Method and system for collecting mine wind speed and direction based on ultrasonic pulse reflection technology

Abstract

The present invention belongs to the field of mining safety monitoring and intelligent sensing technology, and specifically provides a method and system for collecting wind speed and direction in mines based on ultrasonic pulse reflection technology. The method mainly includes: obtaining basic data for ultrasonic pulse reflection monitoring, dividing the basic data into a pulse matching feature set and a sensor working timing set; generating a beam path correlation matrix based on the pulse matching feature set; generating an echo optimization parameter set based on the beam path correlation matrix; generating a pulse frequency adjustment instruction set based on the echo optimization parameter set; generating a wind speed prediction period and abnormal data marking results based on the sensor working timing set and the pulse frequency adjustment instruction set. The present application can effectively eliminate monitoring blind spots in complex tunnels, improve the efficiency of equipment resource utilization, optimize the stability and timeliness of wind speed trend prediction, and provide accurate and reliable technical support for mine ventilation safety management.

Description

Technical Field

[0001] The present invention belongs to the field of mining safety monitoring and intelligent sensing technology, and in particular relates to a method and system for collecting wind speed and direction in mining based on ultrasonic pulse reflection technology. Background Art

[0002] Currently, the field of mine wind speed and direction monitoring generally adopts an ultrasonic reflection monitoring method based on fixed threshold judgment. By presetting pulse parameters to match conventional tunnel working conditions, the tunnel structure parameters and sensor data are usually processed independently, and the pulse emission characteristics are stored in the form of static templates. The equipment operating status is recorded through periodic calibration or discrete sampling.

[0003] The above scheme can meet basic monitoring needs in horizontal tunnels with simple structures. However, when faced with dynamic working conditions such as complex branch tunnels, variable ventilation disturbances and sensor array collaborative failure, it is difficult to quantify the impact of tunnel topology on the ultrasonic propagation path, resulting in continuous deterioration of the signal-to-noise ratio of the reflected signal and expansion of the monitoring blind area, which seriously threatens mine ventilation safety and the timeliness of disaster warning. Summary of the Invention

[0004] This application provides a method and system for collecting wind speed and direction in mines based on ultrasonic pulse reflection technology, which effectively solves the problem that the existing technology relies on manual estimation of the matching relationship between equipment load and process time, makes it difficult to quantify the impact of tunnel topology on the ultrasonic propagation path, resulting in continuous deterioration of the signal-to-noise ratio of the reflected signal, expansion of the monitoring blind area, and serious threats to mine ventilation safety and disaster warning timeliness. It can effectively eliminate the monitoring blind areas of complex tunnels, improve the efficiency of equipment resource utilization, optimize the stability and timeliness of wind speed trend prediction, and provide accurate and reliable technical support for mine ventilation safety management.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] In the first aspect, a method for collecting wind speed and direction in mines based on ultrasonic pulse reflection technology includes: obtaining basic data of ultrasonic pulse reflection monitoring, dividing the basic data into a pulse matching feature set and a sensor working timing set; generating a beam path correlation matrix according to the pulse matching feature set; generating an echo optimization parameter set according to the beam path correlation matrix; generating a pulse frequency adjustment instruction set according to the echo optimization parameter set; generating a wind speed prediction period and abnormal data marking results according to the sensor working timing set and the pulse frequency adjustment instruction set.

[0007] Furthermore, the basic data of the ultrasonic pulse reflection monitoring includes monitoring task data, reflection waveform template data, sensor status data and historical calibration data.

[0008] Furthermore, the basic data is divided into a pulse matching feature set and a sensor working timing set, including:

[0009] Based on the monitoring task data and the reflected waveform template data, the deviation value between the real-time wind speed value and the template and the tunnel structure complexity classification data are generated. Based on the sensor status data and the historical calibration data, the sensor effective working rate timing curve and the signal distortion statistical sequence are generated. The deviation value between the real-time wind speed value and the template and the tunnel structure complexity classification data are used as the pulse matching feature set, and the sensor effective working rate timing curve and the signal distortion statistical sequence are used as the sensor working timing set.

[0010] Furthermore, a beam path correlation matrix is ​​generated according to the pulse matching feature set, including: calculating the matching correction coefficient of the process template based on the deviation value between the real-time wind speed value and the template and the tunnel structure complexity classification data; re-prioritizing the process execution order according to the matching correction coefficient to generate an acquisition node priority sequence; constructing a dynamic correlation matrix between the acquisition node and the sensor array based on the acquisition node priority sequence and sensor status data; performing integrity verification on the dynamic correlation matrix, converting the verified dynamic correlation matrix into a numerical matrix structure, and generating a beam path correlation matrix.

[0011] Furthermore, an echo optimization parameter set is generated according to the beam path correlation matrix, including: using an improved decision tree to process the correlation matrix, the improved decision tree uses a decision tree as a basic architecture, adds a dynamic merging mechanism of branch paths, and outputs an echo optimization parameter set.

[0012] Furthermore, an improved decision tree is used to process the association matrix. The improved decision tree is based on the decision tree architecture, adds a dynamic merging mechanism for branch paths, and outputs an echo optimization parameter set, including:

[0013] The row and column structure of the beam path correlation matrix is ​​analyzed to extract the acquisition node information in the matrix row vectors and the sensor array information in the column vectors. The first-level decision branches of the improved decision tree are divided based on the lane type in the acquisition node information. Under the first-level decision branches, the second-level decision branches of the improved decision tree are divided according to the monitoring emergency level in the acquisition node information. Under the second-level decision branches, the third-level decision branches of the improved decision tree are divided based on the sensor workload in the sensor array information. At the end of the third-level decision branch, the discrete decision paths are dynamically merged to generate a continuous parameter space. The pulse emission interval and sensor collaboration weight parameters are extracted from the continuous parameter space to generate an echo optimization parameter set.

[0014] Furthermore, a pulse frequency adjustment instruction set is generated based on the echo optimization parameter set, including: extracting the pulse emission interval and sensor collaborative weight parameters in the echo optimization parameter set; calculating the effective monitoring time window based on the sensor effective working rate timing curve in the sensor working timing set; performing timing matching on the pulse emission interval parameters and the effective monitoring time window to generate a basic speed regulation instruction; combining the basic speed regulation instruction and the sensor collaborative weight parameters to generate a multi-probe collaborative acquisition instruction, and encoding it into a pulse frequency adjustment instruction set according to the sensor array type.

[0015] Furthermore, according to the sensor working timing set and the pulse frequency adjustment instruction set, a wind speed prediction period and abnormal data marking results are generated, including: using an improved differential autoregressive moving average model to process the sensor working timing set and the pulse frequency adjustment instruction set, the improved differential autoregressive moving average model uses the differential autoregressive moving average model as the basic architecture, adds a parallel parsing module for multi-source instructions, and outputs the wind speed prediction period and abnormal data marking results.

[0016] Furthermore, an improved differential autoregressive moving average model is used to process the sensor working timing set and the pulse frequency adjustment instruction set. The improved differential autoregressive moving average model uses the differential autoregressive moving average model as the basic architecture, adds a parallel parsing module for multi-source instructions, and outputs the wind speed prediction period and abnormal data marking results, including:

[0017] The differential autoregressive moving average model is used to analyze the sensor effective working rate timing curve and signal distortion statistical sequence in the sensor working timing set to extract the real-time operation efficiency characteristics of the equipment; the parallel analysis module is used to analyze the basic speed regulation instructions and multi-probe collaborative acquisition instructions in the pulse frequency adjustment instruction set to extract the instruction timing distribution characteristics; the real-time operation efficiency characteristics of the equipment and the instruction timing distribution characteristics are superimposed and analyzed to generate the sensor array efficiency index; based on the sensor array efficiency index and the pulse emission interval parameters, the completion time of each acquisition node is predicted to generate the wind speed prediction period; according to the timing overlap between the signal distortion statistical sequence and the multi-probe collaborative acquisition instructions, abnormal acquisition nodes are detected and abnormal data marking results are generated.

[0018] In a second aspect, the present application provides a mining wind speed and direction acquisition system based on ultrasonic pulse reflection technology, comprising:

[0019] Data division module: obtains basic data of ultrasonic pulse reflection monitoring, and divides the basic data into a pulse matching feature set and a sensor working timing set.

[0020] Matrix generation module: generates a beam path correlation matrix according to the pulse matching feature set.

[0021] Parameter set generation module: generates an echo optimization parameter set according to the beam path correlation matrix.

[0022] Instruction set generation module: generates a pulse frequency adjustment instruction set according to the echo optimization parameter set.

[0023] Result generation module: generates wind speed prediction cycle and abnormal data marking results according to the sensor working timing set and the pulse frequency adjustment instruction set.

[0024] In the third aspect, the present application provides a mining wind speed and direction collection device based on ultrasonic pulse reflection technology, which includes a memory and a processor; the memory is used to store computer programs; the processor is used to implement the steps of the mining wind speed and direction collection method based on ultrasonic pulse reflection technology as described in the first aspect when executing the computer program.

[0025] In a fourth aspect, the present application provides a storage medium storing computer program instructions. When the computer program instructions are read and executed by a processor, the steps of the mining wind speed and direction acquisition method based on ultrasonic pulse reflection technology as described in the first aspect are executed.

[0026] Beneficial effects of the present invention:

[0027] This application is based on ultrasonic pulse reflection technology. By dynamically associating tunnel topology characteristics with sensor array collaborative strategies, it optimizes the pulse frequency parameters and multi-source data matching accuracy in real time, effectively solving the problem of relying on manual estimation of the matching relationship between equipment load and process time in the existing technology, making it difficult to quantify the impact of tunnel topology on the ultrasonic propagation path, resulting in continuous deterioration of the signal-to-noise ratio of the reflected signal, expansion of the monitoring blind area, and serious threats to mine ventilation safety and disaster warning timeliness. It effectively eliminates the monitoring blind areas of complex tunnels, improves the efficiency of equipment resource utilization, optimizes the stability and timeliness of wind speed trend prediction, and provides accurate and reliable technical support for mine ventilation safety management.

[0028] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures pointed out in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0030] Figure 1 A schematic flow chart of a method for collecting wind speed and direction in a mine based on ultrasonic pulse reflection technology is shown in the present invention;

[0031] Figure 2 The module schematic diagram of a mining wind speed and direction acquisition system based on ultrasonic pulse reflection technology of the present invention is shown. DETAILED DESCRIPTION

[0032] In order to solve the problems raised by the background technology, this application is based on ultrasonic pulse reflection technology. By dynamically correlating tunnel topology characteristics with sensor array collaborative strategies, it optimizes pulse frequency parameters and multi-source data matching accuracy in real time, effectively eliminates monitoring blind spots in complex tunnels, improves equipment resource utilization efficiency, and optimizes the stability and timeliness of wind speed trend prediction, providing accurate and reliable technical support for mine ventilation safety management.

[0033] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0034] In some embodiments, as Figure 1 As shown, the present application provides a method for collecting wind speed and direction for mining based on ultrasonic pulse reflection technology, comprising:

[0035] S100. Obtain basic data for ultrasonic pulse reflection monitoring, and divide the basic data into a pulse matching feature set and a sensor working timing set.

[0036] S200. Generate a beam path correlation matrix according to the pulse matching feature set.

[0037] S300. Generate an echo optimization parameter set according to the beam path correlation matrix.

[0038] S400. Generate a pulse frequency adjustment instruction set according to the echo optimization parameter set.

[0039] S500. Generate a wind speed prediction period and abnormal data marking results according to the sensor working timing set and the pulse frequency adjustment instruction set.

[0040] In some embodiments, the basic data for ultrasonic pulse reflection monitoring in S100 includes monitoring task data, reflection waveform template data, sensor status data, and historical calibration data.

[0041] The monitoring task data includes the location of the monitoring point, the wind speed value actually measured at the monitoring point through ultrasonic pulse reflection, the preset wind speed threshold corresponding to the monitoring point, and the monitoring emergency level.

[0042] The reflection waveform template data stores the standard ultrasonic reflection waveform characteristics of different tunnel types (tunnel curvature, cross-sectional area, and support), including the tortuosity of the tunnel direction, the spatial constraint of ultrasonic reflection, and the reflection waveform distortion.

[0043] Sensor status data represents the real-time collected sensor working parameters, including sensor model, number of valid data packets actually collected within a specified time, theoretical maximum collection capacity, workload (0-100%).

[0044] Historical calibration data represents the sensor calibration record within a specified period of time in the past, including the number of invalid data packets collected within the specified period and the total number of data packets attempted to be collected.

[0045] In some embodiments, dividing the basic data into a pulse matching feature set and a sensor working timing set in S100 includes:

[0046] S110. Generate the deviation value between the real-time wind speed value and the template and the tunnel structure complexity classification data based on the monitoring task data and the reflected waveform template data; generate the sensor effective working rate timing curve and the signal distortion statistical sequence based on the sensor status data and the historical calibration data.

[0047] Deviation between real-time wind speed value and template Reference formula: ;in, represents the wind speed value actually measured by ultrasonic pulse reflection at the i-th monitoring point, represents the preset wind speed threshold corresponding to the i-th monitoring point, and n represents the total number of key measuring points in the current monitoring task.

[0048] By calculating the deviation between the real-time wind speed values ​​at all measurement points and the template, the degree of airflow anomalies within the mine tunnels is quantified. The greater the deviation between the real-time wind speed values ​​and the template, the more severe the interference of the actual environment (such as turbulence and obstacles) on the ultrasonic propagation path.

[0049] The curvature of the tunnel represents the maximum angle at which the tunnel centerline deviates from a straight line, reflecting the tortuosity of the tunnel's direction. The cross-sectional area represents the effective ventilation area of ​​the tunnel's cross section, which determines the spatial constraint of ultrasonic reflection. Steel frame support offers the highest stability, while wooden support or no support can easily lead to distortion of the reflected waveform. The following rules may be used:

[0050] If it is a straight lane, the lane cross-sectional area>10 , the tunnel adopts steel frame support, then =1; if the roadway curvature is less than 30°, the roadway cross-sectional area is 5-10 , the tunnel adopts composite support, then =2; if the roadway curvature 30°, tunnel cross-sectional area <5 , the tunnel has no support or adopts wooden support, then =3.

[0051] The complexity classification data of the tunnel structure directly affects the propagation path and reflection intensity of the ultrasonic pulse. =3 requires higher frequency pulse transmission to overcome multipath interference.

[0052] Sensor effective working rate Reference formula: ,in, represents the number of valid data packets actually collected by sensor j in time window t, Represents the theoretical maximum acquisition capacity of sensor j.

[0053] The effective working rate reflects the actual working efficiency of the sensor. For example, if , indicating that the sensor performance may have degraded due to hardware aging or environmental interference, and calibration or switching to a backup probe is required.

[0054] Count signal distortion events by time window to generate signal distortion statistics series : ;in, Represents a time window The number of invalid data packets caused by internal multipath reflection, electromagnetic noise, etc. Represents a time window The total number of packets the sensor attempted to collect.

[0055] The sensor's effective operating rate timing curve is used to identify periodic interference events (such as electromagnetic pulses caused by equipment startup and shutdown). If the distortion rate exceeds 5% for three consecutive windows, it is considered a persistent anomaly.

[0056] S120. The deviation value between the real-time wind speed value and the template and the tunnel structure complexity classification data are used as the pulse matching feature set, and the sensor effective working rate timing curve and the signal distortion statistical sequence are used as the sensor working timing set.

[0057] In some embodiments, generating a beam path correlation matrix according to the pulse matching feature set in S200 includes:

[0058] S210. Calculate the matching correction coefficient of the process template based on the deviation value between the real-time wind speed value and the template and the tunnel structure complexity classification data.

[0059] The matching correction coefficient can refer to the formula: , where K represents the matching correction coefficient, Represents the complexity level of the tunnel structure, with a value range of 1-3.

[0060] It is used to measure the matching degree between the actual wind speed and the theoretical reflection threshold. The smaller the value, the larger the K value, indicating that the current monitoring environment has a high degree of matching with the reflection waveform template; the complexity of the tunnel structure is graded. The higher it is, the smaller the correction coefficient K is, and a higher frequency ultrasonic pulse is required to compensate for signal attenuation.

[0061] S220. Re-prioritize the process execution sequence according to the matching degree correction coefficient to generate a collection node priority sequence.

[0062] All collection nodes can be arranged in descending order according to the matching degree correction coefficient K to generate a collection node priority sequence.

[0063] For example, if the correction coefficients of the three acquisition nodes (acquisition node 1, acquisition node 2, and acquisition node 3) are 0.25 ( )、0.17( )、0.10( ), the priority sequence is collection node 1, collection node 2, and collection node 3.

[0064] S230. Based on the priority sequence of the acquisition nodes and the sensor status data, a dynamic association matrix between the acquisition nodes and the sensor array is constructed.

[0065] The behavior of the dynamic correlation matrix is ​​based on the priority sorting of the acquisition nodes. The number of rows is the total number of nodes, the number of columns is the available sensors in the sensor array, the number of columns is the total number of sensors, and the matrix element value is is the association strength between the i-th collection node and the j-th sensor.

[0066] To ensure that the association weight between the node and the sensor is automatically reduced when the signal reliability is insufficient, and to avoid excessively high thresholds that lead to idle sensor resources, the first ratio value can be set. If the workload of the jth sensor is less than the first ratio value (such as 80%), then ,in, represents the matching degree correction coefficient of the i-th node; in other cases, such as sensor overload or unavailable, The first ratio value is determined based on historical calibration data of the workload in the sensor status data and the lane structure complexity classification data.

[0067] S240. Perform integrity verification on the dynamic correlation matrix, convert the verified dynamic correlation matrix into a numerical matrix structure, and generate a beam path correlation matrix.

[0068] Integrity verification such as checking whether each collection node is associated with at least one >0 sensors, such as checking that the total load of the nodes associated with a single sensor does not exceed its maximum workload.

[0069] The dynamic association matrix that has passed integrity verification is converted into a two-dimensional numerical matrix. The row and column indexes correspond to the collection nodes and sensors, respectively. The row index maps the collection node ID and its attributes (monitoring point location, lane structure complexity classification data), and the column index maps the sensor ID, sensor model (high frequency, medium frequency, low frequency) and its status (workload, valid packet count).

[0070] K-value driven priority sorting ensures that monitoring resources are allocated first in complex tunnel areas, high K-value nodes are allocated to high-sensitivity sensors, and low-load sensors receive high-priority tasks first to achieve load balancing.

[0071] In some embodiments, generating an echo optimization parameter set according to the beam path correlation matrix in S300 includes: using an improved decision tree to process the correlation matrix, the improved decision tree is based on the decision tree, adds a dynamic merging mechanism of branch paths, and outputs the echo optimization parameter set.

[0072] In some embodiments, an improved decision tree is used to process the association matrix. The improved decision tree is based on a decision tree architecture, adds a dynamic merging mechanism for branch paths, and outputs an echo optimization parameter set, including:

[0073] S310. Analyze the row and column structure of the beam path correlation matrix, and extract the acquisition node information in the matrix row vectors and the sensor array information in the column vectors.

[0074] Row parsing is used to extract acquisition node information, such as roadway type (coal rock, hard rock, composite layer), monitoring emergency level (emergency (1), routine (0)).

[0075] Column parsing is used to extract sensor array information, such as sensor model (high-frequency type, low-frequency type, workload).

[0076] S320. Divide the first-level decision branches of the improved decision tree based on the lane type in the collected node information.

[0077] The rules of the first-level decision branch may be as follows: if it is a coal rock roadway, then the roadway type = 1; if it is a hard rock roadway, then the roadway type = 2; if it is a composite layer, then the roadway type = 3.

[0078] Coal rock tunnels have a high absorption rate of ultrasonic waves, so the pulse energy needs to be increased (intervals need to be shortened). Hard rock tunnels have strong reflections but a lot of multipath interference, so the frequency needs to be dynamically adjusted. Composite layers have mixed reflection characteristics, so the transmission parameters need to be balanced.

[0079] S330. Under the first-level decision branch, divide the second-level decision branch of the improved decision tree according to the monitoring emergency level in the acquisition node information.

[0080] If it is an emergency monitoring, the emergency level = 1, if it is a routine detection, the emergency level = 0.

[0081] For burst monitoring, the pulse interval can be shortened by 20%, and for routine monitoring, the reference interval can be maintained.

[0082] S340. Under the second-level decision branch, the third-level decision branch of the improved decision tree is divided in combination with the sensor workload in the sensor array information.

[0083] Based on the fluctuation range of sensor signal strength in historical calibration data, combined with the influence of tunnel structure complexity classification data on signal transmission, a balance point is statistically analyzed. For example, if the balance point is 0.6, , then it is low load, if , it is a high load.

[0084] For low load, more tasks can be assigned and the collaborative weight can be improved. For high load, task assignment can be restricted to avoid overload.

[0085] S350. At the end of the third-level decision branch, the discrete decision paths are dynamically merged to generate a continuous parameter space.

[0086] The discrete paths (lane type-urgency-load intensity) are dynamically merged into a continuous parameter space, including coupling the matching correction coefficient with the signal distortion statistical series to eliminate redundant paths. According to the sensor effective working rate timing curve, the discrete load allocation decision (the independent decision result generated by the beam path correlation matrix, such as assigning sensor X to node A and sensor Z to node B at a certain moment) is expanded into a continuous parameter combination within the time window. The continuous space parameters include the pulse transmission interval and the sensor cooperation weight parameter.

[0087] S360. Extract the pulse transmission interval and sensor cooperation weight parameters from the continuous parameter space to generate an echo optimization parameter set.

[0088] Continuous space parameters include the pulse emission interval reference formula: , , where represents the pulse emission interval obtained after merging the paths, Represents the safety factor, which can be 0.85. represents the weight of path k, represents the pulse emission interval predicted by path k, represents the depth of path k in the decision tree, and m represents the total number of decision paths involved in dynamic merging.

[0089] pass Giving priority to shallow paths can effectively avoid accidental errors caused by overly deep branches, and the setting of m allows for flexible increase or decrease in the number of merged paths, thus adapting to different production scenarios.

[0090] The sensor collaboration weight parameter can be calculated based on the sensor workload and process correlation strength, refer to the formula: ,in, Represents the j-th sensor collaborative weight parameter, ranging from 0 to 1. The higher the value, the higher the sensor priority. represents the association strength between the i-th collection node and the j-th sensor, represents the workload of the jth sensor.

[0091] Output echo optimization parameter set including pulse transmission interval , sensor collaborative weight parameters .

[0092] In some embodiments, generating a pulse frequency adjustment instruction set according to the echo optimization parameter set in S400 includes:

[0093] S410. Extract the pulse transmission interval and sensor cooperation weight parameters in the echo optimization parameter set; calculate the effective monitoring time window based on the sensor effective working rate timing curve in the sensor working timing set.

[0094] Setting effective work rate threshold , effective working rate threshold This can be determined by establishing and analyzing the distortion rate-effective working rate relationship curve and maximizing sensor utilization while ensuring data reliability.

[0095] The effective monitoring time window can be: , if and only if the sensor effective working rate , Represents the start time of the effective monitoring time window, Indicates the end time of the effective monitoring time window.

[0096] S420. Match the pulse emission interval parameter with the effective monitoring time window to generate a basic speed control instruction.

[0097] In the effective monitoring time window Inside, press Set the pulse transmission cycle and suspend pulse transmission during non-effective periods.

[0098] S430. Combine the basic speed control instructions and sensor collaborative weight parameters to generate multi-probe collaborative acquisition instructions, and encode them into a pulse frequency adjustment instruction set according to the sensor array type.

[0099] The multi-probe collaborative acquisition instruction represents the collaboration rules of multiple sensors within the time window. The collaboration rules are as follows: within the same time window, sensors with high sensor collaboration weight parameters are given priority in allocating monitoring point positions, and sensors with low sensor collaboration weight parameters delay pulse emission time to avoid signal interference.

[0100] The coordination rules are encoded as binary instructions according to the column index of the dynamic incidence matrix.

[0101] In some embodiments, generating a wind speed prediction period and abnormal data marking results according to the sensor working timing set and the pulse frequency adjustment instruction set in S500 includes:

[0102] An improved differential autoregressive moving average model is used to process the sensor working timing set and the pulse frequency adjustment instruction set. The improved differential autoregressive moving average model uses the differential autoregressive moving average model as the basic architecture, adds a parallel parsing module for multi-source instructions, and outputs the wind speed prediction period and abnormal data marking results.

[0103] In some embodiments, an improved differential autoregressive moving average model is used to process the sensor working timing set and the pulse frequency adjustment instruction set. The improved differential autoregressive moving average model uses the differential autoregressive moving average model as a basic architecture, adds a parallel parsing module for multi-source instructions, and outputs wind speed prediction period and abnormal data marking results, including:

[0104] S510. Use the differential autoregressive moving average model to analyze the sensor effective working rate timing curve and signal distortion statistical series in the sensor working timing set to extract the real-time operation efficiency characteristics of the equipment; use the parallel analysis module to analyze the basic speed control instructions and multi-probe collaborative acquisition instructions in the pulse frequency adjustment instruction set to extract the instruction timing distribution characteristics.

[0105] The autoregressive integrated moving average (ARIMA) model is used to analyze the sensor working time series data, and the parallel analysis module is used to simultaneously process basic speed control instructions and multi-probe collaborative acquisition instructions.

[0106] The real-time operation efficiency characteristics of the equipment include the equipment operation stability index and the signal distortion frequency.

[0107] Based on the sensor's effective operating rate time series curve, the average operating rate variance within the sliding window is calculated to obtain the device operation stability index. Based on the signal distortion statistical series, the distortion event density per unit time is counted to obtain the signal distortion frequency.

[0108] The command timing distribution characteristics include the basic speed control command timing distribution and the correlation of multi-probe collaborative commands.

[0109] The distribution pattern of the pulse emission interval parameters on the time axis is extracted to obtain the basic speed control instruction timing distribution, such as the 45ms interval instructions are concentrated between 08:00 and 12:00.

[0110] According to the sensor coordination weight parameters, the temporal overlap of multi-sensor commands is analyzed to obtain the correlation of multi-probe coordination commands, such as sensor 1 and sensor 2 executing high-weight tasks at the same time at time t1.

[0111] S520. Superimpose and analyze the real-time operation efficiency characteristics of the device and the instruction timing distribution characteristics to generate a sensor array efficiency index.

[0112] The instruction density is obtained by dividing the number of pulse transmissions by the length of the effective monitoring time window. The sensor array effectiveness index E is obtained by dividing the stability index by the instruction density, which quantifies the overall efficiency of the sensor array under specific instructions.

[0113] S530. Based on the sensor array efficiency index and pulse emission interval parameters, the completion time of each acquisition node is predicted to generate a wind speed prediction cycle; based on the timing overlap between the signal distortion statistical sequence and the multi-probe collaborative acquisition instructions, abnormal acquisition nodes are detected and abnormal data marking results are generated.

[0114] According to the efficiency index, the pulse interval parameter is modified and the wind speed prediction period is output. The reference formula is: ,in, Represents the wind speed prediction period, which is the data completion time of each collection node.

[0115] If a monitoring point is associated with multiple (greater than or equal to 2) high-load sensors simultaneously during the pulse emission interval, it is marked as a timing overlap anomaly;

[0116] If a collection node's timestamp coincides with a distortion event in the signal distortion statistical sequence during the time series overlap anomaly period, it is marked as a valid anomaly. This double verification reduces the false positive rate and ensures the accuracy of anomaly marking.

[0117] In some embodiments, as Figure 2As shown, this application provides a clothing production management system based on template operation, including:

[0118] Data division module: obtains basic data of ultrasonic pulse reflection monitoring, and divides the basic data into a pulse matching feature set and a sensor working timing set.

[0119] Matrix generation module: generates a beam path correlation matrix according to the pulse matching feature set.

[0120] Parameter set generation module: generates an echo optimization parameter set according to the beam path correlation matrix.

[0121] Instruction set generation module: generates a pulse frequency adjustment instruction set according to the echo optimization parameter set.

[0122] Result generation module: generates wind speed prediction cycle and abnormal data marking results according to the sensor working timing set and the pulse frequency adjustment instruction set.

[0123] This embodiment has all the advantages of a method for collecting wind speed and direction in a mine based on ultrasonic pulse reflection technology, and can automatically execute the steps of a method for collecting wind speed and direction in a mine based on ultrasonic pulse reflection technology.

[0124] In some embodiments, the present application provides a mining wind speed and direction collection device based on ultrasonic pulse reflection technology, which includes a memory and a processor; the memory is used to store computer programs; the processor is used to implement the steps of a mining wind speed and direction collection method based on ultrasonic pulse reflection technology when executing the computer program.

[0125] In some embodiments, the present application provides a storage medium storing computer program instructions. When the computer program instructions are read and executed by a processor, the steps of a method for collecting wind speed and direction for mining based on ultrasonic pulse reflection technology are executed.

[0126] Any reference to memory, storage, database, or other media used in the embodiments provided herein may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory.

[0127] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device that includes a series of elements includes not only those elements, but also other elements that are not explicitly listed, or elements that are inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device that includes the element.

[0128] Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for collecting wind speed and direction in mines based on ultrasonic pulse reflection technology, characterized in that: include: Basic data from ultrasonic pulse reflection monitoring is acquired and classified to obtain a pulse matching feature set and a sensor operating time series set. The pulse matching feature set represents the deviation between the real-time wind speed measurement and the reflection waveform template, as well as the influence of the tunnel structure complexity on signal matching. The sensor operating time series set represents the statistical characteristics of the sensor operating status and signal quality over time. Based on the pulse matching feature set, path correlation analysis and optimization processing are performed to obtain a beam path correlation matrix; wherein the beam path correlation matrix represents the dynamic collaborative mapping relationship between the acquisition node and the sensor array; Based on the beam path correlation matrix, signal optimization parameters are calculated to obtain an echo optimization parameter set; wherein the echo optimization parameter set is used to optimize the quality of the ultrasonic echo signal; Based on the echo optimization parameter set, instruction conversion processing is performed to obtain a pulse frequency adjustment instruction set; wherein the pulse frequency adjustment instruction set is used to coordinate the sensor operating frequency and the coordination mode; Based on the sensor working timing set and the pulse frequency adjustment instruction set, working state integration and abnormality judgment processing are performed to obtain wind speed prediction period and abnormal data marking results.

2. The method for collecting wind speed and direction in mining based on ultrasonic pulse reflection technology according to claim 1 is characterized in that: The basic data of the ultrasonic pulse reflection monitoring includes monitoring task data, reflection waveform template data, sensor status data and historical calibration data.

3. The method for collecting wind speed and direction in mining based on ultrasonic pulse reflection technology according to claim 2 is characterized in that: The basic data is classified and processed to obtain a pulse matching feature set and a sensor working timing set, including: Based on the monitoring task data and the reflected waveform template data, the deviation value between the real-time wind speed value and the template and the tunnel structure complexity classification data are generated. Based on the sensor status data and historical calibration data, the sensor effective working rate time series curve and signal distortion statistical series are generated. The deviation between the real-time wind speed value and the template and the tunnel structure complexity classification data are used as the pulse matching feature set, and the sensor effective working rate timing curve and signal distortion statistical sequence are used as the sensor working timing set.

4. The method for collecting wind speed and direction in mining based on ultrasonic pulse reflection technology according to claim 3 is characterized in that: Based on the pulse matching feature set, path correlation analysis and optimization processing are performed to obtain a beam path correlation matrix, including: Calculate the matching correction coefficient of the process template based on the deviation between the real-time wind speed value and the template and the tunnel structure complexity classification data; Re-prioritize the process execution order according to the matching degree correction coefficient to generate a collection node priority sequence; Based on the priority sequence of acquisition nodes and sensor status data, a dynamic association matrix between acquisition nodes and sensor arrays is constructed; The integrity of the dynamic correlation matrix is ​​verified, and the verified dynamic correlation matrix is ​​converted into a numerical matrix structure to generate a beam path correlation matrix.

5. The method for collecting wind speed and direction in mining based on ultrasonic pulse reflection technology according to claim 4 is characterized in that: Based on the beam path correlation matrix, signal optimization parameters are calculated to obtain an echo optimization parameter set, including: An improved decision tree is used to process the association matrix. The improved decision tree takes the decision tree as the basic architecture, adds a dynamic merging mechanism of branch paths, and outputs an echo optimization parameter set.

6. The method for collecting wind speed and direction in mining based on ultrasonic pulse reflection technology according to claim 5 is characterized in that: An improved decision tree is used to process the association matrix. The improved decision tree is based on a decision tree as a basic architecture, adds a dynamic merging mechanism for branch paths, and outputs an echo optimization parameter set, including: Analyze the row and column structure of the beam path correlation matrix and extract the acquisition node information in the matrix row vectors and the sensor array information in the column vectors; Based on the lane type in the collected node information, the first-level decision branches of the improved decision tree are divided; Under the first-level decision branch, the second-level decision branch of the improved decision tree is divided according to the monitoring emergency level in the collected node information; Under the second-level decision branch, the third-level decision branch of the improved decision tree is divided in combination with the sensor workload in the sensor array information; At the end of the third-level decision branch, the discrete decision paths are dynamically merged to generate a continuous parameter space; The pulse transmission interval and sensor cooperation weight parameters are extracted from the continuous parameter space to generate the echo optimization parameter set.

7. The method for collecting wind speed and direction in mining based on ultrasonic pulse reflection technology according to claim 4 is characterized in that: Based on the echo optimization parameter set, instruction conversion processing is performed to obtain a pulse frequency adjustment instruction set, including: Extract the pulse emission interval and sensor coordination weight parameters from the echo optimization parameter set; calculate the effective monitoring time window based on the sensor effective working rate timing curve from the sensor working timing set; Match the pulse emission interval parameters with the effective monitoring time window to generate basic speed control instructions; Combining the basic speed control instructions and sensor collaborative weight parameters, multi-probe collaborative acquisition instructions are generated and encoded into a pulse frequency adjustment instruction set according to the sensor array type.

8. The method for collecting wind speed and direction in mining based on ultrasonic pulse reflection technology according to claim 1 is characterized in that: Based on the sensor working timing set and the pulse frequency adjustment instruction set, working state integration and abnormality judgment processing are performed to obtain wind speed prediction period and abnormal data marking results, including: An improved differential autoregressive moving average model is used to process the sensor working timing set and the pulse frequency adjustment instruction set. The improved differential autoregressive moving average model uses the differential autoregressive moving average model as the basic architecture, adds a parallel parsing module for multi-source instructions, and outputs the wind speed prediction period and abnormal data marking results.

9. The method for collecting wind speed and direction in mining based on ultrasonic pulse reflection technology according to claim 8, characterized in that: An improved differential autoregressive moving average model is used to process the sensor working timing set and the pulse frequency adjustment instruction set. The improved differential autoregressive moving average model uses the differential autoregressive moving average model as the basic architecture, adds a parallel parsing module for multi-source instructions, and outputs the wind speed prediction period and abnormal data marking results, including: The differential autoregressive moving average model is used to analyze the sensor effective working rate timing curve and signal distortion statistical series in the sensor working timing set to extract the real-time operation efficiency characteristics of the equipment. The parallel analysis module is used to analyze the basic speed regulation instructions and multi-probe collaborative acquisition instructions in the pulse frequency adjustment instruction set to extract the instruction timing distribution characteristics. Superimpose and analyze the real-time operation efficiency characteristics of the equipment and the instruction timing distribution characteristics to generate the sensor array effectiveness index; Based on the sensor array efficiency index and pulse emission interval parameters, the completion time of each acquisition node is predicted and the wind speed prediction period is generated. According to the timing overlap between the signal distortion statistical sequence and the multi-probe collaborative acquisition instructions, abnormal acquisition nodes are detected and abnormal data marking results are generated.

10. A mining wind speed and direction acquisition system based on ultrasonic pulse reflection technology, characterized in that: include: Data partitioning module: This module obtains basic data from ultrasonic pulse reflection monitoring and classifies it to obtain a pulse matching feature set and a sensor working sequence set. The pulse matching feature set represents the deviation between the real-time wind speed measurement and the reflection waveform template, as well as the influence of the tunnel structure complexity on signal matching. The sensor working sequence set represents the statistical characteristics of the sensor working status and signal quality over time. Matrix generation module: Based on the pulse matching feature set, path correlation analysis and optimization are performed to obtain a beam path correlation matrix; wherein the beam path correlation matrix represents the dynamic collaborative mapping relationship between the acquisition node and the sensor array; A parameter set generation module is configured to calculate signal optimization parameters based on the beam path correlation matrix to obtain an echo optimization parameter set; wherein the echo optimization parameter set is used to optimize the quality of the ultrasonic echo signal; An instruction set generation module: performing instruction conversion processing based on the echo optimization parameter set to obtain a pulse frequency adjustment instruction set; wherein the pulse frequency adjustment instruction set is used to coordinate the sensor operating frequency and the coordination mode; Result generation module: Based on the sensor working timing set and the pulse frequency adjustment instruction set, the module performs working state integration and abnormality judgment processing to obtain wind speed prediction period and abnormal data marking results.

Images