Underground coal machine equipment working condition monitoring method based on multi-sensor fusion

Abstract

The invention relates to the technical field of coal mine safety production, and discloses an underground coal machine equipment working condition monitoring method based on multi-sensor fusion, which comprises the following steps: step 1, collecting heading machine equipment state data, roadway environment safety data and transportation load data, applying a unified time stamp to all the collected data, establishing a geographic position coordinate mark based on the real-time propelling distance of the heading machine, and completing space-time alignment of the three types of data; step 2, based on the collected alignment data, performing normalization processing and weighted combination to obtain an equipment load index, an environment pressure index and a transportation load index; and 3, dynamically dividing a front target section and a rear reference section. By adopting the technical scheme of dynamic self-contrast and relative size relation discrimination of three types of index deviations, the technical effect of clearly discriminating the abnormal working condition as a risk type is achieved, and the defect that in the prior art, abnormal causes cannot be accurately distinguished is overcome.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of coal mine safety production, in particular to an underground coal machine equipment working condition monitoring method based on multi-sensor fusion. BACKGROUND

[0002] Underground coal machine equipment, especially heading machine, is the key equipment for efficient mining of coal mine, and the working condition is directly related to production efficiency and operation safety. However, the underground environment has the characteristics of complex and changeable geological conditions, high gas and dust concentration, and large temperature and humidity fluctuation, so that the running working condition of the coal machine equipment is a complex system with high coupling, nonlinearity and strong time variation.

[0003] The traditional coal machine equipment working condition monitoring system mainly relies on a small number of sensors such as current and oil pressure sensors installed on the equipment to monitor the running state of the equipment. The system usually takes the equipment state data, roadway environment safety data and transportation load data as independent and dispersed monitoring objects, and independently collects and alarms each other. This kind of isolated monitoring method has fundamental technical defects:

[0004] The existing working condition monitoring system independently monitors the equipment, environment and transportation data, so that the on-duty personnel cannot easily perform multi-factor cross analysis when facing an abnormality, and cannot accurately distinguish whether the abnormality is caused by geological changes, environmental factors, or equipment failure or wear, thereby delaying the best opportunity to take targeted measures.

[0005] The underground roadway environment is complex and narrow, and it is difficult and easy to damage and fail to arrange sensors on the key operation surface. A small number of monitoring points are arranged away from the key operation surface, resulting in incomplete or serious lag of working condition data information in the key operation sections such as excavation and transportation, and lack of discovery and capture ability for local transient abnormalities.

[0006] The existing monitoring system lacks intelligent recognition and automatic diagnosis function for complex working conditions such as abnormal wear of cutting tool, unbalanced load of cutting head, and asynchronous operation of multi-section belt, and the discovery and judgment of abnormality mainly rely on the experience-based judgment of the on-site driver through hearing, vision and touch. The judgment method of artificial perception has low reliability and strong subjectivity, and once the judgment is wrong, it is easy to cause shutdown or safety production accidents.

[0007] The existing monitoring system can issue a simple alarm for single monitoring point data transient overrun, lacks trend tracking analysis capability for gradual change and accumulation of working condition indexes in continuous multiple sections in the advancing direction along the roadway, and is not easy to discover the implicit risk that the geological conditions or gas environment of the front section to be excavated are gradually deteriorating, which is not conducive to adjusting the excavation parameters and ventilation scheme in advance to control the risk.

[0008] To solve the above problems, the application provides a kind of underground coal machine equipment operating condition monitoring method based on multi-sensor fusion.The method realizes the accurate classification identification of operating condition risk, the risk trend tracking and early warning along the roadway mileage change by the spatio-temporal alignment of multi-source heterogeneous data, the construction of operating condition risk index and dynamic self-contrast, and outputs the classification linkage instruction combined with the risk type. SUMMARY

[0009] In view of the deficiencies of the prior art, the present application provides a kind of underground coal machine equipment operating condition monitoring method based on multi-sensor fusion to solve the problems raised in the above background.

[0010] To achieve the above purpose, the present application is realized by the following technical scheme: a kind of underground coal machine equipment operating condition monitoring method based on multi-sensor fusion, comprising:

[0011] Step one, collect the tunneling machine equipment state data, roadway environment safety data and transportation load data, apply uniform time mark to all collected data, establish geographical position coordinate mark based on the real-time advancing distance of the tunneling machine, complete the spatio-temporal alignment of three types of data;

[0012] Step two, based on the collected aligned data, normalization processing and weighted combination are carried out, and equipment load index, environmental pressure index and transportation load index are obtained;

[0013] Step three, dynamically divide the front target section and the rear reference section, calculate the average value difference of target section and reference section based on three types of indexes, obtain three types of deviation indexes, and weighted combination is carried out on three types of deviation indexes, and the operating condition risk index along the roadway mileage change is outputted;

[0014] Step four, according to the relative size relationship between three types of deviation indexes, the abnormal operating condition is judged as one of environment dominant risk, tunneling load dominant risk or transportation load dominant risk;

[0015] Step five, based on the comparison of operating condition risk index and preset classification threshold, the corresponding classification early warning is outputted, and the classification linkage instruction is issued combined with the identified risk type;

[0016] Step six, when the operating condition risk index is continuously stable and lower than the preset safety threshold, the data of the current front target section is automatically updated as the new rear reference section data.

[0017] Preferably, the tunneling machine equipment state data includes tunneling machine motor current, hydraulic oil pressure and cutting head vibration index; The roadway environment safety data includes gas concentration, dust concentration and temperature index; The transportation load data includes belt load and belt speed index.

[0018] Preferably, in step one, further comprising:

[0019] sub-step , collecting tunneling machine equipment state data in real time , roadway environment safety data and transportation load data Three types of heterogeneous data, and impose uniform timestamps on each group of sample points of each type of data ;

[0020] wherein, is the set of equipment state data collected at time , is the set of environment safety data collected at time , is the set of transportation load data collected at time , is the uniform timestamp, is the serial number of the current collection sequence;

[0021] sub-step , based on the real-time pushing distance of the tunneling machine , calculate and establish the unique geographical position coordinates for each group of data samples collected with uniform timestamps , The calculation formula of the geographical position coordinates is: ,

[0022] wherein, is the geographical position coordinates of the tunneling starting point, is the real-time pushing distance of the tunneling machine, is the geographical position coordinates;

[0023] sub-step , based on the uniform timestamp and the geographical position coordinates , align the tunneling machine equipment state data , roadway environment safety data and transportation load data Three types of heterogeneous data on the same time and the same roadway section, forming a section time two-dimensional data table , the construction of the section time two-dimensional data table satisfies:

[0024] ,

[0025] wherein, is the set of all uniform timestamps, is the set of all geographical position coordinates, is the section time two-dimensional data table, The data sample is contained in a set of all uniform timestamps. middle, The data sample contains a set of all geographic location coordinates. middle.

[0026] Preferably, step two further includes:

[0027] Sub-step The output two-dimensional data table of time intervals The raw monitoring signals in Normalization is performed to obtain the normalized signal. The normalization formula is:

[0028] ,

[0029] in, The category to which the original signal belongs. Two-dimensional data table for time intervals The sample number in the sample, Original monitoring signal The preset safety limits for the monitoring items are as follows: Two-dimensional data table for time intervals The Middle The first sample point Real-time value of the original monitoring signal. For the first The first sample point The signal value after normalization of the original monitoring signal;

[0030] Sub-step Based on device status data Related normalized signal The weighted combination calculation yields the first... Equipment load index of sample points The formula for calculating the index is as follows:

[0031] ,

[0032] in, This represents the total number of original monitoring items in the equipment status data. For the first Normalized signals for equipment condition monitoring projects For the first Equipment load indicators at sample points For the first The preset weighting coefficients for equipment condition monitoring items when calculating equipment load indicators, and all weights satisfy the limiting condition;

[0033] sub-step based on the normalized signal related to the environmental safety data and the normalized signal related to the transportation load data , the environmental stress index and the transportation load index of the i-th sample point are calculated by weighted combination, and the calculation formula is:

[0034] ,

[0035] ,

[0036] wherein, n is the total number of environmental safety monitoring items, n is the total number of transportation load monitoring items, is the environmental stress index of the i-th sample point, is the transportation load index of the i-th sample point, and is a single quantitative value reflecting the load and stability of the transportation system, , , is the preset weight coefficient of the corresponding normalized signal.

[0037] Preferably, in step three, further comprising:

[0038] sub-step dynamically dividing a front target section with a length of Lf and a rear reference section with a length of Lr, located at the current tunneling working face, located at the rear area where tunneling has been completed and the working condition is stable, and based on the output device load index, the environmental stress index and the transportation load index, the average values of the three types of indexes on the two sections are calculated. the calculation formula of the front target section average value is:

[0039]

[0040] ,

[0041] the calculation formula of the rear reference section average value is:

[0042] ​​​​​​​​​​​​​​​​ ,

[0043] wherein, is the front target section, is the rear reference section, , is the preset length of the front target section and the rear reference section, is the value of the first class index at the first sample point, , is the total number of sample points contained in the target section and the reference section, , is the average value of the first class index on the target section and the reference section;

[0044] Sub-step , based on the average value of the front target section and the average value of the rear reference section , the three types of deviation indexes are calculated by the difference between the average value of the front target section and the average value of the rear reference section :

[0045] The device load deviation index , the environmental stress deviation index and the transportation load deviation index , the calculation formula is: ,

[0046] wherein, is the deviation value of the first class index, is the average value of the first class index on the target section, is the average value of the first class index on the reference section;

[0047] Sub-step , the device load deviation index , the environmental stress deviation index and the transportation load deviation index are combined by weighting, and the working condition risk index is calculated, the calculation formula of the working condition risk index is:

[0048] ,

[0049] wherein, , , is the preset weight coefficient, is the working condition risk index.

[0050] Preferably, in step four, further comprising:

[0051] sub-step performing significance determination on the outputted equipment load deviation index , environmental stress deviation index and transportation load deviation index to determine whether there is an abnormality that needs to be identified, the significance determination condition being:

[0052] ,

[0053] wherein, is the first class deviation index, is the minimum significance threshold of the first class deviation index;

[0054] sub-step when it is determined that there is a significant abnormality, performing risk type classification determination according to the relative size relationship among the equipment load deviation index , environmental stress deviation index and transportation load deviation index :

[0055] environmental dominant risk determination condition:

[0056] and ,

[0057] if this condition is met, it is determined to be an environmental dominant risk;

[0058] excavation load dominant risk determination condition:

[0059] and ,

[0060] if this condition is met, it is determined to be an excavation load dominant risk;

[0061] transportation load dominant risk determination condition:

[0062] and ,

[0063] if this condition is met, it is determined to be a transportation load dominant risk;

[0064] wherein, , , , , , are preset relative discrimination coefficients;

[0065] sub-step , if the equipment load deviation index , the environmental stress deviation index and the transportation load deviation index satisfy the following comprehensive increase conditions, it is determined that the comprehensive risk is:

[0066] and and ,

[0067] if any index passes the significance determination but fails to be determined as a specific risk type, it is determined as a no-type risk;

[0068] wherein, is a preset comprehensive risk minimum threshold.

[0069] Preferably, in step five, further comprising:

[0070] sub-step , based on the working condition risk index , compare with the preset graded warning threshold , , to determine the warning level of the current working condition, wherein, ;

[0071] first warning determination condition: ≤ ,

[0072] if this condition is met, output first warning information, and suggest the on-duty personnel to focus on the risk profile change trend;

[0073] second warning determination condition: ≤ ,

[0074] if this condition is met, output second warning information, and trigger forced audible and light alarm;

[0075] third warning determination condition: ≥ ,

[0076] if this condition is met, output third warning information, and trigger the highest level of emergency alarm.

[0077] wherein, is the working condition risk index, , , is a preset graded warning threshold;

[0078] sub-step When determining a secondary or tertiary early warning, according to the output risk type Send a corresponding classification linkage instruction to the control unit The linkage instruction meets the following conditions:

[0079] Environmental dominant risk instruction determination:

[0080] If Environmental dominant risk Enhanced ventilation instruction;

[0081] Excavation load dominant risk instruction determination:

[0082] If Excavation load dominant risk Slow down excavation instruction;

[0083] Transport load dominant risk instruction determination:

[0084] If Transport load dominant risk Check and clean instruction;

[0085] Tertiary early warning instruction forced determination:

[0086] If ≥ Automatic stop instruction;

[0087] Wherein, is the risk type, is the specific linkage instruction, is the working condition risk index, is the tertiary early warning threshold;

[0088] sub-step Real-time display of the constructed working condition risk profile along the tunnel mileage and the determined risk type and the current early warning level, to provide intuitive visual judgment basis for on-site duty personnel.

[0089] Preferably, in step six, further comprising:

[0090] sub-step Continuously monitor the output working condition risk index When the working condition risk index meets the following stable working condition determination conditions within the preset continuous monitoring time , enter the sub-step ;

[0091] Risk value below safety threshold condition:

[0092] For all

[0093] wherein, is a preset safety working condition threshold value;

[0094] The risk fluctuation degree is lower than the stability threshold condition: ,

[0095] wherein, is a preset risk fluctuation stability threshold value, is the working condition risk index at time , is a preset continuous monitoring time length, is the fluctuation degree index of the working condition risk index in the monitoring time;

[0096] Substep When the stable working condition determination condition is met, the automatic updating of the reference section data is performed: all data of the current front target section are automatically assigned to the rear reference section , and the rolling of the section data is completed;

[0097] Data updating logic: ,

[0098] wherein, is the data set of the current front target section, is the updated rear reference section data set;

[0099] Substep After the automatic updating of the reference section data is completed, the reference section average values of the three types of indexes are recalculated based on the data in the new rear reference section , and the recalculation formula is: ,

[0100] wherein, is the updated average value of the th type of index on the rear reference section, is the total number of sample points contained in the new rear reference section, is the value of the th type of index at the th sample point.

[0101] Preferably, the equipment load index , the environmental stress index and the transportation load index ​The weight coefficient is determined based on an analytic hierarchy process or an expert experience method, and the weight coefficient is dynamically adjusted according to the coal seam geological condition and the equipment aging degree.

[0102] Preferably, the environment dominant risk type in the fourth step specifically refers to a significant increase in a deviation value of a gas concentration or a dust concentration index in a roadway, the tunneling load dominant risk type specifically refers to a significant increase in a deviation value of a motor current or a cutting head vibration index of a tunneling machine, and the transportation load dominant risk type specifically refers to a significant increase in a deviation value of a belt load index.

[0103] The application provides an underground coal machine equipment working condition monitoring method based on multi-sensor fusion.

[0104] 1. The application adopts a dynamic self-contrast and relative size relationship discrimination technical scheme of three types of index deviations, achieves the technical effect of clearly distinguishing abnormal working conditions as risk types, and solves the problem of the inability to accurately distinguish abnormal causes in the prior art.

[0105] 2. The application adopts a unified timestamp and advancing distance coordinate marking technical scheme of multi-source heterogeneous data, achieves the technical effect of aligning three types of data in space and time, and solves the problem of incomplete working condition data information in key operation sections in the prior art.

[0106] 3. The application adopts a classified linkage instruction issuing technical scheme according to risk types, achieves the technical effect of converting risk analysis results into automatic instructions, and solves the problem of low reliability of artificial perception and judgment in the prior art.

[0107] 4. The application adopts a technical scheme of dynamically dividing sections and outputting working condition risk indexes changing along the roadway mileage, achieves the technical effect of realizing working condition index trend tracking analysis, and solves the problem of the difficulty in discovering implicit risks in advance in the prior art. BRIEF DESCRIPTION OF DRAWINGS

[0108] Figure 1 is a flowchart of the application;

[0109] Figure 2 is a schematic diagram of the relative size relationship of three types of deviation indexes and risk type discrimination. DETAILED DESCRIPTION

[0110] To enable personnel skilled in the art to understand the application scheme, the technical solutions in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, other embodiments obtained by those skilled in the art without creative labor should be within the protection scope of the application.

[0111] The application will be described in detail below with reference to the drawings:

[0112] Embodiments:

[0113] Please refer to the accompanying Figure 1 -Appendix Figure 2 , the embodiment of the application provides a kind of based on multi-sensor fusion's underground coal machine equipment operating condition monitoring method, comprising:

[0114] Step one, collection heading machine equipment state data, roadway environment safety data and transportation load data, all data collected are subjected to uniform time mark, based on the real-time advancing distance of heading machine, establish geographical position coordinate mark, complete the space-time alignment of three types of data;

[0115] Step two, based on the aligned data collected, normalization processing and weighted combination are carried out, to obtain equipment load index, environmental stress index and transportation load index;

[0116] Step three, dynamically divide front target section and rear reference section, based on three types of indexes, the average value difference of target section and reference section is calculated, three types of deviation indexes are obtained, and three types of deviation indexes are weighted combined, and the operating condition risk index along the roadway mileage change is output;

[0117] Step four, according to the relative size relationship between three types of deviation indexes, the abnormal operating condition is identified as one of environment dominant risk, heading load dominant risk or transportation load dominant risk;

[0118] Step five, based on the comparison between operating condition risk index and preset classification threshold, the corresponding classification early warning is output, and the classified linkage instruction is issued in combination with the identified risk type;

[0119] Step six, when operating condition risk index is continuously stable and lower than preset safety threshold, the data of current front target section is automatically updated as new rear reference section data.

[0120] In step one, further comprising:

[0121] Sub-step , real-time collection heading machine equipment state data , roadway environment safety data And transportation load data Three types of heterogeneous data, and each group of sample points of each type of data is subjected to uniform timestamp ;

[0122] Among them, It is equipment state data set collected at time , It is equipment state data set collected at time The collection of environmental safety data, In time The collection of transportation load data, To standardize timestamps, This is the sequence number of the currently collected sequence;

[0123] Sub-step Based on the real-time advance distance obtained by the tunneling machine To ensure that each group collected has a unified timestamp Calculate and establish unique geographic coordinates for the data samples Geographic coordinates The calculation formula is: ,

[0124] in, The geographical coordinates of the tunneling starting point. This refers to the real-time advance distance of the tunneling machine. Geographic coordinates;

[0125] Sub-step Based on unified timestamp and geographic coordinates The tunneling machine equipment status data Safety data of tunnel environment and transport load data The three types of heterogeneous data are aligned at the same time and in the same roadway segment to form a two-dimensional data table for the segment time. Two-dimensional data table of time intervals The construction satisfies:

[0126] ,

[0127] in, For the set of all uniform timestamps, It is the set of all geographic location coordinates. A two-dimensional data table for a time interval. The data sample is contained in a set of all uniform timestamps. middle, The data sample contains a set of all geographic location coordinates. middle.

[0128] Step two further includes:

[0129] Sub-step The output two-dimensional data table of time intervals The raw monitoring signals in Normalization is performed to obtain the normalized signal. The normalization formula is:

[0130] ,

[0131] wherein, is the category to which the original signal belongs, is the sample serial number in the section-time two-dimensional data table , is the preset safety limit value of the monitoring item to which the original monitoring signal belongs, is the real-time value of the original monitoring signal of the i-th category at the j-th sample point in the section-time two-dimensional data table ,

[0132] Sub-step , based on the normalized signals related to the equipment state data , the equipment load index at the j-th sample point is calculated by weighted combination , and the index calculation formula is:

[0133] ,

[0134] wherein, is the total number of original monitoring items in the equipment state data, is the normalized signal of the i-th equipment state monitoring item, is the equipment load index at the j-th sample point, is the preset weight coefficient of the i-th equipment state monitoring item in the calculation of the equipment load index, and all weights satisfy the limit condition

[0135] Sub-step , based on the normalized signals related to the environmental safety data and the normalized signals related to the transportation load data , the environmental pressure index at the j-th sample point and the transportation load index are calculated by weighted combination , and the calculation formula is:

[0136] , ​​​​​​​​​​​​​​​​

[0137] ,

[0138] wherein, is the total number of environmental safety monitoring items, is the total number of transportation load monitoring items, is the environmental pressure index of the sample point, is the transportation load index of the sample point, which is a single quantitative value reflecting the load and stability of the transportation system, , is the preset weight coefficient of the corresponding normalized signal.

[0139] In step three, further comprising:

[0140] Sub-step dynamically divides a front target section with a length of and a rear reference section with a length of , is located at the current tunneling working face, is located at the rear area where tunneling has been completed and the working condition is stable, and based on the output device load index , the environmental pressure index and the transportation load index , the average values of the three types of indexes on the two sections are calculated;

[0141] The calculation formula of the front target section average value :

[0142] ,

[0143] The calculation formula of the rear reference section average value :

[0144] ,

[0145] wherein, is the front target section, is the rear reference section, , is the preset length of the front target section and the rear reference section, is the value of the type index at the sample point, , is the total number of sample points contained in the target section and the reference section, , is the The average value of the class index on the target section and the reference section;

[0146] Sub-step , based on the average value of the front target section and the average value of the rear reference section , a three-class deviation index is calculated by the difference between the average value of the front target section and the average value of the rear reference section

[0147] Device load deviation index , environmental stress deviation index and transportation load deviation index , the calculation formula is: ,

[0148] Among them, is the deviation value of the first class index, is the average value of the first class index on the target section, is the average value of the first class index on the reference section;

[0149] Sub-step , the device load deviation index , the environmental stress deviation index and the transportation load deviation index are combined by weight, and the working condition risk index is calculated The calculation formula of the working condition risk index

[0150] ,

[0151] Among them, , , is the preset weight coefficient, is the working condition risk index.

[0152] In step four, further comprising:

[0153] Sub-step , the output device load deviation index , environmental stress deviation index and transportation load deviation index are subjected to significance determination to determine whether there is an abnormality that needs to be identified, and the significance determination condition is:

[0154] ,

[0155] Among them, is the first​ class bias indicator, is the first minimum significance threshold of the class bias indicator;

[0156] sub-step When it is determined that there is a significant anomaly, according to the relative size relationship among the equipment load bias indicator , the environmental pressure bias indicator and the transportation load bias indicator , a risk type classification is determined:

[0157] environmental dominant risk determination condition:

[0158] and ,

[0159] If this condition is met, it is determined that the environmental dominant risk is dominant;

[0160] tunneling load dominant risk determination condition:

[0161] and ,

[0162] If this condition is met, it is determined that the tunneling load dominant risk is dominant;

[0163] transportation load dominant risk determination condition:

[0164] and ,

[0165] If this condition is met, it is determined that the transportation load dominant risk is dominant;

[0166] wherein , , , , , are all preset relative discrimination coefficients;

[0167] sub-step If the equipment load bias indicator , the environmental pressure bias indicator and the transportation load bias indicator all meet the following comprehensive increase condition, it is determined that it is a comprehensive risk:

[0168] and and ,

[0169] If any indicator passes the significance determination but fails to be determined as a specific risk type, it is determined as a no-type risk;

[0170] wherein, is a preset minimum threshold of the comprehensive risk.

[0171] Step five further comprises:

[0172] Sub-step , comparing the working condition risk index with a preset hierarchical early warning threshold , , to determine the early warning level of the current working condition, wherein, ;

[0173] Primary early warning determination condition: ≤ ,

[0174] If this condition is met, output a primary early warning information, and suggest the on-duty personnel to focus on the risk profile change trend;

[0175] Secondary early warning determination condition: ≤ ,

[0176] If this condition is met, output a secondary early warning information, and trigger a forced sound and light alarm;

[0177] Tertiary early warning determination condition: ≥ ,

[0178] If this condition is met, output a tertiary early warning information, and trigger the highest level of emergency alarm.

[0179] wherein, is the working condition risk index, , , is a preset hierarchical early warning threshold;

[0180] Sub-step , when determining a secondary or tertiary early warning, according to the output risk type , send a corresponding classification linkage instruction to the control unit, and the linkage instruction meets the following conditions:

[0181] Environmental dominant risk instruction determination:

[0182] If the environmental dominant risk is a strengthened ventilation instruction;

[0183] Tunneling load dominant risk instruction determination:

[0184] If Tunneling load dominates risk Deceleration tunneling command;

[0185] Determination of transport load-driven risk instructions:

[0186] like Transportation load dominates risk Check the cleanup instructions;

[0187] Level 3 Early Warning Instruction Mandatory Judgment:

[0188] like ≥ Automatic shutdown command;

[0189] in, As a risk type, For specific linkage instructions, This is the working condition risk index. The threshold is set at Level 3.

[0190] Sub-step It displays in real time the working condition risk profile along the roadway mileage and the identified risk types. The current warning level and other information provide on-site staff with an intuitive and visual basis for judgment.

[0191] Step six further includes:

[0192] Sub-step The output working condition risk index Continuous monitoring will be conducted when the operating condition risk index... During the preset continuous monitoring time When the following stable operating condition determination conditions are met, proceed to the next sub-step. ;

[0193] Risk value below safety threshold condition:

[0194] For all

[0195] in, The preset safety operating condition threshold;

[0196] Risk volatility below the stability threshold condition: ,

[0197] in, The preset risk fluctuation stability threshold, In time The working condition risk index The preset continuous monitoring time length, Operating condition risk index The fluctuation index in the monitoring time;

[0198] Sub-step When the stable working condition determination condition is met, the automatic updating of the reference section data is performed: all the data of the current front target section are automatically assigned to the rear reference section , and the rolling of the section data is completed;

[0199] Data updating logic: ,

[0200] Wherein, is the data set of the current front target section, is the updated rear reference section data set;

[0201] Sub-step After the automatic updating of the reference section data is completed, the reference section average values of the three types of indexes are recalculated based on the data in the new rear reference section The recalculation formula is: ,

[0202] Wherein, is the updated average value of the type index on the rear reference section, is the total number of sample points contained in the new rear reference section, is the value of the type index at the sample point.

[0203] By collecting three types of heterogeneous data of the tunneling machine equipment state data, the roadway environment safety data and the transportation load data in real time, and applying a unified time mark and a geographical position coordinate mark based on the tunneling advancing distance, the originally dispersed and independent monitoring information is aligned on the same time and the same roadway section to form a section time two-dimensional data table, thereby solving the problem of dispersed data sources and isolated information in the prior art.

[0204] The aligned original monitoring signals are normalized to eliminate the differences between different dimensional sensor data, and the normalized signals are combined by weighting, so that the complex multi-dimensional monitoring data are condensed into the equipment load index, the environment pressure index and the transportation load index, which greatly simplifies the analysis dimension of the data and can comprehensively and comprehensively reflect the real-time working conditions of the coal machine equipment, the roadway environment and the transportation system.

[0205] ​The front target section representing the current working condition and the rear reference section representing the stable historical working condition are dynamically divided, the three types of deviation indexes are obtained by calculating the difference between the average values of the two, the dynamic self-contrast of the working condition is realized, the influence of inherent environmental factors such as geological conditions on the monitoring result is effectively eliminated, the deviation index directly reflects the abnormal change degree of the working condition, the working condition risk index changing along the roadway mileage is calculated by weighted combination, and the risk profile formed by the index directly reveals the accumulation and evolution trend of the risk.

[0206] The existence of a real anomaly is determined through significance determination, false positives are avoided, and classification discrimination is performed according to the relative size relationship among the three types of deviation indexes, and the abnormal working condition is intelligently identified as being caused by environmental dominant risk, excavation load dominant risk or transportation load dominant risk, and if multiple risks simultaneously significantly increase, it is determined as comprehensive risk, solving the problems that the on-duty personnel cannot perform multi-factor cross analysis and cannot distinguish the causes of the anomaly in the prior art.

[0207] The output working condition risk index is compared with the preset multi-level threshold value, the output of one-level, two-level and three-level graded early warning is realized, and meanwhile, combined with the risk type, targeted classification linkage instructions such as the instruction of strengthening ventilation, the instruction of slowing down excavation, the instruction of checking and cleaning or the instruction of automatic shutdown are directly issued, the risk analysis result is converted into automatic control measures, the response speed and disposal efficiency of the risk are greatly improved, the subjectivity and hysteresis of artificial experience judgment are effectively avoided, and the safety margin of production operation is improved.

[0208] The working condition risk index is continuously monitored, when the stable working condition determination conditions of being lower than the safety threshold value and low fluctuation degree are met, the current front target section data is automatically updated as new rear reference section data, the adaptive rolling update of the reference benchmark is realized, the mechanism ensures that the reference section as the dynamic self-contrast benchmark always represents the latest and most stable working condition under the current coal seam geological condition, avoids the benchmark drift caused by the slow change of the geological condition, ensures that the calculation of the three types of deviation indexes is always based on the latest actual working condition, and improves the environmental adaptability and long-term accuracy of the monitoring method.

[0209] Although the embodiments of the present application have been shown and described, it can be understood by those of ordinary skill in the art that various changes, modifications, replacements and variations can be made to the embodiments without departing from the principles and spirits of the present application, the scope of the present application is defined by the appended claims and equivalents.

Claims

1. A method for monitoring the operating conditions of underground coal mining equipment based on multi-sensor fusion, characterized in that, include: Step 1: Collect equipment status data of the tunneling machine, safety data of the tunnel environment, and transportation load data. Apply a unified time stamp to all collected data and establish geographical coordinate marks based on the real-time advance distance of the tunneling machine to complete the spatiotemporal alignment of the three types of data. Step 2: Based on the collected aligned data, normalization and weighted combination are performed to obtain equipment load index, environmental pressure index and transportation load index. Step 3: Dynamically divide the target section ahead and the reference section behind. Calculate the average difference between the target section and the reference section based on three types of indicators to obtain three types of deviation indicators. Then, perform a weighted combination of the three types of deviation indicators to output the working condition risk index that changes along the roadway mileage. Step 4: Based on the relative magnitude of the three types of deviation indicators, the abnormal working condition is identified as one of the following: environmental-dominated risk, tunneling load-dominated risk, or transportation load-dominated risk. Step 5: Based on the comparison between the working condition risk index and the preset classification threshold, output the corresponding classification warning, and at the same time, issue classified linkage instructions in combination with the identified risk type. Step 6: When the working condition risk index remains stable and is below the preset safety threshold, the data of the current target section ahead will be automatically updated to the new reference section data behind.

2. The method for monitoring the operating conditions of underground coal mining equipment based on multi-sensor fusion according to claim 1, characterized in that, The tunneling machine equipment status data includes the tunneling machine motor current, hydraulic oil pressure, and cutting head vibration index; the roadway environmental safety data includes gas concentration, dust concentration, and temperature index; and the transportation load data includes belt load and belt speed index.

3. The method for monitoring the operating conditions of underground coal mining equipment based on multi-sensor fusion according to claim 1, characterized in that, Step one further includes: Sub-step Real-time acquisition of tunneling machine equipment status data Safety data of tunnel environment and transport load data Three types of heterogeneous data, with a unified timestamp applied to each group of sample points in each type of data. ; in, In time The collection of device status data, In time The collection of environmental safety data, In time The collection of transportation load data, To standardize timestamps, This is the sequence number of the currently collected sequence; Sub-step Based on the real-time advance distance obtained by the tunneling machine To ensure that each group collected has a unified timestamp Calculate and establish unique geographic coordinates for the data samples. The geographical coordinates The calculation formula is: , in, The geographical coordinates of the tunneling starting point. This refers to the real-time advance distance of the tunneling machine. Geographic coordinates; Sub-step Based on unified timestamp and geographic coordinates The tunneling machine equipment status data Safety data of tunnel environment and transport load data The three types of heterogeneous data are aligned at the same time and in the same roadway segment to form a two-dimensional data table for the segment time. The two-dimensional data table of the time segment The construction satisfies: ， in, For the set of all uniform timestamps, The set of all geographic location coordinates. A two-dimensional data table for a time interval. The data sample is contained in a set of all uniform timestamps. middle, The data sample contains a set of all geographic location coordinates. middle.

4. The method for monitoring the operating conditions of underground coal mining equipment based on multi-sensor fusion according to claim 1, characterized in that, Step two further includes: Sub-step The output two-dimensional data table of time intervals Various raw monitoring signals Normalization is performed to obtain the normalized signal. The normalization formula is: ， in, The category to which the original signal belongs. Two-dimensional data table for time intervals The sample number in the sample, Original monitoring signal The preset safety limits for the monitoring items are as follows: Two-dimensional data table for time intervals The Middle The first sample point Real-time value of the original monitoring signal. For the first The first sample point The signal value after normalization of the original monitoring signal; Sub-step Based on device status data Related normalized signal The weighted combination calculation yields the first... Equipment load index of sample points The formula for calculating the index is as follows: ， in, This represents the total number of original monitoring items in the equipment status data. For the first Normalized signals for equipment condition monitoring projects For the first Equipment load indicators at sample points For the first The preset weighting coefficients for equipment condition monitoring items when calculating equipment load indicators, and all weights satisfy The limiting conditions; Sub-step Based on environmental safety data Related normalized signal and transport load data Related normalized signal The weighted combination calculation yields the first... Environmental pressure indicators of sample points and transportation load indicators The calculation formula is: ， ， in, This represents the total number of environmental safety monitoring projects. This represents the total number of transportation load monitoring items. For the first Environmental stress indicators of sample points For the first The transportation load index at sample points is a single quantitative value reflecting the load and stability of the transportation system. , These are the preset weighting coefficients for the corresponding normalized signal.

5. The method for monitoring the operating conditions of underground coal mining equipment based on multi-sensor fusion according to claim 1, characterized in that, Step three further includes: Sub-step The dynamic partition length is The target section ahead and length is Rear reference section , Located at the current tunneling face, Located in the rear area where tunneling has been completed and the operating conditions are stable, and based on the output equipment load index Environmental stress indicators and transportation load indicators Calculate the average value of the three types of indicators across the two segments; Average value of the target section ahead The calculation formula is as follows: ， Average value of the rear reference section The calculation formula is as follows: ， in, The target section ahead, For reference section to follow, , The preset lengths for the target section ahead and the reference section behind. For the first Class indicators in the first The values ​​of the sample points, , This represents the total number of sample points contained within the target segment and the reference segment. , For the first The average value of the category indicator in the target segment and the reference segment; Sub-step Based on the average value of the target section ahead and the average value of the reference section behind. Average value of the target segment ahead and the average value of the reference section behind. The difference is used to calculate three types of deviation indices: Equipment load deviation index Environmental pressure deviation index and transport load deviation index The calculation formula is: , in, For the first Deviation value of class indicators, For the first The average value of the class of indicators in the target segment. For the first The average value of the category indicator over the reference range; Sub-step For equipment load deviation index Environmental pressure deviation index and transport load deviation index By performing weighted combination, the working condition risk index is calculated. The working condition risk index The calculation formula is: ， in, , , The preset weighting coefficients, This is the working condition risk index.

6. The method for monitoring the operating conditions of underground coal mining equipment based on multi-sensor fusion according to claim 1, characterized in that, Step four further includes: Sub-step The output equipment load deviation index Environmental pressure deviation index and transport load deviation index A saliency determination is performed to determine whether there are any anomalies that need to be identified. The saliency determination criteria are as follows: ， in, For the first Class deviation index, For the first The minimum significance threshold for class bias indicators; Sub-step When a significant anomaly is determined, it is based on the equipment load deviation index. Environmental pressure deviation index and transport load deviation index Risk type classification is based on the relative size relationship between them: Environmental-dominant risk identification criteria: and , If this condition is met, it is determined to be an environmentally dominant risk; Tunneling load-dominant risk assessment criteria: and , If this condition is met, the risk is determined to be dominated by tunneling load. Transportation load-dominant risk assessment criteria: and , If this condition is met, it is determined to be a risk dominated by transportation load; in, , , , , , All are preset relative discrimination coefficients; Sub-step If the equipment load deviation index Environmental pressure deviation index and transport load deviation index If the following combined conditions are met simultaneously, it is determined to be a comprehensive risk: and and , If any indicator passes the significance test but fails to be classified as a specific risk type, it is classified as a typeless risk. in, This is the preset minimum threshold for comprehensive risk.

7. The method for monitoring the operating conditions of underground coal mining equipment based on multi-sensor fusion according to claim 1, characterized in that, Step five further includes: Sub-step Based on the working condition risk index With respect to the preset graded early warning threshold , , By comparing the data, the warning level of the current operating condition can be determined. ; Level 1 warning criteria: ≤ , If this condition is met, a Level 1 warning will be issued, and on-duty personnel are advised to pay close attention to the changing trends of the risk profile. Level II warning criteria: ≤ , If this condition is met, output a level 2 warning message and trigger a mandatory audible and visual alarm. Criteria for Level 3 Early Warning: ≥ , If this condition is met, a Level 3 warning message will be output, triggering the highest level emergency alarm. in, This is the working condition risk index. , , The preset tiered early warning threshold; Sub-step When a level 2 or 3 warning is issued, the risk type output will be determined accordingly. Send the corresponding classification linkage command to the control unit. The linkage command is determined if it meets the following conditions: Environmental-led risk instruction determination: like Environment-driven risk Strengthen ventilation instructions; Determination of risk command dominated by tunneling load: like Tunneling load dominates risk Deceleration tunneling command; Determination of transport load-driven risk instructions: like Transportation load dominates risk Check the cleanup instructions; Level 3 Early Warning Instruction Mandatory Judgment: like ≥ Automatic shutdown command; in, As a risk type, For specific linkage instructions, This is the working condition risk index. The threshold is set at Level 3. Sub-step It displays in real time the working condition risk profile along the roadway mileage and the identified risk types. The current warning level and other information provide on-site staff with an intuitive and visual basis for judgment.

8. The method for monitoring the operating conditions of underground coal mining equipment based on multi-sensor fusion according to claim 1, characterized in that, Step six further includes: Sub-step The output working condition risk index Continuous monitoring will be conducted when the operating condition risk index... During the preset continuous monitoring time When the following stable operating condition determination conditions are met, proceed to the next sub-step. ; Risk value below safety threshold condition: For all , in, The preset safety operating condition threshold; Risk volatility below the stability threshold condition: , in, The preset risk fluctuation stability threshold, In time The working condition risk index The preset continuous monitoring time length, Operating condition risk index The volatility index during the monitoring period; Sub-step When the stable operating condition criteria are met, the reference segment data is automatically updated: the current target segment ahead is updated. All data is automatically assigned to the subsequent reference segment. This completes the rolling of segment data; Data update logic: , in, For the current target segment, For the updated rear reference segment dataset; Sub-step After the automatic update of the reference segment data is completed, based on the new rear reference segment... Based on the data, recalculate the average reference range for the three types of indicators. The recalculation formula is as follows: , in, For the updated number The average value of the class of indicators over the reference range. This represents the total number of sample points included in the new rear reference segment. For the first Class indicators in the first The value of the sample point.

9. A method for monitoring the operating conditions of underground coal mining equipment based on multi-sensor fusion according to claim 5, characterized in that, The equipment load index Environmental stress indicators and transportation load indicators The weighting coefficients are determined based on the analytic hierarchy process or expert experience method, and are dynamically adjusted according to the geological conditions of the coal seam and the degree of equipment aging.

10. The method for monitoring the operating conditions of underground coal mining equipment based on multi-sensor fusion according to claim 1, characterized in that, The environmentally dominant risk type in step four specifically refers to a significant increase in the deviation value of the gas concentration or dust concentration index in the roadway; the tunneling load dominant risk type specifically refers to a significant increase in the deviation value of the tunneling machine motor current or cutting head vibration index; and the transportation load dominant risk type specifically refers to a significant increase in the deviation value of the belt load index.

Images