Remote intelligent seismograph system based on star network communication and working method thereof

Abstract

The invention discloses a remote intelligent seismograph system based on star network communication and a working method thereof, and belongs to the field of seismic exploration. The remote intelligent seismograph system comprises a node seismograph module which is in two-way communication with a remote management module through a star network communication unit; the node seismograph module, the star network communication unit and the remote management module are electrically connected with the multi-source energy module; wherein the node seismograph module is used for collecting seismic wave vibration signals and preprocessing the collected seismic wave vibration signals; the star network communication unit is used for receiving the pre-processed seismic wave vibration signal and uploading the pre-processed seismic wave vibration signal to the remote management module; and the remote management module is used for receiving the seismic wave vibration signal so as to output a three-dimensional fault model and a health monitoring report. By adopting the remote intelligent seismograph system based on star network communication and the working method thereof, remote downloading and remote monitoring of data are realized, and the efficiency and reliability of seismic data acquisition and management are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of seismic exploration, and particularly to a remote intelligent seismograph system based on satellite network communication and its working method. Background Art

[0002] In the field of seismic exploration, seismic node seismographs are widely used for collecting and analyzing seismic wave signals. However, traditional seismic node seismographs have the following problems, which limit their efficiency and reliability:

[0003] 1. Backward data acquisition method: When traditional seismic node seismographs complete the acquisition task or reach the data storage limit, it is necessary to manually retrieve the equipment and download seismic wave data or write parameters through wired transmission. This method not only relies on frequent manual operations, increasing the workload and maintenance cost, but also may cause data loss or interruption due to untimely maintenance, seriously affecting the operation efficiency and data integrity.

[0004] 2. Signal problems: In remote areas, traditional network transmission methods such as 4G often face problems of unstable network or even no signal, making real-time data transmission unreliable. This further exacerbates the difficulty of data acquisition. Especially in emergency situations, it may lead to delays or omissions of important information.

[0005] 3. Power supply difficulties: Existing seismic node seismographs mostly rely on battery power supply. After the battery runs out, it is necessary to manually retrieve the equipment for charging. Especially in remote areas or unattended environments, there is a lack of reliable power supply, and the continuity and stability of equipment operation are difficult to guarantee. Even by adding external power supply equipment such as generators, it will significantly increase the complexity and operation and maintenance cost of the system.

[0006] 4. High dependence on manual labor: The existing technology highly depends on manual intervention, including equipment parameter adjustment, data download, and power supply maintenance. This dependence not only increases the labor cost and operation difficulty, but also reduces the efficiency and safety of maintenance work in remote and harsh environments. Especially in these special environments, the work risk of maintenance personnel is high and the efficiency is low, affecting the overall operation effect.

[0007] In summary, traditional seismic node seismographs have obvious deficiencies in data acquisition, signal transmission, power supply guarantee, and manual dependence. Summary of the Invention

[0008] The purpose of the present invention is to provide a remote intelligent seismograph system based on satellite network communication and its working method to solve the above technical problems.

[0009] To achieve the above object, the present invention provides a remote intelligent seismograph system based on satellite network communication, including a node seismograph module. The node seismograph module communicates bidirectionally with a remote management module via a satellite network communication unit, and the node seismograph module, the satellite network communication unit, and the remote management module are all electrically connected to a multi-source energy module;

[0010] Among them, the node seismograph module includes a connected three-axis MEMS seismic wave sensor. The three-axis MEMS seismic wave sensor is successively connected to an edge intelligent preprocessing unit via an amplification and filtering unit and an analog-to-digital conversion unit, and is used to collect seismic wave vibration signals by using the three-axis MEMS seismic wave sensor, and preprocess the collected seismic wave vibration signals by using the edge intelligent preprocessing unit;

[0011] The satellite network communication unit includes a modem, a communication protocol processing unit, and a phased array antenna connected in sequence. The phased array antenna is a three-mode link structure composed of an FSO communication unit, an adaptive radio frequency unit, and a Beidou RDSS link, and is used to receive the preprocessed seismic wave vibration signals and upload them to the remote management module, or transmit them to the node seismograph module after receiving the instructions from the remote management module;

[0012] The remote management module is used to receive seismic wave vibration signals, and at the same time remotely monitor and manage the node seismograph module to output a three-dimensional fault model and a health monitoring report.

[0013] Preferably, the multi-source energy module includes a solar power generation unit and a vertical axis wind power generation unit. The solar power generation unit and the vertical axis wind power generation unit are both electrically connected to a lithium iron phosphate battery pack via a battery management unit.

[0014] Preferably, the three-axis MEMS seismic wave sensor integrates a seismic wave sensor, an amplification and filtering unit, an analog-to-digital conversion unit, and a microprocessor connected in sequence. The microprocessor is also respectively connected to the edge intelligent preprocessing unit and the communication unit. At the same time, the edge intelligent preprocessing unit is connected to the communication unit, and the communication unit communicates with the satellite network communication unit, so as to collect seismic wave vibration signals by using the three-axis MEMS seismic wave sensor, transmit them to the amplification and filtering unit for filtering and signal amplification, and then after analog-to-digital conversion by using an analog-to-digital converter, transmit them to the microprocessor, and after data compression based on wavelet transform by the edge intelligent preprocessing unit, upload the collected seismic data by using the communication unit.

[0015] Preferably, the uplink of the FSO communication unit uses a 1550nm wavelength laser modulation, and combines the wavelength-mode division multiplexing modulation of the compressed data, with a transmission rate ≥ 20Gbps and a bit error probability ≤ 5×10 -9 ;

[0016] The adaptive radio frequency unit operates in the frequency band of 1.4 GHz - 5.8 GHz, supports OFDM modulation and polarization diversity, and has a transmission rate ≥ 50 Mbps when the rain attenuation ≤ 30 dB;

[0017] The RDSS link integrates the Beidou-3 B1I + B2a dual-frequency reception, and the end-to-end delay ≤ 2.5 s in the emergency mode.

[0018] A working method of a remote intelligent seismograph system based on satellite network communication includes the following steps:

[0019] S1. The three-axis MEMS seismic wave sensor collects vibration signals at a set sampling rate. After amplification, filtering, and analog-to-digital conversion, the edge intelligent preprocessing unit performs wavelet denoising and data compression to generate a feature data set;

[0020] S2. Determine the communication transmission link, and upload the feature data set to the remote management module based on the determined communication transmission link;

[0021] S3. Based on the feature data set, use the remote management module to generate a three-dimensional tomography model and output it.

[0022] Preferably, in step S1, the specific steps of performing wavelet denoising and data compression are as follows:

[0023] The first step: Use the Daubechies9 orthogonal wavelet basis to perform discrete wavelet decomposition on the digital signal after analog-to-digital conversion to obtain the approximate coefficients and detail coefficients of each layer, and the decomposition formula is as follows:

[0024] A j+1 [k] = ∑ n h[n] × A j [2k - n];

[0025] D j+1 [k] = ∑ n g[n] × A j [2k - n];

[0026] In the formula, A j+1 and A j respectively represent the approximate coefficients of the (j + 1)-th layer and the j-th layer; n represents the decomposition layer number; h[n] represents the low-pass filter; D j+1 represents the detail coefficient of the (j + 1)-th layer; [2k - n] represents the downsampling operation;

[0027] The second step: Threshold denoising, and set the threshold λ j of the j-th layer to be:

[0028]

[0029] In the formula, σ jDenote the noise standard deviation of the detail coefficient of the j-th layer; N j Denote the length of the detail coefficient of the j-th layer;

[0030] Step 3, based on the threshold λ j Correct the wavelet coefficient W j,k :

[0031]

[0032] wherein, Denote the corrected wavelet coefficient;

[0033] Step 4, quantize the wavelet coefficients by using a hierarchical compression strategy, wherein 5% of the maximum amplitude coefficients are reserved for high-frequency noise and 95% of the amplitude coefficients are reserved for low-frequency noise;

[0034] Step 5, perform Huffman coding on the reserved coefficients to achieve compression.

[0035] Preferably, step S2 specifically includes the following steps:

[0036] S21, screen the candidate link set, and the screening conditions are:

[0037]

[0038] wherein, B candidate Denote the candidate link bandwidth; S data Denote the size of the feature data set; Δt max Denote the maximum allowable transmission time;

[0039] S22, evaluate the channel quality by using the signal-to-noise ratio SNR and the bit error rate BER:

[0040]

[0041] wherein, P signal Denote the signal power; P noise Denote the noise power; N error Denote the number of bit errors; N total Denote the complete transmission code;

[0042] S23, calculate the transmission performance scores Score of the FSO communication unit, the adaptive radio frequency unit, and the Beidou RDSS link respectively based on the comprehensive weight scoring model:

[0043]

[0044] wherein, w1, w2, and w3 denote the weights of bandwidth, reliability, and delay respectively; B max Denote the maximum bandwidth of the candidate link; Latency denotes the link transmission delay;

[0045] S24. Sort the transmission performance scores Score of the FSO communication unit, the adaptive radio frequency unit, and the Beidou RDSS link, and select the link corresponding to the maximum transmission performance score Score as the optimal link;

[0046] S25. Modulate the received seismic wave vibration signal using a modem, and then encrypt and packetize the modulated signal using the communication protocol processing unit, and upload it to the remote management module using the determined optimal link;

[0047] Or receive the remote management module instruction using the optimal link, demodulate it using a modem, and then encrypt and packetize the modulated signal using the communication protocol processing unit and transmit it to the node seismograph module.

[0048] Preferably, step S3 specifically includes the following steps:

[0049] S31. Perform time-frequency domain normalization on the original signal in the received feature dataset:

[0050]

[0051] Among them,

[0052]

[0053] In the formula, χ norm represents the normalized data; x represents the original signal; μ represents the signal mean; σ represents the standard deviation; N represents the total number of samples; x i represents the i-th data sample;

[0054] S32. Extract tomographic multi-scale features through the MultiRes-Unet3D neural network:

[0055]

[0056] Among them,

[0057]

[0058] In the formula, L weighted represents the weighted loss function; C represents the total number of categories; W c represents the weight of category c; y c represents the true label; p c represents the probability of being predicted as true; f c represents the frequency of category c appearing in the feature dataset; ∈ represents a constant;

[0059] S33. Generate a tomographic point cloud based on the stereo vision method, and calculate the depth information using multi-view seismic data:

[0060]

[0061] Among them,

[0062] d = argmin d ∑ p∈Ω ∣I left (p) - I right (p + d)∣ 2 ;

[0063] In the formula, Z represents the depth of the fault point; B represents the binocular baseline distance; f represents the focal length; d represents the parallax; I left (p) and I right (p + d) respectively represent the intensity values at point p in the left view and the right view; Ω represents the parallax search range;

[0064] S34. Align the fault point cloud using the Iterative Closest Point (ICP) algorithm, where the objective function E of the ICP algorithm ICP is as follows:

[0065]

[0066] Among them, p i represents the source point cloud; q i represents the target point cloud; T represents the rigid body transformation matrix; I represents the number of points in the fault point cloud;

[0067] S35. Convert the optimized point cloud into a meshed surface model to obtain a three-dimensional fault model.

[0068] Therefore, the present invention adopts the above-mentioned remote intelligent seismograph system based on star network communication and its working method, and the beneficial effects are as follows:

[0069] 1. Real-time remote monitoring and management: (1) Instantaneity: The star network communication unit enables the node seismograph module to communicate with the remote management module in real time, ensuring the instant transmission of seismic wave vibration signals, which greatly shortens the time from data acquisition to processing and improves the emergency response speed; (2) Remote control: The remote management module not only receives seismic wave vibration signals but also can remotely configure parameters and monitor the status of the node seismograph module, reducing the need for on-site maintenance, lowering the operation and maintenance costs. At the same time, users do not need to go to the site to recover equipment or manually download data, thus significantly reducing the need for manual intervention and related workload, and avoiding data lag and loss problems caused by manual intervention;

[0070] 2. High reliability and stability: (1) Star network communication: Stable data transmission is achieved by using phased array antennas and modems, enabling real-time transmission of seismic data globally. It is especially suitable for remote areas such as mountainous regions. Even in environments without network coverage, data can be stably uploaded, greatly improving the timeliness and accuracy of seismic monitoring. It has higher reliability and anti-interference ability compared to traditional network methods such as 4G; (2) Power supply guarantee: The multi-source energy module ensures that the device can continue to operate in remote areas lacking power supply. The design of the battery management unit and energy storage battery enables the system to maintain stable power supply under different lighting conditions, avoiding data interruption caused by power shortage;

[0071] 3. Intelligence and automation: (1) Automatic preprocessing: The built-in amplification and filtering unit and analog-to-digital conversion unit in the node seismograph module can automatically complete signal filtering, amplification, and conversion, reducing the burden of subsequent data analysis; (2) Intelligent analysis: The data processing unit in the remote management module can not only store and analyze the received seismic wave vibration signals but also generate a health monitoring report to help users quickly understand the seismic activity situation;

[0072] 4. Multi-dimensional information feedback: In addition to seismic wave vibration signals, the system can also provide real-time feedback on the operating status information of the device (such as battery power, storage space, and communication quality), helping managers promptly discover potential problems and take measures to prevent monitoring interruption caused by device loss, damage, or operational failure;

[0073] 5. By adopting a multi-source energy module, energy self-sufficiency is achieved, eliminating the need for frequent battery charging. Even in low-light conditions, the energy storage battery can ensure the continuous operation of the device, reducing the maintenance frequency and operating costs, thereby realizing long-term stable seismic monitoring.

[0074] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Description of the Drawings

[0075] Figure 1 It is a structural block diagram of a remote intelligent seismograph system based on star network communication of the present invention. Detailed Embodiments

[0076] In order to make the objectives, technical solutions, and advantages of the embodiments disclosed in the present invention clearer and more understandable, the following further details the embodiments of the present invention in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the embodiments of the present invention and are not used to limit the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of this application. The examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals represent the same or similar elements or elements with the same or similar functions from beginning to end.

[0077] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0078] The following details the embodiments of the present invention in conjunction with the accompanying drawings.

[0079] As Figure 1 shown, a remote intelligent seismograph system based on star network communication includes a node seismograph module. The node seismograph module communicates bidirectionally with a remote management module via a star network communication unit, and the node seismograph module, the star network communication unit, and the remote management module are all electrically connected to a multi-source energy module. Among them, the node seismograph module includes a connected three-axis MEMS seismic wave sensor. The three-axis MEMS seismic wave sensor is sequentially connected to an edge intelligent preprocessing unit via an amplification and filtering unit and an analog-to-digital conversion unit, and is used to collect seismic wave vibration signals using the three-axis MEMS seismic wave sensor and preprocess the collected seismic wave vibration signals using the edge intelligent preprocessing unit. The star network communication unit includes a modulator-demodulator, a communication protocol processing unit, and a phased array antenna connected in sequence. The phased array antenna is a three-mode link structure composed of an FSO communication unit, an adaptive radio frequency unit, and a Beidou RDSS link, and is used to receive the preprocessed seismic wave vibration signals and upload them to the remote management module, or transmit them to the node seismograph module after receiving instructions from the remote management module. The remote management module is used to receive seismic wave vibration signals and simultaneously remotely monitor and manage the node seismograph module to output a three-dimensional fault model and a health monitoring report.

[0080] The multi-source energy module includes a solar power generation unit and a vertical axis wind power generation unit. The solar power generation unit and the vertical axis wind power generation unit are both electrically connected to a lithium iron phosphate battery pack via a battery management unit.

[0081] The three-axis MEMS seismic wave sensor integrates a seismic wave sensor, an amplification and filtering unit, an analog-to-digital conversion unit, and a microprocessor that are electrically connected in sequence. The microprocessor is also respectively connected to an edge intelligent preprocessing unit and a communication unit. At the same time, the edge intelligent preprocessing unit is connected to the communication unit, and the communication unit communicates with the star network communication unit to realize the acquisition of seismic wave vibration signals using the three-axis MEMS seismic wave sensor, and transmit them to the amplification and filtering unit for filtering and signal amplification. After analog-to-digital conversion using the analog-to-digital converter, the signals are transmitted to the microprocessor, and after data compression based on wavelet transform by the edge intelligent preprocessing unit, the acquired seismic data is uploaded using the communication unit.

[0082] The uplink of the FSO communication unit uses a 1550nm wavelength laser modulation and combines the wavelength-mode division multiplexing modulation of the compressed data, with a transmission rate ≥ 20Gbps and a bit error probability ≤ 5×10 -9 ; The adaptive radio frequency unit operates in the 1.4GHz - 5.8GHz frequency band, supports OFDM modulation and polarization diversity, and has a transmission rate ≥ 50Mbps when the rain attenuation ≤ 30dB; The RDSS link integrates the Beidou-3 B1I + B2a dual-frequency reception, and the end-to-end delay ≤ 2.5s in the emergency mode.

[0083] It should be noted that the above electronic components are all mature products on the market. In this embodiment, only procurement and connection according to the instruction manual are required, and no improvement has been made to them. Therefore, the circuit connection structure and principle will not be elaborated here.

[0084] A working method of a remote intelligent seismograph system based on star network communication includes the following steps:

[0085] S1. The three-axis MEMS seismic wave sensor acquires vibration signals at a set sampling rate. After amplification, filtering, and analog-to-digital conversion, the edge intelligent preprocessing unit performs wavelet denoising and data compression to generate a feature data set;

[0086] In step S1, the specific steps of performing wavelet denoising and data compression are as follows:

[0087] The first step: Using the Daubechies9 orthogonal wavelet basis, perform discrete wavelet decomposition on the digital signal after analog-to-digital conversion to obtain the approximate coefficients and detail coefficients of each layer, and the decomposition formula is as follows:

[0088] A j+1 [k] = ∑ n h[n] × A j [2k - n];

[0089] D j+1 [k] = ∑ n g[n] × A j [2k - n];

[0090] Wherein, A j+1 and A j respectively represent the approximation coefficients of the (j + 1)-th layer and the j-th layer; n represents the decomposition level; h[n] represents the low-pass filter; D j+1 represents the detail coefficients of the (j + 1)-th layer; [2k - n] represents the downsampling operation;

[0091] Step 2: Threshold denoising, and set the threshold λ of the j-th layer j as:

[0092]

[0093] Wherein, σ j represents the noise standard deviation of the detail coefficients of the j-th layer; N j represents the length of the detail coefficients of the j-th layer;

[0094] Step 3: Based on the threshold λ j correct the wavelet coefficients W j,k :

[0095]

[0096] Wherein, represents the corrected wavelet coefficients;

[0097] Step 4: Quantize the wavelet coefficients by using a hierarchical compression strategy, wherein 5% of the maximum amplitude coefficients are reserved for high-frequency noise, and 95% of the amplitude coefficients are reserved for low-frequency noise;

[0098] Step 5: Perform Huffman coding on the reserved coefficients to achieve compression.

[0099] S2. Determine the communication transmission link, and upload the feature data set to the remote management module based on the determined communication transmission link;

[0100] Step S2 specifically includes the following steps:

[0101] S21. Screen the link candidate set, and the screening conditions are:

[0102]

[0103] Wherein, B candidate represents the candidate link bandwidth; S data represents the size of the feature data set; Δt max represents the maximum allowable transmission time;

[0104] S22. Evaluate the channel quality by using the signal-to-noise ratio SNR and the bit error rate BER:

[0105]

[0106] Wherein, P signal represents the signal power; P noise represents the noise power; N error represents the number of error codes; N total represents the complete transmission code;

[0107] S23. Calculate the transmission performance scores Score of the FSO communication unit, the adaptive radio frequency unit, and the Beidou RDSS link respectively based on the comprehensive weight scoring model:

[0108]

[0109] Wherein, w1, w2, and w3 respectively represent the weights of bandwidth, reliability, and delay; B max represents the maximum bandwidth of the candidate link; Latency represents the link transmission delay;

[0110] S24. Sort the transmission performance scores Score of the FSO communication unit, the adaptive radio frequency unit, and the Beidou RDSS link, and take the link corresponding to the maximum transmission performance score Score as the optimal link;

[0111] S25. Modulate the received seismic wave vibration signal by using a modem, and then encrypt and packetize the modulated signal by using a communication protocol processing unit, and upload it to the remote management module by using the determined optimal link;

[0112] Or receive the remote management module instruction by using the optimal link, demodulate it by using a modem, and then encrypt and packetize the modulated signal by using a communication protocol processing unit and transmit it to the node seismograph module.

[0113] S3. Based on the feature dataset, generate a three-dimensional fault model by using the remote management module and output it.

[0114] Step S3 specifically includes the following steps:

[0115] S31. Perform time-frequency domain normalization on the original signal in the received feature dataset:

[0116]

[0117] Among them,

[0118]

[0119] Wherein, χ norm represents the normalized data; x represents the original signal; μ represents the signal mean; σ represents the standard deviation; N represents the total number of samples; x i represents the i-th data sample;

[0120] S32. Extract tomographic multi-scale features through the MultiRes-Unet3D neural network:

[0121]

[0122] Among them,

[0123]

[0124] In the formula, L weighted represents the weighted loss function; C represents the total number of categories; W c represents the weight of category c; y c represents the true label; p c represents the probability of being predicted as true; f c represents the frequency of category c appearing in the feature dataset; ∈ represents a constant;

[0125] S33. Generate tomographic point clouds based on the stereo vision method and calculate depth information using multi-view seismic data:

[0126]

[0127] Among them,

[0128] d = argmin d ∑ p∈Ω ∣I left (p) - I right (p + d)∣ 2 ;

[0129] In the formula, Z represents the tomographic point depth; B represents the binocular baseline distance; f represents the focal length; d represents the disparity; I left (p) and I right (p + d) represent the intensity values of the left view and the right view at point p respectively; Ω represents the disparity search range;

[0130] S34. Align the tomographic point clouds using the Iterative Closest Point (ICP) algorithm, where the objective function E ICP of the ICP algorithm is as follows:

[0131]

[0132] Among them, p i represents the source point cloud; q i represents the target point cloud; T represents the rigid body transformation matrix; I represents the number of points in the tomographic point cloud;

[0133] S35. Convert the optimized point cloud into a meshed surface model to obtain a three-dimensional tomographic model.

[0134] Embodiment

[0135] Step 1: Seismic wave data acquisition and preprocessing:

[0136] When conducting exploration work in a seismically active area, node seismograph modules are deployed at multiple key positions (the node seismograph module includes triaxial MEMS seismic wave sensors arranged in a multi-point array, which can accurately capture seismic wave vibration signals from different directions). When the multi-point array triaxial MEMS seismic wave sensors detect seismic waves and the vibration signal intensity reaches the preset threshold, the built-in microprocessor immediately starts the acquisition process and transmits the signal to the amplification and filtering unit.

[0137] In the signal amplification link, the adaptive gain control (AGC) technology is used. For example, when a weak seismic wave signal is detected, the system automatically increases the amplification factor to ensure that the signal can be clearly presented; when encountering a strong earthquake signal, the amplification factor is reasonably adjusted to prevent signal overload and distortion. In the filtering stage, an adaptive Kalman filtering algorithm combined with a band-pass filter is adopted. Through the dynamic correction of the measured signal by Kalman filtering, environmental noise and irrelevant interference are effectively removed. At the same time, the band-pass filter is used to allow seismic wave signals in the main frequency range to pass through smoothly, filter out low-frequency and high-frequency noise, and then transmit them to the analog-to-digital conversion unit.

[0138] The analog-to-digital conversion unit uses a 24-bit high-precision successive approximation register (SAR) analog-to-digital converter to accurately convert the analog signal into a digital signal. The converted digital signal is further preprocessed by the edge intelligent preprocessing unit, such as denoising and data correction, and then the data volume is reduced by using the built-in data compression algorithm and stored in the high-speed solid-state storage unit.

[0139] Step 2: Determine the transmission link and upload the data to the satellite network communication unit:

[0140] The preprocessed data is transmitted from the edge intelligent preprocessing unit to the communication unit. The communication unit uses LoRaWAN wireless communication technology to transmit the data to the satellite network communication unit.

[0141] The modem of the satellite network communication unit performs QPSK modulation on the data and uses the forward error correction (FEC) technology. In a test in a mountainous area, despite signal interference, the accuracy of data transmission is still over 98% after being processed by the FEC technology, effectively ensuring data integrity. The communication protocol processing unit encrypts the modulated data using the TLS / SSL encryption protocol and then packs it according to a specific protocol format.

[0142] With the automatic orientation function of the phased array antenna, the packed signal is stably transmitted to the remote management module through the low-earth orbit satellite network. In tests in remote areas, the antenna can quickly lock the satellite signal, and data can be stably uploaded even under complex terrain and adverse weather conditions.

[0143] Step 3: The remote management module receives and processes

[0144] The remote management module receives seismic data via the StarNet satellite, demodulates it using a modem, and decrypts it through the communication protocol processing unit. Then it analyzes the decrypted data, calculates parameters such as signal strength, frequency, and earthquake source direction, and finally outputs a three-dimensional fault model. During a seismic monitoring, the system quickly and accurately calculates the earthquake source direction, providing an important reference for subsequent rescue work. At the same time, the generated health monitoring report updates the device operation status in real time. Once abnormal situations such as too low device power are detected, the alarm mechanism is immediately triggered.

[0145] The working parameters such as the sampling rate and gain of the node seismograph module can also be adjusted in real time by means of the parameter configuration unit of the remote management module. When testing in different earthquake source intensity regions, the system automatically adjusts the parameters according to the actual situation to ensure the data acquisition effect. And it monitors the power, storage space, communication quality, etc. of all modules in real time to ensure the stable operation of the device.

[0146] Step 4: Data storage and backup

[0147] The remote management module stores all the received seismic data in the distributed storage system and performs redundant backup (During the long-term operation process, after multiple data recovery tests, it is proved that this storage and backup mechanism can effectively avoid data loss and ensure the security and integrity of the data).

[0148] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that they can still modify or equivalently replace the technical solutions of the present invention, and these modifications or equivalent replacements do not make the modified technical solutions deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A remote intelligent seismograph system based on star network communication, characterized in that: It includes a node seismograph module, which communicates bidirectionally with the remote management module via the star network communication unit, and the node seismograph module, the star network communication unit and the remote management module are all electrically connected to the multi-source energy module; The node seismograph module includes a connected three-axis MEMS seismic wave sensor, which is connected to the edge intelligent preprocessing unit via an amplification and filtering unit and an analog-to-digital conversion unit in sequence, and is used to collect seismic wave vibration signals using the three-axis MEMS seismic wave sensor, and preprocess the collected seismic wave vibration signals using the edge intelligent preprocessing unit; The star network communication unit includes a modem, a communication protocol processing unit and a phased array antenna connected in sequence. The phased array antenna is a three-mode link structure composed of an FSO communication unit, an adaptive radio frequency unit and a Beidou RDSS link. It is used to receive the pre-processed seismic wave vibration signal and upload it to the remote management module, or receive the remote management module's instructions and transmit it to the node seismograph module; The remote management module is used to receive seismic wave vibration signals and remotely monitor and manage the node seismometer modules to output a three-dimensional fault model and health monitoring report.

2. A remote intelligent seismograph system based on star network communication according to claim 1, characterized in that: The multi-source energy module includes a solar power generation unit and a vertical axis wind power generation unit, and both the solar power generation unit and the vertical axis wind power generation unit are electrically connected to the lithium iron phosphate battery pack via a battery management unit.

3. A remote intelligent seismograph system based on star network communication according to claim 1, characterized in that: The three-axis MEMS seismic wave sensor integrates a seismic wave sensor, an amplifying and filtering unit, an analog-to-digital conversion unit and a microprocessor which are electrically connected in sequence. The microprocessor is also connected to the edge intelligent preprocessing unit and the communication unit respectively. At the same time, the edge intelligent preprocessing unit is connected to the communication unit, and the communication unit communicates with the star network communication unit, so as to realize the collection of seismic wave vibration signals by the three-axis MEMS seismic wave sensor, and transmit them to the amplifying and filtering unit for filtering and signal amplification, and then transmit them to the microprocessor after analog-to-digital conversion by the analog-to-digital converter, and after data compression based on wavelet transform by the edge intelligent preprocessing unit, the collected seismic data is uploaded by the communication unit.

4. A remote intelligent seismograph system based on star network communication according to claim 3, characterized in that: The uplink of the FSO communication unit uses 1550nm wavelength laser modulation and combines the data compressed by wavelength-mode division multiplexing modulation. The transmission rate is ≥20Gbps and the bit error probability is ≤5×10 -9 ; The adaptive radio unit operates in the 1.4GHz-5.8GHz frequency band, supports OFDM modulation and polarization diversity, and the transmission rate is ≥50Mbps when the rain attenuation is ≤30dB; The RDSS link integrates BeiDou-3 B1 I+B2a dual-frequency reception, and the end-to-end delay in emergency mode is ≤2.5s.

5. A working method of a remote intelligent seismograph system based on star network communication as described in any one of claims 1 to 4, characterized in that: The following steps are involved: S1, the three-axis MEMS seismic wave sensor collects vibration signals at a set sampling rate, and after amplification, filtering and analog-to-digital conversion, uses the edge intelligent pre-processing unit to perform wavelet denoising and data compression to generate a feature data set; S2. Determine a communication transmission link, and upload the feature data set to a remote management module based on the determined communication transmission link; S3. Based on the feature data set, a three-dimensional fault model is generated using a remote management module and output.

6. The working method of a remote intelligent seismograph system based on star network communication according to claim 5, characterized in that: In step S1, the specific steps of performing wavelet denoising and data compression are as follows: The first step is to use the Daubechies9 orthogonal wavelet basis to perform discrete wavelet decomposition on the digital signal after analog-to-digital conversion to obtain the approximate coefficients and detail coefficients of each layer, and the decomposition formula is as follows: A j+1 [k]=∑ n h[n]×A j [2k-n]; <h2 style=";text-align:left;direction:ltr">D<h2 style=";text-align:left;direction:ltr"> j+1 <h2 style=";text-align:left;direction:ltr"> [k]=∑<h2 style=";text-align:left;direction:ltr"> n <h2 style=";text-align:left;direction:ltr"> g[n]×A<h2 style=";text-align:left;direction:ltr"> j <h2 style=";text-align:left;direction:ltr"> [2k-n]; In the formula, A j+1 and A j denote the approximate coefficients of the j+1th layer and the jth layer respectively; n denotes the number of decomposition layers; h[n] denotes a low-pass filter; D j+1 represents the detail coefficient of the j+1th layer; [2k-n] represents the downsampling operation; The second step is threshold denoising, and setting the threshold λ of the jth layer j for: In the formula, σ j N represents the noise standard deviation of the detail coefficient of the jth layer; j Indicates the length of the j-th layer detail coefficient; Step 3: Based on threshold λ j Modified wavelet coefficient W j,k : In the formula, represents the modified wavelet coefficients; Step 4: quantize the wavelet coefficients using a hierarchical compression strategy, where 5% of the maximum amplitude coefficient is retained for high-frequency noise and 95% of the amplitude coefficient is retained for low-frequency noise; Step 5: Perform Huffman encoding on the retained coefficients to achieve compression.

7. The working method of a remote intelligent seismograph system based on star network communication according to claim 5, characterized in that: Step S2 specifically includes the following steps: S21. Filter the link candidate set, and the filtering conditions are: In the formula, B candidate represents the candidate link bandwidth; S data represents the size of the feature dataset; Δt max Indicates the maximum allowed transmission time; S22. Use signal-to-noise ratio (SNR) and bit error rate (BER) to evaluate channel quality: Where P signal Indicates signal power; P noise Represents noise power; N error Indicates the number of bit errors; N total Indicates the complete transmission code; S23. Calculate the transmission performance scores of the FSO communication unit, adaptive radio unit and Beidou RDSS link based on the comprehensive weight scoring model: Where w1, w2 and w3 represent the weights of bandwidth, reliability and delay respectively; B max Indicates the maximum bandwidth of the candidate link; Latency indicates the link transmission delay; S24, sorting the transmission performance scores Score of the FSO communication unit, the adaptive radio frequency unit and the Beidou RDSS link, and taking the link corresponding to the largest transmission performance score Score as the optimal link; S25, modulating the received seismic wave vibration signal using the modem, encrypting and packaging the modulated signal using the communication protocol processing unit, and uploading it to the remote management module using the determined optimal link; Alternatively, the remote management module command is received by using the optimal link, and after demodulation by the modem, the modulated signal is encrypted and packaged by the communication protocol processing unit and then transmitted to the node seismograph module.

8. The working method of a remote intelligent seismograph system based on star network communication according to claim 5, characterized in that: Step S3 specifically includes the following steps: S31, performing time-frequency domain standardization on the original signal in the received feature data set: in, In the formula, χ norm represents the standardized data; x represents the original signal; μ represents the signal mean; σ represents the standard deviation; N represents the total number of samples; x i represents the i-th data sample; S32. Extract multi-scale features of faults through MultiRes-Unet3D neural network: in, Where, L weighted represents the weighted loss function; C represents the total number of categories; W c represents the weight of category c; y c represents the true label; p c represents the probability that the prediction is true; f c Represents the frequency of category c in the feature data set; ∈ represents a constant; S33. Generate fault point cloud based on stereo vision method and calculate depth information using multi-view seismic data: in, d=argmin d ∑ p∈Ω ∣I left (p)-I right (p+d)∣ 2 ; Where Z is the depth of the fault point; B is the binocular baseline distance; f is the focal length; d is the parallax; I left (p) and I right (p+d) represents the intensity values ​​of the left view and the right view at point p respectively; Ω represents the disparity search range; S34, using the iterative closest point algorithm to align the fault point cloud, where the objective function E of the iterative closest point algorithm is ICP as follows: Among them, p i represents the source point cloud; q i represents the target point cloud; T represents the rigid body transformation matrix; I represents the number of points in the fault point cloud; S35, converting the optimized point cloud into a gridded surface model to obtain a three-dimensional fault model.

Images