause serious information loss or distortion, thereby causing the control system to make dangerous decisions based on incorrect confidence, such as false start of equipment, false alarm and the like. Therefore, in the embodiment of the present application, by analyzing the coupling relationship between the contradictory data that may occur, the quality function in the D-S evidence theory is optimized, the accuracy and stability of identifying coal wall hanging using the D-S evidence theory are greatly improved, and a high-confidence state recognition result is provided for intelligent management and control of the coal mine coal bin.

[0043] In step S102, the coupling relationship between the material level and the material weight, and the coupling relationship between the material adhesion volume of the bin wall and the bin wall vibration signal are analyzed according to the monitoring data of each evidence source at the current time, and the wall hanging characteristic index of the coal mine coal bin at the current time is obtained.

[0044] In the set evidence sources, the material level and the material weight may be contradictory. For example, the coal position detected by a level measurement sensor is very high, showing a "full bin" state, but the weight detected by a weighing sensor may be decreasing, showing an "empty bin" state. At this time, the bin states displayed by the measurement sensor and the weighing sensor are seriously inconsistent. The material adhesion volume of the bin wall and the bin wall vibration signal may be contradictory. For example, the image collected by an image sensor shows that the bin wall is smooth and clean, and the material adhesion volume on the bin wall is small, and the visual evidence supports "normal (no wall hanging)", but the bin wall signal detected by a vibration sensor shows abnormal damping increase, inherent frequency offset or new resonance peak, and the vibration evidence supports "wall hanging occurs". Therefore, in the embodiment of the present application, the coupling relationship between the material level and the material weight, and the coupling relationship between the material adhesion volume of the bin wall and the bin wall vibration signal are analyzed, and the wall hanging characteristic index of the coal mine coal bin at the current time is obtained, so as to optimize the quality function in the D-S evidence theory.

[0045] obtain monitoring data of each evidence source at the current moment, wherein the monitoring data of the bin wall vibration signal at the current moment is amplitude, which can also be frequency or energy, and the present disclosure does not limit this, and the implementer can set this according to the specific scene, linearly normalize the monitoring data of each evidence source at the current moment to obtain the normalized value of each evidence source at the current moment;

[0046] calculate the reciprocal of the sum of the normalized value of the material weight at the current moment and the preset constant, linearly normalize the product between the reciprocal and the normalized value of the material level at the current moment to obtain the first coupling characteristic value between the material level and the material weight;

[0047] calculate the difference between the constant 1 and the normalized value of the bin wall vibration signal at the current moment, and calculate the product between the difference and the normalized value of the bin wall material adhesion volume at the current moment to obtain the second coupling characteristic value between the bin wall material adhesion volume and the bin wall vibration signal;

[0048] weight and sum the first coupling characteristic value and the second coupling characteristic value to obtain the wall-hanging characteristic index of the coal mine coal bunker at the current moment.

[0049] In an embodiment, the calculation formula of the wall-hanging characteristic index of the coal mine coal bunker at the current moment is:

[0050]

[0051] wherein H represents the wall-hanging characteristic index of the coal mine coal bunker at the current moment, L represents the normalized value of the material level at the current moment, W represents the normalized value of the material weight at the current moment, represents a preset constant to prevent the denominator from being zero, in the embodiment of the present disclosure, the present disclosure does not limit this, and the implementer can set this according to the specific scene, V represents the normalized value of the bin wall vibration signal at the current moment, and S represents the normalized value of the bin wall material adhesion volume at the current moment, represents a linear normalization function, represents a first weight, represents a second weight.

[0052] It should be noted that when the material level is constant, if the material weight decreases, it is likely that coal wall-hanging occurs at this time, that is, when L is constant, the smaller W is, the greater the possibility of coal wall-hanging is, and thus H is greater; when the bin wall material adhesion volume is large and the bin wall vibration amplitude is small, it is likely that coal wall-hanging occurs at this time, that is, the smaller V is and the greater S is, the greater the possibility of coal wall-hanging is, and thus H is greater. Since the two coupling relationships are complementary in space and physics, the same weight is allocated, that is, ​There are no restrictions here; implementers can set them according to the specific scenario.

[0053] Thus, the analysis of the coupling relationship between material level and material weight, as well as the coupling relationship between the volume of material adhering to the silo wall and the silo wall vibration signal, was completed, and the wall adhesion characteristic index of the coal mine silo at the current moment was obtained.

[0054] Step S103: During the iterative fusion of various evidence sources using the Dempster combination rule, the wall-hanging feature index is used to optimize all quality functions in each iterative fusion to obtain all optimized quality functions in each iterative fusion. Based on all optimized quality functions in each iterative fusion, the various evidence sources are iteratively fused to obtain the fusion probability of coal hanging in the coal bunker at the current moment.

[0055] In the process of iteratively fusing various evidence sources using Dempster's combination rule, it is assumed that the material location is fused first. and material weight Taking proposition A {occurrence of wall hanging} as an example, then and The mass function of A after fusion (i.e.) and The formula for calculating the reliability of A after fusion is: ,in, Indicates fusion and The conflict coefficient of A at time A, The construction of the quality function and the calculation of the conflict coefficient are existing techniques in Dempster's combination rules, and will not be elaborated here.

[0056] Furthermore, using the coal mine bunker's wall-hanging characteristic index obtained in step S102 at the current moment, the following is performed: and After fusion, the conflict coefficient in the mass function of A is dynamically adjusted to obtain the final conflict coefficient, thereby optimizing the mass function. Specifically:

[0057] Will Conflict coefficient in The initial conflict coefficient is denoted as , and the product between the preset intensity adjustment factor and the wall-hanging characteristic index is calculated. The constant 1 is subtracted from the product to obtain the adjustment coefficient. The product between the initial conflict coefficient and the adjustment coefficient is calculated to obtain the final conflict coefficient.

[0058] In one implementation, the formula for calculating the final conflict coefficient is:

[0059]

[0060] Wherein, represents fusion and the final conflict coefficient of A, represents fusion and the initial conflict coefficient of A, H represents the wall hanging characteristic index of the coal bunker in the coal mine at the current time, represents the preset strength adjustment factor.

[0061] It should be noted that since the contradictory data will greatly increase the conflict coefficient, and then cause the fusion information obtained in the final D-S evidence theory to be distorted, the greater H is, the greater the possibility of coal wall hanging in the coal bunker at the current time, at this time, the initial conflict coefficient should be appropriately reduced, that is The smaller, the more the fusion information obtained in the final D-S evidence theory tends to the wall hanging state indicated by the wall hanging characteristic index, and reduces the negative influence of the conflict evidence; for controlling the degree of influence of H on the conflict coefficient, In the embodiment of the present application, the value of H is set to Here, no limitation is made, and the implementer can set it according to the specific scene.

[0062] Further, the value of in the formula is replaced by , to obtain the optimized quality function of A after fusion and , denoted as , that is . Similarly, the final conflict coefficient of B when and fuse is obtained , and the final conflict coefficient of C when and fuse is obtained , and then the optimized quality function of B after fusion and is obtained , and the optimized quality function of C after fusion and is obtained .

[0063] Further, the fusion of and is obtained and the optimized quality function of A when fuse is , the optimized quality function of B when fuse is , and Optimized quality function of fusion time C ; finally, the and fusions are obtained, that is, the fusion probability of each proposition after the fusion of all evidence sources, that is, the fusion probability of A , the fusion probability of B , and the fusion probability of C .

[0064] In step S104, the fusion probability of coal wall hanging in the coal mine coal bunker at a preset number of historical moments before the current moment is obtained, and the intelligent management and control of the coal wall hanging state of the coal mine coal bunker at the current moment is performed according to the current moment and the fusion probability of coal wall hanging in the coal mine coal bunker at each historical moment.

[0065] Considering that the fusion result of a single moment may be affected by instantaneous interference and produce jitter, in the embodiment of the present application, the fusion probability of coal wall hanging in the coal mine coal bunker at the current moment is smoothed and optimized by introducing time sequence context information, and the final probability of coal wall hanging in the coal mine coal bunker at the current moment is obtained, so as to improve the stability and reliability of system decision.

[0066] The specific way of obtaining the final probability of coal wall hanging in the coal mine coal bunker at the current moment is as follows:

[0067] The fusion probability of coal wall hanging in the coal mine coal bunker at 9 historical moments before the current moment is obtained, and the number of historical moments is not limited here. The implementer can set the number of historical moments according to the monitoring frequency of the coal wall hanging state in actual application, take a preset decay factor as the base, take the time interval between each historical moment and the current moment (the number of moments between the historical moment and the current moment) as the index, obtain the decay weight of the fusion probability of coal wall hanging in the coal mine coal bunker at each historical moment corresponding to each time interval, and the preset decay factor is less than a constant 1.

[0068] The constant 1 is taken as the decay weight of the fusion probability of coal wall hanging in the coal mine coal bunker at the current moment, the weighted average of all fusion probabilities is calculated according to the decay weights of all fusion probabilities, and the final probability of coal wall hanging in the coal mine coal bunker at the current moment is obtained.

[0069] In an embodiment, the calculation formula of the final probability of coal wall hanging in the coal mine coal bunker at the current moment is as follows:

[0070]

[0071] Wherein, represents the final probability of coal wall hanging in the coal mine coal bunker at the current moment, represents a preset attenuation factor, N is the total number of all historical moments and the current moment, N=10 in the embodiment of the application, represents the time interval between the i-th moment (including all historical moments and the current moment) and the current moment, since the time interval between the current moment and the current moment is 0, when the i-th moment is the current moment, , that is, , represents the fusion probability of coal wall-hanging in the coal bunker at the i-th moment.

[0072] It should be noted that, in order to make the reference value of the fusion probability at the historical moment farther away from the current moment smaller, the value of is less than the constant 1, the typical value range of is [0.8, 0.95], and in the embodiment of the application, , which is not limited here, and the implementer can set it according to the specific scene.

[0073] Further, according to the final probability of coal wall-hanging in the coal bunker at the current moment, the coal wall-hanging state of the coal bunker at the current moment is intelligently managed and controlled, specifically:

[0074] The preset wall-hanging probability threshold corresponding to different preset wall-hanging management and control strategies is obtained, the final probability of coal wall-hanging in the coal bunker at the current moment is compared with each preset wall-hanging probability threshold, and the preset wall-hanging management and control strategy of the coal bunker at the current moment is obtained. For example: when , it is considered that the coal bunker is in normal operation and no wall-hanging occurs, at this time, no management and control is needed; when , it is considered that the coal bunker has slight wall-hanging or poor flow, at this time, the wall vibrator is set to operate in low intensity and intermittent mode, or a few air cannons located in the easy wall-hanging area are triggered to operate in low frequency (such as once every 5 minutes), to prevent the wall-hanging from continuing to deteriorate and develop into serious blockage; when , it is considered that the coal bunker has wall-hanging or blockage has been formed, at this time, the working intensity and working cycle of the wall vibrator should be increased, or a group of air cannons are triggered in a specific order and time sequence (such as from bottom to top, with an interval of 2 seconds), and the frequency of the coal feeder is adjusted at the same time, to reduce the load, reduce wall-hanging, and restore flow; when At this time, it is considered that the coal bunker of the coal mine has occurred serious wall hanging, and there is a structural risk, at this time, the coal feeder should be immediately stopped running, all air cannons and bunker wall vibrators are triggered in strong sequence, forced ventilation (gas prevention) is started, inert gas (explosion and fire prevention) is injected, and operation and maintenance personnel are notified to handle the emergency site to ensure the safety of the coal bunker of the coal mine and prevent accidents. The setting of the preset wall hanging probability threshold corresponding to different preset wall hanging control strategies is not limited, and the implementer can set it according to the specific scene.

[0075] Intelligent control of the coal wall hanging state of the coal bunker of the coal mine at the current time according to the final probability of coal wall hanging in the coal bunker of the coal mine at the current time is the prior art, which will not be repeated here.

[0076] In summary, the present application obtains a wall hanging characteristic index for quantifying the possibility of wall hanging by analyzing the coupling relationship between the material level and the material weight, and the coupling relationship between the material adhesion volume of the bunker wall and the bunker wall vibration signal. In the process of iterative fusion of each evidence source using the Dempster combination rule, the wall hanging characteristic index is used to optimize all quality functions at each iteration fusion, effectively reducing the decision-making error caused by the defects of the combination rule of the traditional D-S evidence theory in the high conflict scene, obtaining all optimized quality functions with stronger anti-interference ability at each iteration fusion, and then obtaining the fusion probability of coal wall hanging in the coal bunker of the coal mine at the current time, improving the accuracy of identifying the state of coal wall hanging using the D-S evidence theory. At the same time, the fusion probability of coal wall hanging in the coal bunker of the coal mine at the current time and each historical time is obtained to combine the time sequence weighted smoothing to suppress transient interference and obtain a reliable conclusion, and finally realize accurate identification and robust early warning of the wall hanging state of the bunker.

[0077] The above embodiments are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. An intelligent management and control system for a coal mine bunker multi-parameter fusion, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, The processor implements the following steps when executing the computer program: In the process of monitoring the coal wall hanging state of the coal bunker at the current time by using the D-S evidence theory, each evidence source for monitoring the coal wall hanging state of the coal bunker is obtained, the evidence sources include the material level, the material weight, the volume of material adhesion on the wall and the wall vibration signal of the coal bunker; According to the monitoring data of each evidence source at the current time, the coupling relationship between the material level and the material weight, and the coupling relationship between the volume of material adhesion on the wall and the wall vibration signal are analyzed, and a wall hanging characteristic index of the coal bunker at the current time is obtained; In the process of iteratively fusing each evidence source by using the Dempster combination rule, the wall hanging characteristic index is used to optimize all mass functions at each iteration fusion, and all optimized mass functions at each iteration fusion are obtained, and according to all optimized mass functions at each iteration fusion, each evidence source is iteratively fused, and a fusion probability of the coal wall hanging in the coal bunker at the current time is obtained; The fusion probabilities of the coal wall hanging in the coal bunker at a preset number of historical times before the current time are obtained, and according to the fusion probability of the coal wall hanging in the coal bunker at the current time and the fusion probability of the coal wall hanging in the coal bunker at each historical time, the coal wall hanging state of the coal bunker at the current time is intelligently controlled; The wall hanging characteristic index of the coal bunker at the current time is obtained according to the monitoring data of each evidence source at the current time, including: The monitoring data of each evidence source at the current time is linearly normalized to obtain the normalized value of each evidence source at the current time; According to the normalized values of the material level and the material weight at the current time, a first coupling characteristic value between the material level and the material weight is obtained; According to the normalized values of the volume of material adhesion on the wall and the wall vibration signal at the current time, a second coupling characteristic value between the volume of material adhesion on the wall and the wall vibration signal is obtained; The first coupling characteristic value and the second coupling characteristic value are weighted and summed to obtain the wall hanging characteristic index of the coal bunker at the current time; wherein the calculation formula of the wall hanging characteristic index of the coal bunker at the current time is: ； wherein H represents a wall-hanging characteristic index of the coal bunker of the coal mine at the current time, L represents a normalized value of the material level at the current time, and W represents a normalized value of the material weight at the current time, represents a preset constant, V represents a normalized value of the bunker wall vibration signal at the current time, and S represents a normalized value of the material adhesion volume of the bunker wall at the current time, represents a linear normalization function, represents a first weight, represents a second weight; The wall hanging characteristic index is used to optimize all mass functions at each iteration fusion to obtain all optimized mass functions at each iteration fusion, including: For any mass function at any iteration fusion, the conflict coefficient in the mass function is obtained, denoted as an initial conflict coefficient, the product of a preset intensity adjustment factor and the wall hanging characteristic index is calculated, the constant 1 is subtracted from the product to obtain an adjustment coefficient, the product of the initial conflict coefficient and the adjustment coefficient is calculated to obtain a final conflict coefficient; The initial conflict coefficient in the mass function is replaced by the final conflict coefficient to obtain the optimized mass function of the mass function.

2. The intelligent management and control system of a coal mine bunker multi-parameter fusion according to claim 1, characterized in that, The first coupling characteristic value between the material level and the material weight is obtained according to the normalized value of the material level and the normalized value of the material weight at the current moment, and includes: The reciprocal of the sum between the normalized value of the material weight at the current moment and a preset constant is calculated, and the product between the reciprocal and the normalized value of the material level at the current moment is linearly normalized to obtain the first coupling characteristic value between the material level and the material weight.

3. The intelligent management and control system of a coal mine bunker multi-parameter fusion according to claim 1, characterized in that, The second coupling characteristic value between the wall material adhesion volume and the wall vibration signal is obtained according to the normalized value of the wall material adhesion volume and the normalized value of the wall vibration signal at the current moment, and includes: The difference between constant 1 and the normalized value of the wall vibration signal at the current moment is calculated, and the product between the difference and the normalized value of the wall material adhesion volume at the current moment is calculated to obtain the second coupling characteristic value between the wall material adhesion volume and the wall vibration signal.

4. The intelligent management and control system of a coal mine bunker multi-parameter fusion according to claim 1, characterized in that, The coal wall hanging state of the coal mine bunker at the current moment is intelligently controlled according to the fusion probability of coal wall hanging in the coal mine bunker at the current moment and each historical moment, and includes: The time interval between each historical moment and the current moment is obtained, a preset attenuation factor is taken as the base, and each time interval is taken as the index to obtain the attenuation weight of the fusion probability of coal wall hanging in the coal mine bunker at the corresponding historical moment of each time interval, and the preset attenuation factor is less than constant 1; Constant 1 is taken as the attenuation weight of the fusion probability of coal wall hanging in the coal mine bunker at the current moment, and the weighted average value of all fusion probabilities is calculated according to the attenuation weights of all fusion probabilities to obtain the final probability of coal wall hanging in the coal mine bunker at the current moment; The coal wall hanging state of the coal mine bunker at the current moment is intelligently controlled according to the final probability of coal wall hanging in the coal mine bunker at the current moment.

5. The intelligent management and control system of a coal mine bunker multi-parameter fusion according to claim 4, characterized in that, The coal wall hanging state of the coal mine bunker at the current moment is intelligently controlled according to the final probability of coal wall hanging in the coal mine bunker at the current moment, and includes: Different preset wall hanging probability thresholds corresponding to preset wall hanging control strategies are obtained, and the final probability of coal wall hanging in the coal mine bunker at the current moment is compared with each preset wall hanging probability threshold to obtain the preset wall hanging control strategy of the coal mine bunker at the current moment.

Images