A coal mine underground node seismic data real-time monitoring system

Abstract

This invention provides a real-time monitoring system for underground coal mine seismic data, comprising a MEMS all-digital three-component geophone for real-time acquisition of seismic wave signals in three-dimensional space underground; a node acquisition unit for real-time reception, local storage, and preliminary processing of seismic wave data; an intrinsically safe power supply module employing intrinsically safe circuit design compliant with protection standards to provide safe power supply to the node devices; a high-precision time synchronization module using dual-crystal high-precision clock synchronization technology with AI temperature compensation to monitor and compensate for frequency drift caused by temperature changes in the crystal oscillators in real time; and an adaptive wireless communication module that automatically selects the optimal network for real-time data transmission based on the actual underground communication environment. This invention features high dynamic range, high sensitivity, large storage capacity, high reliability, low power consumption, low weight, and real-time data transmission capabilities, meeting the real-time acquisition and processing needs of seismic data in the complex environment of underground coal mines.

Description

Technical Field

[0001] This invention relates to the field of coal mine seismic exploration technology, specifically to a real-time monitoring system for underground coal mine node seismic data, which is suitable for real-time acquisition, transmission and processing of seismic data in complex underground coal mine environments. Background Technology

[0002] Seismic exploration technology is an indispensable tool in coal resource exploration and mining, effectively discovering coal resources and solving structural, stratigraphic, and lithological problems encountered during mining. However, traditional seismic acquisition equipment has many limitations, such as being bulky, complex to install, having low trigger sensitivity, and weak resistance to electromagnetic interference, making it difficult to meet the high-precision, high-efficiency seismic data acquisition needs of the complex underground environment of coal mines.

[0003] With the development of the Internet of Things, wireless communication, and high-precision sensor technologies, nodal seismic data acquisition systems have gradually become a research hotspot. However, existing nodal devices still face many challenges when applied in underground coal mines, such as insufficient clock synchronization accuracy, large device size, high operating costs, and weak resistance to electromagnetic interference. Therefore, developing a high-precision, low-power, and electromagnetically interference-resistant nodal seismic data real-time monitoring system suitable for the complex environment of underground coal mines is of significant practical importance. Summary of the Invention

[0004] The purpose of this invention is to provide a real-time monitoring system for seismic data at underground nodes in coal mines. This system has high dynamic range, high sensitivity, large storage capacity, high reliability, low power consumption, low weight, and real-time data transmission capabilities, which can meet the needs of real-time acquisition and processing of seismic data in the complex environment of underground coal mines.

[0005] The present invention achieves the above objectives through the following technical solutions: A real-time monitoring system for underground seismic data in coal mines includes: MEMS fully digital three-component geophone is used to acquire seismic wave signals in three-dimensional space underground in coal mines in real time. The node acquisition unit is connected to the MEMS all-digital three-component geophone and is used for real-time reception, local storage and preliminary processing of seismic wave data. The intrinsically safe power supply module adopts an intrinsically safe circuit design that complies with GB 3836.1-2010 and GB 3836.4-2010 standards to provide a safe power supply for node devices; The high-precision time synchronization module adopts dual-crystal high-precision clock synchronization technology that integrates AI temperature compensation, and monitors and compensates for frequency drift of the crystal oscillator caused by temperature changes in real time. The adaptive wireless communication module automatically selects the optimal network for real-time data transmission based on the actual underground communication environment. Specifically, it matches and calculates the collected environmental parameters with the service QoS requirements to generate a comprehensive score for each candidate network standard. The network standard with the highest comprehensive score is then determined as the current optimal network, and the multi-mode communication chip is controlled to access this network for data transmission.

[0006] According to the present invention, a real-time monitoring system for underground seismic data in a coal mine includes an adaptive wireless communication module comprising: The multi-mode communication chip adopts an integrated chip that supports multi-band radio frequency transceivers and baseband processors. It has an integrated hardware acceleration module for directly processing physical layer signaling parsing and writing the parsed channel quality index (CQI) into the register of the intelligent handover control unit in real time through shared memory or direct memory access mechanism. The environmental sensing unit, including a broadband spectrum scanning sub-circuit and an auxiliary sensor interface, is connected to the intelligent switching control unit via an I2C or SPI high-speed serial bus. It is used to collect electromagnetic noise floor noise, tunnel vibration characteristics, and temperature and humidity data of the underground environment. The intelligent switching control unit is implemented based on a field-programmable gate array or a high-performance ARM Cortex series microcontroller. It communicates bidirectionally with the application processor (AP) side of the multi-mode communication chip through a PCIe or UART interface, thus constructing a transmission channel that separates the control plane and the data plane. The power management and isolation unit, including a DC-DC buck converter circuit and a magnetic isolation chip, is used to convert the wide downhole voltage input into the stable operating voltage required by each chip and to achieve electrical isolation between the control logic ground and the power ground. The multi-mode communication chip, environmental sensing unit, and intelligent switching control unit are integrated and packaged in the same metal shield. The shield is made of nickel-plated copper alloy or stainless steel, and an LC filter circuit and transient voltage suppression diode are set at the radio frequency signal input end to filter out pulse interference generated by downhole electromechanical equipment.

[0007] According to the present invention, a real-time monitoring system for underground seismic data in coal mines includes an intelligent switching control unit comprising: The environmental parameter acquisition subunit is used to acquire physical layer parameters and network layer parameters of the downhole communication environment in real time. The parameters include at least the received signal strength indication, signal-to-noise ratio, reference signal reception quality, channel quality indication, network load rate, transmission delay and packet loss rate for each network standard. The business requirements analysis subunit is used to analyze the business characteristics of the data to be transmitted and determine the corresponding business QoS requirement thresholds, including minimum bandwidth requirements, maximum allowable latency and reliability level. The multi-dimensional decision subunit is used to match and calculate the collected environmental parameters with the service QoS requirements based on a preset weighted scoring algorithm or fuzzy logic model, and generate a comprehensive score for each candidate network standard; among them, the weighted scoring algorithm pre-configures weight coefficients for the signal attenuation characteristics of different network standards according to the underground roadway topology.

[0008] According to the present invention, a real-time monitoring system for underground seismic data in coal mines includes a business requirements analysis subunit comprising: The packet feature extraction module is used to scan the IP packets of the data to be transmitted in real time and extract the five-tuple information and application layer payload features of the packets. The business feature matching module is used to compare the extracted features with the pre-set business feature library to identify the specific type of the current business. The business types include at least video surveillance stream, gas concentration monitoring data, equipment control signaling, voice call and file transfer. The QoS parameter mapping module is used to query the corresponding QoS policy table based on the identified service type and dynamically determine the service quality requirement threshold. The QoS policy table predefines priority queues, minimum guaranteed bandwidth, maximum retransmission count, and latency jitter tolerance for different service types.

[0009] According to the real-time monitoring system for underground seismic data in coal mines provided by the present invention, the multi-dimensional decision-making subunit performs the following weighted scoring calculation process: The parameter normalization process maps the collected physical layer parameters and network layer parameters of each network type to dimensionless evaluation values ​​in the interval [0, 1]. For negative indices, the reciprocal method or linear interpolation method is used for normalization. The dynamic weight vector is determined by calling the preset underground topology weight table and determining the weight coefficients of each evaluation dimension based on the current roadway type and movement speed of the module. W =[ w 1, w 2,..., w n ]; In long, straight tunnels, the signal coverage weight of 5G / 4G is w cov Higher than WiFi; at junctions or areas with dense devices, network load balancing weight w load Upgrade to the highest level; The service matching degree calculation involves performing Euclidean distance calculation or cosine similarity matching between the normalized network parameter matrix and the service QoS requirement vector to generate a service matching factor. α ; A comprehensive score is generated using a formula. Score i =∑(w j × P ij )× α Calculate the first i The overall score of each candidate network standard, among which P ij For the first i The network in the first j Normalized values ​​in each dimension; Among them, the comprehensive score is selected. Score The network type that is the largest and greater than the preset switching threshold is selected as the current optimal network.

[0010] According to the present invention, a real-time monitoring system for underground seismic data in coal mines is provided, wherein the multi-dimensional decision sub-unit is implemented based on a type-II fuzzy logic model, specifically including: The fuzzification interface is used to convert environmental parameters into fuzzy linguistic variables, define fuzzy sets as {very poor (VB), poor (B), medium (M), good (G), very good (VG)}, and establish corresponding trapezoidal or Gaussian membership functions; A fuzzy rule base, pre-stored with IF-THEN inference rules built based on downhole expert experience; The fuzzy inference engine uses the Mamdani or Sugeno inference mechanism. Based on the current input fuzzy variables, it activates multiple rules in the fuzzy rule base and performs logical synthesis to obtain the fuzzy output set of each candidate network. The defuzzing module uses the centroid method or the maximum membership method to convert the fuzzy output set into priority values, and sorts the candidate networks in descending order according to the size of the priority values, selecting the network with the highest ranking as the optimal network.

[0011] According to the present invention, a real-time monitoring system for underground seismic data in coal mines includes a MEMS all-digital three-component detector with a single-chip system-in-package structure, specifically comprising: The MEMS sensor uses deep reactive ion etching to fabricate three orthogonally arranged micromechanical cantilever beam mass block structures on the same silicon wafer, corresponding to vibration sensing along the X, Y, and Z axes, respectively. The capacitance detection array, located below the micromechanical cantilever beam mass block structure, is used to detect the differential capacitance change caused by the displacement of the mass block and convert the mechanical displacement signal into an analog charge signal. The digital conversion ASIC circuit is flip-chip bonded to the MEMS sensor and integrates a low-noise charge amplifier, a high-order Sigma-Delta analog-to-digital converter, and a digital filtering unit. Among them, the digital conversion ASIC circuit directly converts the analog charge signal into a 24-bit digital signal inside the chip, and outputs it directly to the intelligent switching control unit through the I2S or PDM digital audio interface. The temperature compensation module, integrated into the ASIC circuit, is used to correct the zero-point drift and sensitivity coefficient of the MEMS sensor in real time according to the downhole ambient temperature.

[0012] According to the present invention, a real-time monitoring system for underground seismic data in coal mines includes a node acquisition unit comprising: The dual-level storage architecture includes a volatile cache and a non-volatile flash array. During normal operation, the real-time received seismic wave data is first stored in the cache. When the cache level reaches a preset threshold or a network interruption signal is detected, a disk flushing operation is automatically triggered to write the data blocks to the flash array. The power failure protection circuit is connected to the main power supply terminal of the node acquisition unit and includes a supercapacitor or backup lithium battery. It is used to maintain the operation of the storage control circuit when the main power supply is unexpectedly disconnected. The log index file system generates an index header file with a high-precision timestamp for each segment of stored data, recording the acquisition time, detector ID, and data length; When the network connection is restored, the node acquisition unit automatically calculates the data segments that have not been uploaded based on the index file, prioritizes the uploading of historical data in chronological order, and then transmits real-time data to avoid data out of order.

[0013] According to the present invention, a real-time monitoring system for underground seismic data in coal mines includes an intrinsically safe power supply module comprising: Intrinsically safe conversion circuit, used to convert external input power supply to intrinsically safe output, includes voltage clamping circuit and current limiting circuit; The safety parameter sampling unit acquires the output voltage of the intrinsically safe side in real time through a high-precision ADC. V out and output current I out And the surface temperature of key components; The energy integration calculation module calculates the cumulative released energy of the circuit in real time based on sampled data. E =∫ V ( t ) I ( t ) dt When the calculated cumulative energy approaches 80% of the minimum ignition energy specified in GB 3836.4 standard, a first-level warning signal is generated. The dynamic power limiting logic module, connected to the intelligent switching control unit, automatically reduces the output voltage or cuts off power to non-critical loads when it receives a high power consumption warning or detects an abnormal internal temperature, retaining only the minimum power supply to the MEMS detector and the main control chip.

[0014] According to the present invention, a real-time monitoring system for underground seismic data in coal mines includes a high-precision time synchronization module comprising: The dual crystal clock source consists of a high-stability temperature-controlled crystal oscillator (OCXO) and a low-power temperature-compensated crystal oscillator (TCXO) connected in parallel. The OCXO serves as the master clock source, providing a high-precision reference frequency, while the TCXO serves as the slave clock source, used for timekeeping during low-power standby or OCXO warm-up. A multi-dimensional sensing unit is used to collect temperature gradient data on the surface of the crystal oscillator housing and ripple noise of the power supply voltage in real time. The AI ​​thermal compensation engine is deployed with a frequency drift prediction model based on a BP neural network or a long short-term memory network; the input layer of this model includes the current temperature. T Temperature change rate dT / dt and historical frequency deviation Δ f hist The output layer is the predicted correction value Δ for the crystal oscillator frequency. f pred The model training process incorporates the thermal conductivity differential equation of the crystal oscillator as a physical constraint. The frequency locking and calibration unit, based on the Δ output of the AI ​​thermal compensation engine, f pred The output frequency of the OCXO can be finely adjusted in real time using a numerically controlled oscillator or a phase-locked loop; When an OCXO fault or excessive power consumption is detected, the system seamlessly switches to a TCXO and uses the A thermal compensation engine to perform the same temperature compensation on the TCXO.

[0015] Therefore, compared with the prior art, the real-time monitoring system for underground seismic data in coal mines proposed in this invention has the following advantages: 1. This invention uses a high-precision MEMS sensor and a high dynamic range A / D converter to achieve high-precision acquisition and processing of seismic wave signals.

[0016] 2. This invention integrates AI temperature compensation dual-crystal high-precision clock synchronization technology to ensure time synchronization accuracy in environments without GPS signals downhole.

[0017] 3. The multi-mode wireless communication module of the present invention supports multiple communication protocols to ensure real-time and stable transmission of earthquake data.

[0018] 4. The node equipment adopts an intrinsically safe circuit design, which meets the safety standards for underground coal mines and reduces the risk of accidents.

[0019] 5. The node data management system of this invention enables remote monitoring and intelligent data analysis of node devices, thereby improving work efficiency and data quality.

[0020] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments. Attached Figure Description

[0021] Figure 1 This is a structural schematic diagram of an embodiment of a real-time monitoring system for underground seismic data in coal mines according to the present invention.

[0022] Figure 2 This is a schematic diagram of the structure of a MEMS all-digital three-component detector in an embodiment of a real-time monitoring system for underground seismic data in a coal mine, according to the present invention.

[0023] Figure 3 This is a structural schematic diagram of the node acquisition unit in an embodiment of a real-time monitoring system for underground seismic data in a coal mine according to the present invention.

[0024] Figure 4 This is a structural schematic diagram of an intrinsically safe power supply module in an embodiment of a real-time monitoring system for underground seismic data in a coal mine, according to the present invention.

[0025] Figure 5 This is a structural schematic diagram of the adaptive wireless communication module in an embodiment of a real-time monitoring system for underground seismic data in a coal mine according to the present invention.

[0026] Figure 6 This is a structural schematic diagram of the intelligent switching control unit in an embodiment of a real-time monitoring system for underground seismic data in a coal mine according to the present invention. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0028] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0029] See Figure 1 This embodiment provides a real-time monitoring system for underground seismic data in coal mines, including: MEMS fully digital three-component geophone is used to acquire seismic wave signals in three-dimensional space underground in coal mines in real time. The node acquisition unit is connected to the MEMS all-digital three-component geophone and is used for real-time reception, local storage and preliminary processing of seismic wave data. The intrinsically safe power supply module adopts an intrinsically safe circuit design that complies with GB 3836.1-2010 and GB 3836.4-2010 standards to provide a safe power supply for node devices; The high-precision time synchronization module adopts dual-crystal high-precision clock synchronization technology that integrates AI temperature compensation, and monitors and compensates for frequency drift of the crystal oscillator caused by temperature changes in real time. The adaptive wireless communication module automatically selects the optimal network for real-time data transmission based on the actual underground communication environment. Specifically, it matches and calculates the collected environmental parameters with the service QoS requirements to generate a comprehensive score for each candidate network standard. The network standard with the highest comprehensive score is then determined as the current optimal network, and the multi-mode communication chip is controlled to access this network for data transmission.

[0030] As shown in Table 1, Table 1 presents the nodal data acquisition technical specifications of the MEMS all-digital three-component detector. Table 1: Technical Specifications of Node Data Acquisition System

[0031] In this embodiment, as Figure 2 As shown, the MEMS all-digital three-component detector adopts a single-chip system-in-package (SiP) structure, specifically including: The MEMS sensor uses deep reactive ion etching (DRIE) to fabricate three orthogonally arranged micromechanical cantilever beam mass blocks on the same silicon wafer, corresponding to vibration sensing along the X, Y, and Z axes, respectively, with sensitivity coefficient deviations of less than ±1% across the three axes. The capacitance detection array, located below the micromechanical cantilever beam mass block structure, is used to detect the differential capacitance change caused by the displacement of the mass block and convert the mechanical displacement signal into an analog charge signal. The digital conversion ASIC circuit is flip-chip bonded to the MEMS sensor and integrates a low-noise charge amplifier, a high-order Sigma-Delta analog-to-digital converter (ADC), and a digital filtering unit. The digital conversion ASIC circuit directly converts the analog charge signal into a 24-bit digital signal inside the chip, and outputs it directly to the intelligent switching control unit through the I2S or PDM digital audio interface, without the need for an external ADC chip. The temperature compensation module, integrated into the ASIC circuit, is used to correct the zero drift and sensitivity coefficient of the MEMS sensor in real time according to the downhole ambient temperature (0-60℃).

[0032] In this embodiment, the node acquisition unit includes: The dual-level storage architecture includes a volatile cache (DDR3 / DDR4) and a non-volatile flash array. During normal operation, the real-time received seismic wave data is first stored in the cache. When the cache level reaches a preset threshold or a network interruption signal is detected, a disk flushing operation is automatically triggered to write the data blocks to the flash array. The power failure protection circuit is connected to the main power supply terminal of the node acquisition unit and includes a supercapacitor or backup lithium battery. When the main power supply is unexpectedly disconnected, the storage control circuit is maintained to work for no less than 10 seconds to ensure that the critical seismic wave data of the last 10 seconds in the cache is completely written to the flash memory. The log index file system generates an index header file with a high-precision timestamp (accurate to the microsecond level) for each segment of stored data, recording the acquisition time, detector ID, and data length; When the network connection is restored, the node acquisition unit automatically calculates the data segments that have not been uploaded based on the index file, prioritizes the uploading of historical data in chronological order, and then transmits real-time data to avoid data out of order.

[0033] In this embodiment, as Figure 4 As shown, the intrinsically safe power supply module includes: Intrinsically safe conversion circuit, used to convert external input power supply to intrinsically safe output, includes voltage clamping circuit and current limiting circuit; The safety parameter sampling unit acquires the output voltage of the intrinsically safe side in real time through a high-precision ADC. V out and output current I out And the surface temperature of key components (such as current-limiting resistors and Zener diodes); The energy integration calculation module calculates the cumulative released energy of the circuit in real time based on sampled data. E =∫ V ( t ) I ( t ) dt When the calculated cumulative energy approaches 80% of the minimum ignition energy (e.g., 200 μJ) specified in GB 3836.4 standard, a first-level warning signal is generated. The dynamic power limiting logic module, connected to the intelligent switching control unit, automatically reduces the output voltage or cuts off power to non-critical loads (such as video transmitters) when it receives a high power consumption warning or detects an abnormal internal temperature, while only retaining the minimum power supply to the MEMS detector and the main control chip, ensuring that the core monitoring functions operate in an intrinsically safe state.

[0034] Due to the harsh working environment in coal mines, including strong electromagnetic interference, highly conductive carbon dust, high humidity, and flammable and explosive gases, the equipment must possess high electromagnetic interference resistance. Simultaneously, the power supply section of the node equipment must adopt an intrinsically safe circuit design, strictly complying with the relevant specifications of GB 3836.1—2010 "Explosive Atmospheres Part 1: Equipment General Requirements" and GB3836.4—2010 "Explosive Atmospheres Part 4: Equipment Protected by Intrinsically Safe 'i'". In the intrinsically safe power module, to adapt to the needs of working in confined underground spaces, the equipment structure dimensions have been optimized to 122mm × 108mm × 92mm (length × width × height), with a weight controlled to ≤1.0kg, facilitating manual transportation, rapid deployment, and retrieval. The housing is made of polycaprolactam (PA6) / glass fiber (GF) 15% reinforced plastic, with a wear-resistant, corrosion-resistant, and anti-static coating. The sealing structure uses a combination of double O-rings and sealant, achieving an IP protection rating of IP68. It can operate continuously for 180 days without failure in environments with carbon dust concentrations of 50g / m³ and relative humidity of 95% (40℃), ensuring stable operation in harsh environments. The housing surface is coated with an electromagnetic shielding coating, effectively resisting electromagnetic interference in the frequency range of 30MHz-1GHz and a field strength of 200V / m, ensuring the stability of signal acquisition and transmission.

[0035] In this embodiment, the high-precision time synchronization module includes: The dual crystal clock source consists of a high-stability temperature-controlled crystal oscillator (OCXO) and a low-power temperature-compensated crystal oscillator (TCXO) connected in parallel. The OCXO serves as the master clock source, providing a high-precision reference frequency, while the TCXO serves as the slave clock source, used for timekeeping during low-power standby or OCXO warm-up. The multi-dimensional sensing unit is used to collect temperature gradient data (including current temperature, temperature change rate and historical temperature curve) and power supply voltage ripple noise on the surface of the crystal oscillator in real time. The AI ​​thermal compensation engine is deployed with a frequency drift prediction model based on a BP neural network or a Long Short-Term Memory (LSTM) network; the input layer of this model includes the current temperature. TTemperature change rate dT / dt and historical frequency deviation Δ f hist The output layer is the predicted correction value Δ for the crystal oscillator frequency. f pred The model training process incorporates the thermal conduction differential equation of the crystal oscillator as a physical constraint to ensure that the AI ​​prediction results conform to the thermal inertia characteristics of the crystal oscillator material. The frequency locking and calibration unit, based on the Δ output of the AI ​​thermal compensation engine, f pred The output frequency of the OCXO is finely adjusted in real time using a numerically controlled oscillator (NCO) or a phase-locked loop (PLL) to control the frequency stability within 1×10⁻⁶. Within the order of 10; When an OCXO fault or excessive power consumption is detected, the system seamlessly switches to a TCXO and uses the A thermal compensation engine to perform the same temperature compensation on the TCXO, ensuring that the time phase jump at the moment of switching is less than 10ns.

[0036] As can be seen, the high-precision time synchronization module in this embodiment can overcome the technical bottleneck of no GPS signal downhole and clock drift caused by temperature fluctuations. For the first time, it combines the AI ​​temperature compensation model with the dual crystal oscillator redundancy design. Through BP neural network training at multiple temperature points, it establishes a precise temperature-frequency error compensation mechanism, achieving a timing error of ≤10μs, which is far superior to the industry standard of ±1ms.

[0037] In this embodiment, as Figure 5 As shown, the adaptive wireless communication module includes: The multi-mode communication chip adopts an integrated chip that supports multi-band radio frequency transceivers and baseband processors. It has an integrated hardware acceleration module for directly processing physical layer signaling parsing and writing the parsed channel quality index (CQI) into the register of the intelligent handover control unit in real time through shared memory or direct memory access mechanism. The environmental sensing unit, including a broadband spectrum scanning sub-circuit and an auxiliary sensor interface, is connected to the intelligent switching control unit via an I2C or SPI high-speed serial bus. It is used to collect electromagnetic noise floor noise, tunnel vibration characteristics, and temperature and humidity data of the underground environment. The intelligent switching control unit is implemented based on a field-programmable gate array or a high-performance ARM Cortex series microcontroller. It communicates bidirectionally with the application processor (AP) side of the multi-mode communication chip through a PCIe or UART interface, thus constructing a transmission channel that separates the control plane and the data plane. The power management and isolation unit, including a DC-DC buck converter circuit and a magnetic isolation chip, is used to convert the wide downhole voltage input (9V-36V) into the stable operating voltage required by each chip and to achieve electrical isolation between the control logic ground and the power ground. The multi-mode communication chip, environmental sensing unit, and intelligent switching control unit are integrated and packaged in the same metal shield. The shield is made of nickel-plated copper alloy or stainless steel, and an LC filter circuit and transient voltage suppression diode (TVS) are set at the radio frequency signal input end to filter out pulse interference generated by downhole electromechanical equipment.

[0038] In this embodiment, as Figure 6 As shown, the intelligent switching control unit includes: The environmental parameter acquisition subunit is used to acquire physical layer parameters and network layer parameters of the downhole communication environment in real time. The parameters include, but are not limited to, the received signal strength indication (RSSI), signal-to-noise ratio (SNR), reference signal reception quality (RSRQ), channel quality indication (CQI), network load rate, transmission delay, and packet loss rate for each network standard. The business requirements analysis subunit is used to analyze the business characteristics of the data to be transmitted and determine the corresponding business QoS (Quality of Service) requirement thresholds, including minimum bandwidth requirements, maximum allowable latency, and reliability level. The multi-dimensional decision subunit is used to match and calculate the collected environmental parameters with the service QoS requirements based on a preset weighted scoring algorithm or fuzzy logic model, and generate a comprehensive score for each candidate network standard; among them, the weighted scoring algorithm pre-configures weight coefficients for the signal attenuation characteristics of different network standards according to the underground roadway topology.

[0039] Furthermore, the business requirements analysis sub-unit includes: The packet feature extraction module is used to scan the IP packets of the data to be transmitted in real time and extract the five-tuple information (source IP, destination IP, source port, destination port, protocol type) and application layer payload features of the packets. The business feature matching module is used to compare the extracted features with the pre-set business feature library to identify the specific type of the current business. The business types include at least video surveillance stream, gas concentration monitoring data, equipment control signaling, voice call and file transfer. The QoS parameter mapping module is used to query the corresponding QoS policy table based on the identified service type and dynamically determine the service quality requirement threshold. The QoS policy table predefines priority queues, minimum guaranteed bandwidth, maximum retransmission count, and latency jitter tolerance for different service types. For example, gas concentration monitoring data is mapped to the highest reliability level (URLLC) and low bandwidth requirements, while high-definition video streams are mapped to high bandwidth requirements and medium reliability level (eMBB).

[0040] In this embodiment, the multidimensional decision-making subunit performs the following weighted scoring calculation process: The parameter normalization process maps the collected physical layer parameters (RSSI, SNR) and network layer parameters (latency, packet loss rate, bandwidth) of each network standard to dimensionless evaluation values ​​in the range [0, 1]. Among them, negative indicators (such as latency and packet loss rate) are normalized by reciprocal method or linear interpolation method. The dynamic weight vector is determined by calling the preset underground topology weight table and determining the weight coefficients of each evaluation dimension based on the current roadway type (straight roadway, bifurcation, mining face) and movement speed of the module. W =[ w 1, w 2,..., w n ]; In long, straight tunnels, the signal coverage weight of 5G / 4G is w cov Higher than WiFi; at junctions or areas with dense devices, network load balancing weight w load Upgrade to the highest level; The service matching degree calculation involves performing Euclidean distance calculation or cosine similarity matching between the normalized network parameter matrix and the service QoS requirement vector to generate a service matching factor. α ; A comprehensive score is generated using a formula. Score i =∑( w j × P ij )× α Calculate the first i The overall score of each candidate network standard, among which P ij For the first i The network in the first j Normalized values ​​in each dimension; Among them, the comprehensive score is selected. Score The network type that is the largest and greater than the preset switching threshold (e.g., 60 points) is the current optimal network.

[0041] Alternatively, the multidimensional decision-making subunits are implemented based on a type-two fuzzy logic model, specifically including: The fuzzification interface is used to convert environmental parameters (RSSI, latency, bandwidth) into fuzzy linguistic variables, define fuzzy sets as {very poor (VB), poor (B), medium (M), good (G), very good (VG)}, and establish corresponding trapezoidal or Gaussian membership functions; The fuzzy rule base pre-stores IF-THEN inference rules built based on the experience of downhole experts; for example: "IF roadway type is 'mining face' AND service type is 'control signaling' AND 5G signal is 'medium' AND WiFi load is 'high' THEN 5G priority is 'high'"; "IF roadway type is 'blind spot' AND service type is 'video' AND LTE Cat 1 signal is 'good' THEN LTE Cat 1 priority is the highest"; The fuzzy inference engine uses the Mamdani or Sugeno inference mechanism. Based on the current input fuzzy variables, it activates multiple rules in the fuzzy rule base and performs logical synthesis to obtain the fuzzy output set of each candidate network. The defuzzing module uses the centroid method or the maximum membership method to convert the fuzzy output set into precise priority values, and sorts the candidate networks in descending order according to the magnitude of these values, selecting the network ranked first as the optimal network.

[0042] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0043] The above embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of protection of the present invention. Any non-substantial changes and substitutions made by those skilled in the art based on the present invention shall fall within the scope of protection claimed by the present invention.

Claims

1. A real-time monitoring system for underground seismic data in coal mines, characterized in that, include: MEMS fully digital three-component geophone is used to acquire seismic wave signals in three-dimensional space underground in coal mines in real time. The node acquisition unit is connected to the MEMS all-digital three-component geophone and is used for real-time reception, local storage and preliminary processing of seismic wave data. The intrinsically safe power supply module adopts an intrinsically safe circuit design that complies with GB 3836.1-2010 and GB 3836.4-2010 standards to provide a safe power supply for node devices; The high-precision time synchronization module adopts dual-crystal high-precision clock synchronization technology that integrates AI temperature compensation, and monitors and compensates for frequency drift of the crystal oscillator caused by temperature changes in real time. The adaptive wireless communication module automatically selects the optimal network for real-time data transmission based on the actual underground communication environment. Specifically, it matches and calculates the collected environmental parameters with the service QoS requirements to generate a comprehensive score for each candidate network standard. The network standard with the highest comprehensive score is then determined as the current optimal network, and the multi-mode communication chip is controlled to access this network for data transmission.

2. The system according to claim 1, characterized in that, The adaptive wireless communication module includes: The multi-mode communication chip adopts an integrated chip that supports multi-band radio frequency transceivers and baseband processors. It has an integrated hardware acceleration module for directly processing physical layer signaling parsing and writing the parsed channel quality index (CQI) into the register of the intelligent handover control unit in real time through shared memory or direct memory access mechanism. The environmental sensing unit, including a broadband spectrum scanning sub-circuit and an auxiliary sensor interface, is connected to the intelligent switching control unit via an I2C or SPI high-speed serial bus. It is used to collect electromagnetic noise floor noise, tunnel vibration characteristics, and temperature and humidity data of the underground environment. The intelligent switching control unit is implemented based on a field-programmable gate array or a high-performance ARM Cortex series microcontroller. It communicates bidirectionally with the application processor (AP) side of the multi-mode communication chip through a PCIe or UART interface, thus constructing a transmission channel that separates the control plane and the data plane. The power management and isolation unit, including a DC-DC buck converter circuit and a magnetic isolation chip, is used to convert the wide downhole voltage input into the stable operating voltage required by each chip and to achieve electrical isolation between the control logic ground and the power ground. The multi-mode communication chip, environmental sensing unit, and intelligent switching control unit are integrated and packaged in the same metal shield. The shield is made of nickel-plated copper alloy or stainless steel, and an LC filter circuit and transient voltage suppression diode are set at the radio frequency signal input end to filter out pulse interference generated by downhole electromechanical equipment.

3. The system according to claim 1, characterized in that, The intelligent switching control unit includes: The environmental parameter acquisition subunit is used to acquire physical layer parameters and network layer parameters of the downhole communication environment in real time. The network layer parameters include at least the received signal strength indication, signal-to-noise ratio, reference signal reception quality, channel quality indication, network load rate, transmission delay and packet loss rate for each network standard. The business requirements analysis subunit is used to analyze the business characteristics of the data to be transmitted and determine the corresponding business QoS requirement thresholds, including minimum bandwidth requirements, maximum allowable latency and reliability level. The multi-dimensional decision subunit is used to match and calculate the collected environmental parameters with the service QoS requirements based on a preset weighted scoring algorithm or fuzzy logic model, and generate a comprehensive score for each candidate network standard; among them, the weighted scoring algorithm pre-configures weight coefficients for the signal attenuation characteristics of different network standards according to the underground roadway topology.

4. The system according to claim 3, characterized in that, The business requirements analysis sub-unit includes: The packet feature extraction module is used to scan the IP packets of the data to be transmitted in real time and extract the five-tuple information and application layer payload features of the packets. The business feature matching module is used to compare the extracted features with the pre-set business feature library to identify the specific type of the current business. The business types include at least video surveillance stream, gas concentration monitoring data, equipment control signaling, voice call and file transfer. The QoS parameter mapping module is used to query the corresponding QoS policy table based on the identified service type and dynamically determine the service quality requirement threshold. The QoS policy table predefines priority queues, minimum guaranteed bandwidth, maximum retransmission count, and latency jitter tolerance for different service types.

5. The system according to claim 3, characterized in that, The multidimensional decision-making subunit performs the following weighted scoring calculation process: The parameter normalization process maps the collected physical layer parameters and network layer parameters of each network type to dimensionless evaluation values ​​in the interval [0, 1]. For negative indices, the reciprocal method or linear interpolation method is used for normalization. The dynamic weight vector is determined by calling the preset underground topology weight table and determining the weight coefficients of each evaluation dimension based on the current roadway type and movement speed of the module. W =[ w 1, w 2,..., w n ]; In long, straight tunnels, the signal coverage weight of 5G / 4G is w cov Higher than WiFi; at junctions or areas with dense devices, network load balancing weight w load Upgrade to the highest level; The service matching degree calculation involves performing Euclidean distance calculation or cosine similarity matching between the normalized network parameter matrix and the service QoS requirement vector to generate a service matching factor. α ; A comprehensive score is generated using a formula. Score i =∑( w j × P ij )× α Calculate the first i The overall score of each candidate network standard, among which P ij For the first i The network in the first j Normalized values ​​in each dimension; Among them, the comprehensive score is selected. Score The network type that is the largest and greater than the preset switching threshold is selected as the current optimal network.

6. The system according to claim 3, characterized in that, The multidimensional decision-making subunit is implemented based on a type-two fuzzy logic model, specifically including: The fuzzification interface is used to convert environmental parameters into fuzzy linguistic variables, define fuzzy sets as {very poor (VB), poor (B), medium (M), good (G), very good (VG)}, and establish corresponding trapezoidal or Gaussian membership functions; A fuzzy rule base, pre-stored with IF-THEN inference rules built based on downhole expert experience; The fuzzy inference engine uses the Mamdani or Sugeno inference mechanism. Based on the current input fuzzy variables, it activates multiple rules in the fuzzy rule base and performs logical synthesis to obtain the fuzzy output set of each candidate network. The defuzzing module uses the centroid method or the maximum membership method to convert the fuzzy output set into priority values, and sorts the candidate networks in descending order according to the size of the priority values, selecting the network with the highest ranking as the optimal network.

7. The system according to claim 2, characterized in that, The MEMS all-digital three-component detector adopts a single-chip system-in-package structure, specifically including: The MEMS sensor uses deep reactive ion etching to fabricate three orthogonally arranged micromechanical cantilever beam mass block structures on the same silicon wafer, corresponding to vibration sensing along the X, Y, and Z axes, respectively. The capacitance detection array, located below the micromechanical cantilever beam mass block structure, is used to detect the differential capacitance change caused by the displacement of the mass block and convert the mechanical displacement signal into an analog charge signal. The digital conversion ASIC circuit is flip-chip bonded to the MEMS sensor and integrates a low-noise charge amplifier, a high-order Sigma-Delta analog-to-digital converter, and a digital filtering unit. Among them, the digital conversion ASIC circuit directly converts the analog charge signal into a 24-bit digital signal inside the chip, and outputs it directly to the intelligent switching control unit through the I2S or PDM digital audio interface. The temperature compensation module, integrated into the ASIC circuit, is used to correct the zero-point drift and sensitivity coefficient of the MEMS sensor in real time according to the downhole ambient temperature.

8. The system according to any one of claims 1 to 7, characterized in that, The node acquisition unit includes: The dual-level storage architecture includes a volatile cache and a non-volatile flash array. During normal operation, the real-time received seismic wave data is first stored in the cache. When the cache level reaches a preset threshold or a network interruption signal is detected, a disk flushing operation is automatically triggered to write the data blocks to the flash array. The power failure protection circuit is connected to the main power supply terminal of the node acquisition unit and includes a supercapacitor or backup lithium battery. It is used to maintain the operation of the storage control circuit when the main power supply is unexpectedly disconnected. The log index file system generates an index header file with a high-precision timestamp for each segment of stored data, recording the acquisition time, detector ID, and data length; When the network connection is restored, the node acquisition unit automatically calculates the data segments that have not been uploaded based on the index file, prioritizes the uploading of historical data in chronological order, and then transmits real-time data to avoid data out of order.

9. The system according to claim 2, characterized in that, Intrinsically safe power modules include: Intrinsically safe conversion circuit, used to convert external input power supply to intrinsically safe output, includes voltage clamping circuit and current limiting circuit; The safety parameter sampling unit acquires the output voltage of the intrinsically safe side in real time through a high-precision ADC. V out and output current I out And the surface temperature of key components; The energy integration calculation module calculates the cumulative released energy of the circuit in real time based on sampled data. E =∫ V ( t ) I ( t ) dt When the calculated cumulative energy approaches 80% of the minimum ignition energy specified in GB 3836.4 standard, a first-level warning signal is generated. The dynamic power limiting logic module, connected to the intelligent switching control unit, automatically reduces the output voltage or cuts off power to non-critical loads when it receives a high power consumption warning or detects an abnormal internal temperature, retaining only the minimum power supply to the MEMS detector and the main control chip.

10. The system according to any one of claims 1 to 7, characterized in that, The high-precision time synchronization module includes: The dual crystal clock source consists of a high-stability temperature-controlled crystal oscillator (OCXO) and a low-power temperature-compensated crystal oscillator (TCXO) connected in parallel. The OCXO serves as the master clock source, providing a high-precision reference frequency, while the TCXO serves as the slave clock source, used for timekeeping during low-power standby or OCXO warm-up. A multi-dimensional sensing unit is used to collect temperature gradient data on the surface of the crystal oscillator housing and ripple noise of the power supply voltage in real time. The AI ​​thermal compensation engine is deployed with a frequency drift prediction model based on a BP neural network or a long short-term memory network; the input layer of this model includes the current temperature. T Temperature change rate dT / dt and historical frequency deviation Δ f hist The output layer is the predicted correction value Δ for the crystal oscillator frequency. f pred The model training process incorporates the thermal conductivity differential equation of the crystal oscillator as a physical constraint. The frequency locking and calibration unit, based on the Δ output of the AI ​​thermal compensation engine, f pred The output frequency of the OCXO can be finely adjusted in real time using a numerically controlled oscillator or a phase-locked loop; When an OCXO fault or excessive power consumption is detected, the system seamlessly switches to a TCXO and uses the A thermal compensation engine to perform the same temperature compensation on the TCXO.

Images