Data noise reduction and error code reduction method and system for mining rock burst monitoring sensor

Abstract

The invention relates to the technical field of mining safety monitoring, in particular to a data noise reduction and error code reduction method and system for a mining rock burst monitoring sensor. The method comprises the following steps: firstly, collecting a vibration signal of a mine working face through a sensor, then carrying out wavelet analysis and decomposition on the vibration signal to obtain a high-frequency noise component and a low-frequency effective signal component, removing the high-frequency noise component, and retaining the low-frequency effective signal component; performing wavelet reconstruction on the reserved low-frequency effective signal component, generating an effective vibration signal after noise reduction, encoding the effective vibration signal, transmitting the effective vibration signal to a data receiving end, decoding encoded data at the data receiving end, identifying and correcting an error code generated in a transmission process through a verification rule of a data error correction code, and finally performing data transmission. And outputting effective vibration monitoring data without error codes. According to the invention, integrated cooperative processing of noise separation, hardware anti-interference and transmission error correction is realized, and the reliability and stability of a monitoring system are enhanced.

Description

Technical Field

[0001] This invention relates to the field of mine safety monitoring technology, and in particular to a method and system for noise reduction and error reduction of mine rockburst monitoring sensor data. Background Technology

[0002] Rockbursts are a common and serious disaster during deep coal mining. Essentially, they are violent vibrations caused by the sudden release of energy from coal and rock masses under stress. These vibrations can lead to roadway collapses, equipment damage, and even casualties during mining operations. To prevent serious damage from rockbursts during production, vibration sensors are used in real-time monitoring of rockbursts during mine construction.

[0003] However, the mining environment is complex and harsh, with multiple interferences such as high voltage (e.g., underground high-voltage power supply systems), strong electromagnetic fields (e.g., electromagnetic radiation generated by mining equipment and locomotives), and mechanical noise (e.g., vibrations from personnel movement, belt conveyors, and scraper conveyors). This leads to the following core problems facing rockburst monitoring systems: 1. Severe signal noise interference: The vibration signals collected by the sensors are mixed with a large amount of non-target noise (such as low-frequency noise from people walking and periodic noise from belt conveyor operation). Traditional filtering methods (such as low-pass filtering and mean filtering) cannot accurately distinguish between noise and effective rockburst signals, resulting in distortion of the effective signals and affecting the accuracy of rockburst early warning. 2. Poor interference resistance during on-site installation: When sensors and communication cables are installed underground, they are easily exposed to strong electromagnetic environments. Existing sensors mostly use a single metal shielding layer, which has weak resistance to electromagnetic radiation. Communication cables lack standardized grounding design, which causes electromagnetic interference to enter the signal circuit through the sensor shell or cable. 3. High data transmission error rate: High-voltage pulses underground (such as pulse signals generated by high-voltage switch operation or motor start-stop) can interfere with the data transmission link, causing bit flips in the data collected by the sensors during transmission. Traditional data transmission protocols do not have error correction mechanisms designed for the strong interference environment in mines, resulting in an error rate as high as 10%. -3 -10 -2 This severely reduces data quality.

[0004] In existing technologies, some solutions attempt to improve anti-interference capabilities by increasing the thickness of the shielding layer or optimizing the filtering parameters, but no integrated solution of "signal noise reduction - installation anti-interference - transmission error correction" has been formed. Summary of the Invention

[0005] The present invention aims to at least partially solve one of the technical problems in the related art.

[0006] Therefore, the first objective of this invention is to provide a method for noise reduction and error reduction of data from a mine rockburst monitoring sensor, comprising: S1, which collects vibration signals from the mine working face through sensors; S2, perform wavelet analysis decomposition on the vibration signal to obtain high-frequency noise components and low-frequency effective signal components, and use an artificial intelligence model to identify and classify the noise components, remove high-frequency noise components and retain low-frequency effective signal components. S3 performs wavelet reconstruction on the retained low-frequency effective signal components to generate a denoised effective vibration signal, and uses data error correction code to encode the effective vibration signal, and transmits the encoded data to the data receiving end. S4 decodes the encoded data at the data receiving end, identifies and corrects errors generated during transmission through the verification rules of the data error correction code, and outputs error-free and valid vibration monitoring data.

[0007] In one embodiment of the present invention, S2 further includes, S21, the vibration signal collected by the sensor is decomposed into noise components of different frequencies using the db4 wavelet basis function; S22 uses a CNN-LSTM hybrid model to classify noise components of different frequencies and remove high-frequency noise components.

[0008] In one embodiment of the present invention, S22 further includes extracting frequency features of each noise component using a CNN model, extracting time series features of each noise component using an LSTM model, and outputting the classification results of the noise components through a fully connected layer. The frequency features include peak frequency and spectral energy distribution, and the time series features include vibration duration and amplitude variation trend.

[0009] In one embodiment of the present invention, the data error correction code in S3 is an RS (255,223) code, where 255 is the length of the encoded data and 223 is the length of the original valid data.

[0010] In one embodiment of the present invention, a composite shielding layer is provided outside the sensor, the composite shielding layer comprising an outer polyethylene insulating plastic layer and an inner tin foil layer.

[0011] In one embodiment of the present invention, the thickness of the composite shielding layer is The thickness of the tin foil is The thickness of the polyethylene insulating plastic is The electromagnetic shielding effectiveness of the shielding layer Breakdown voltage .

[0012] In one embodiment of the present invention, the frequency range corresponding to the high-frequency noise component is 125-1000Hz, and the frequency range corresponding to the low-frequency effective signal component is... .

[0013] To achieve the above objectives, a second aspect of the present invention provides a system for noise reduction and error reduction of data from a mine rockburst monitoring sensor, comprising a software system for data processing and transmission and a hardware device for noise interference suppression, wherein the software system includes: The wavelet analysis and classification module is used to perform wavelet analysis decomposition on the vibration signals collected at the mine site to obtain high-frequency noise components and low-frequency effective signal components. The trained artificial intelligence model is then used to identify and classify the components, removing the high-frequency noise components and retaining the low-frequency effective signal components. The wavelet reconstruction and error correction coding module is used to reconstruct the retained low-frequency effective signal components using wavelets, generate the noise-reduced effective vibration signal, and encode the effective vibration signal using data error correction codes, and transmit the encoded data to the data receiving end. The error correction decoding and data output module is used to decode the encoded data at the data receiving end, identify and correct the errors generated during transmission through the verification rules of the data error correction code, and output error-free valid vibration monitoring data.

[0014] In one embodiment of the present invention, the wavelet analysis and classification module is further used for: The vibration signal acquired by the sensor is decomposed into noise components of different frequencies using the db4 wavelet basis function; A CNN-LSTM hybrid model is used to classify noise components of different frequencies, removing high-frequency noise components. This model comprises a CNN model and an LSTM model. The CNN model extracts the frequency features of each noise component, while the LSTM model extracts the time-series features. The frequency and time-series features corresponding to the noise classification are input into a fully connected layer to obtain the classification result for that noise component. The frequency features include peak frequency and spectral energy distribution, and the time-series features include vibration duration and amplitude variation trend. In one embodiment of the present invention, the hardware device includes, A composite shielding layer sensor module is used to collect vibration signals at the mine site through a sensor with a composite shielding layer, wherein the composite shielding layer is composed of an outer layer of polyethylene insulating plastic and an inner layer of tin foil. The communication cable is used to transmit the effective vibration signal encoded by the wavelet reconstruction and error correction coding module to the data receiving end. The communication cable is a mine flame-retardant shielded cable with a composite shielding layer on the outside. A grounding electrode is set on the cable every 50-100m. The thickness of the composite shielding layer is The thickness of the tin foil is The thickness of the polyethylene insulating plastic is The electromagnetic shielding effectiveness of the shielding layer Breakdown voltage .

[0015] The method and system of this invention significantly improve the signal-to-noise ratio and recognition accuracy of mine rockburst monitoring signals, effectively reduce the data transmission error rate, realize the integrated collaborative processing of noise separation, hardware anti-interference and transmission error correction, and enhance the reliability and stability of the monitoring system.

[0016] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0017] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart of a method for noise reduction and error reduction of data from a mine rockburst monitoring sensor according to an embodiment of the present invention; Figure 2 This is a hardware device structure diagram of a mine rockburst monitoring sensor data noise reduction and error reduction system according to an embodiment of the present invention; Figure 3 This is a structural diagram of a composite shielding layer for a mine rockburst monitoring sensor data noise reduction and error reduction system according to an embodiment of the present invention; Figure 4 This is a structural diagram of a communication cable for a mine rockburst monitoring sensor data noise reduction and error reduction system according to an embodiment of the present invention. Detailed Implementation

[0018] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0019] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0020] The following description, with reference to the accompanying drawings, describes a method and system for noise reduction and error reduction of data from a mine rockburst monitoring sensor, as proposed in an embodiment of the present invention.

[0021] Example 1 Figure 1 This is a flowchart of a method for noise reduction and error reduction of data from a mine rockburst monitoring sensor according to an embodiment of the present invention.

[0022] like Figure 1 As shown, the method for noise reduction and error reduction of mine rockburst monitoring sensor data includes the following steps: S1 collects vibration signals from the mine working face using sensors.

[0023] Specifically, in some implementations, vibration signals from the mine working face are acquired using a sensor with a composite shielding layer. This composite shielding layer consists of an outer layer of polyethylene insulating plastic and an inner layer of tin foil. The technical principle behind this system is based on the synergistic effect of electromagnetic shielding and insulation protection. This sensor structure design aims to effectively isolate the influence of strong electromagnetic interference (EMI) and high-voltage electric fields commonly found in mines on the vibration signal acquisition process, thereby improving the signal-to-noise ratio and acquisition stability.

[0024] In the specific operation method, the inner side of the sensor housing is sequentially set with a thickness of... The polyethylene insulating plastic layer with a thickness of The aluminum foil layer forms a composite shielding structure of "polyethylene insulating plastic + aluminum foil". The aluminum foil acts as an electromagnetic shielding layer, utilizing its metallic properties to block high-frequency electromagnetic waves (such as...) The shielding effect is achieved by reflecting and absorbing mechanical noise and electromagnetic radiation within the range. Polyethylene insulating plastic serves as the insulating layer and has a breakdown voltage. It can effectively prevent signal distortion caused by electric field coupling between the sensor and the downhole high-pressure equipment, and at the same time, it provides physical protection for the tin foil, preventing its performance from deteriorating due to environmental factors such as moisture and dust.

[0025] Regarding parameter settings, the sensor module uses an intrinsically safe accelerometer for mining (such as the KGY81 model), with a measurement range of [missing information]. The frequency response range is The sampling frequency is set to To ensure the low-frequency vibration signal generated by rockburst is controlled ( Complete capture of ) . In practical applications, this structure is suitable for the middle of the sidewall of a fully mechanized mining face (distance from the floor). In installation scenarios, expansion bolts are used for fixing to avoid introducing additional vibration and noise due to loose installation.

[0026] This step plays a crucial role in the front-end signal acquisition and interference suppression in the entire technical solution, providing high-quality raw data input for subsequent wavelet analysis and artificial intelligence recognition. It is a fundamental link in realizing accurate monitoring and reliable data transmission of rockburst.

[0027] S2 performs wavelet analysis decomposition on the vibration signal to obtain high-frequency noise components and low-frequency effective signal components. Then, an artificial intelligence model is used to identify and classify the noise components, remove the high-frequency noise components, and retain the low-frequency effective signal components.

[0028] Specifically, in some implementations, wavelet analysis decomposition is performed on the vibration signal to obtain high-frequency noise components and low-frequency effective signal components. The trained artificial intelligence model is then used to identify and classify the components, removing the high-frequency noise components and retaining the low-frequency effective signal components. This is one of the core steps in the data noise reduction method of this invention, aiming to achieve accurate separation of complex mine noise and effective rockburst signals, thereby significantly improving the signal-to-noise ratio and the reliability of the monitoring system.

[0029] Further, step S2 includes, S21, the vibration signal collected by the sensor is decomposed into noise components of different frequencies using the db4 wavelet basis function.

[0030] This step first uses the db4 wavelet basis function to perform multi-scale decomposition on the raw vibration signal acquired by the sensor. The db4 wavelet has good time-frequency localization characteristics and is suitable for the analysis of non-stationary signals. In specific implementation, the number of wavelet decomposition levels is set to 3-5 to cover the noise frequency range commonly found in mines. For example, in a 4-level decomposition, the signal is decomposed into 4 high-frequency detail components (d1-d4) and 1 low-frequency approximation component (a4). The high-frequency components typically correspond to... Noise signals within the range, such as people walking and conveyor belt operation; while the low-frequency component a4 mainly contains the effective vibration signals generated by rockbursts, and its frequency range is usually [missing information]. The amplitude range is .

[0031] S22 uses a CNN-LSTM hybrid model to classify noise components of different frequencies and remove high-frequency noise components.

[0032] This step introduces a trained CNN-LSTM hybrid artificial intelligence model to identify and classify the decomposed components. CNN is used to extract the frequency features of the signal (such as peak frequency and spectral energy distribution), while LSTM is used to capture the time-series features of the signal (such as vibration duration and amplitude variation trend). The classification results are output through fully connected layers, identifying high-frequency components as noise and removing them, while retaining low-frequency effective signal components. During model training, the sample data undergoes normalization (amplitude normalized to [0,1]) and data augmentation (adding random Gaussian noise to simulate downhole interference) to improve the model's generalization ability. The training samples include three categories: personnel movement noise signals (…). Amplitude ), conveyor belt mechanical vibration noise signal ( Amplitude ) and effective rockburst signals ( Amplitude Model recognition accuracy It can effectively distinguish between noise and valid signals.

[0033] At the application level, this step is suitable for rockburst monitoring systems in deep fully mechanized mining faces, especially in environments with strong electromagnetic interference and frequent mechanical vibrations. By combining wavelet analysis with AI recognition, it is possible to achieve... The vibration signals collected at the sampling frequency are subjected to real-time noise reduction processing, which significantly improves the extraction accuracy of the effective signal.

[0034] The technical advantage of this step lies in the fact that, through the synergistic effect of multi-scale wavelet decomposition and artificial intelligence classification and recognition, high-frequency noise components can be accurately removed while retaining low-frequency effective signal components, thereby improving the signal-to-noise ratio to [value missing]. Compared to traditional filtering methods, this improves performance by 15-25 dB, providing a high-quality input signal for subsequent data encoding and transmission, which is a key step in achieving accurate early warning of rockbursts.

[0035] S3 performs wavelet reconstruction on the retained low-frequency effective signal components to generate a denoised effective vibration signal, and uses data error correction code to encode the effective vibration signal, and transmits the encoded data to the data receiving end.

[0036] Specifically, in the method of this invention, wavelet reconstruction of the retained low-frequency effective signal components and encoding with data error correction codes are key steps in achieving signal noise reduction and improving data transmission reliability. Technically, this step first utilizes the inverse transform process of wavelet analysis to reconstruct the retained effective low-frequency components (such as the a4 component, corresponding to a frequency range) after artificial intelligence identification. The wavelet reconstruction process reconstructs the vibration signal in the time domain, restoring its morphology. The wavelet reconstruction uses the same wavelet basis function db4 as the decomposition process, mapping the low-frequency signal back to the original signal space from the multi-scale decomposition space through a layer-by-layer inverse transform, thereby removing high-frequency noise components (such as d1-d4, corresponding to...). This process is used to obtain the effective vibration signal after noise reduction. The amplitude range of the reconstructed signal is... The frequency is concentrated in The signal-to-noise ratio has been improved to It is significantly better than traditional filtering methods. level.

[0037] In terms of data error correction coding, this invention uses RS (255, 223) code to encode the reconstructed effective vibration signal. RS code is a linear block code based on a finite field, with a coding length of 255 bytes, containing 223 bytes of effective data and 32 bytes of redundancy check code. This coding method has strong burst error correction capability, correcting up to 16 erroneous bytes in every 255 bytes of data. It is suitable for bit flipping problems caused by sudden electromagnetic interference in mines, such as high-voltage switch operation and motor start-up / stop. The encoding process is executed by an STM32H743 processor in the signal processing module, and the encoded data is transmitted through a mine-grade flame-retardant shielded cable (outer layer wrapped with...). (Thick composite shielding layer) transmits data to the ground monitoring center.

[0038] In practical applications, this step is typically deployed in the signal processing box of a deep fully mechanized mining face to process vibration data from multiple sensor nodes. By combining wavelet reconstruction and RS coding, not only is the signal purity improved, but the bit error rate during data transmission is also effectively reduced, keeping the overall system bit error rate within a certain range. The magnitude of the signal is reduced by 3-4 orders of magnitude compared to traditional solutions. Technically, this step enhances end-to-end reliability from signal processing to data transmission, providing high-quality, low-error-rate data support for real-time monitoring and early warning of rockbursts, demonstrating significant engineering practical value.

[0039] S4 decodes the encoded data at the data receiving end, identifies and corrects errors generated during transmission through the verification rules of the data error correction code, and outputs error-free and valid vibration monitoring data.

[0040] Specifically, the encoded data is decoded at the data receiving end. Errors generated during transmission are identified and corrected using the verification rules of the data error correction code, resulting in the output of error-free and valid vibration monitoring data. This step is a key component of the data transmission error correction mechanism of this invention. Its technical implementation is based on the decoding algorithm of RS (255,223) error correction code, aiming to ensure that the vibration data collected by the sensor is not distorted due to sudden errors during transmission in the strong electromagnetic interference environment of the mine, thereby ensuring the high reliability and integrity of the signals received by the ground monitoring system.

[0041] In some implementations, the decoding and error correction process first verifies the received encoded data using an RS code decoder. RS code is a non-binary linear block code with a length of 255 bytes, containing 223 bytes of valid data and 32 bytes of check information. During decoding, the system first calculates the syndrome of the received data; if the syndrome is non-zero, it indicates an error in the data. Subsequently, the Berlekamp-Massey algorithm or the Euclidean algorithm is used to identify the error location polynomial. The Chien search and Forney algorithm are then used to determine the error location and value, and correction is performed. This process can correct burst errors of up to 16 bytes and is suitable for multi-bit errors caused by sudden electromagnetic interference in mines, such as high-voltage switch operation or motor start-up and shutdown.

[0042] Optionally, the decoder can be implemented in hardware using an FPGA (Field-Programmable Gate Array) or ASIC (Application-Specific Integrated Circuit) architecture to meet the real-time and stability requirements of the mine monitoring system. In software implementation, the decoding algorithm can run on an STM32H743 processor or a higher-performance embedded platform to ensure efficient error correction even with limited resources.

[0043] Furthermore, in practical applications, this step involves deploying a data receiving terminal at the ground monitoring center. Its built-in RS decoder is connected to a communication cable, which is a mining-grade flame-retardant shielded cable. The outer layer is wrapped with the same polyethylene insulation plastic and tin foil composite shielding structure as the sensor. Grounding electrodes are installed every 50-100m along the tunnel, with a grounding resistance of [missing information]. To reduce the impact of electromagnetic interference on the transmission link, the decoding and error correction module successfully corrected three burst errors caused by the operation of the underground high-voltage switch in the experimental test. The final output bit error rate detection value was [value missing]. This reduces transmission speed by 3-4 orders of magnitude compared to traditional methods.

[0044] The technical value of this step lies in the fact that by introducing the decoding and error correction mechanism of RS (255,223) code, the transmission reliability of mine vibration monitoring data is effectively improved, and signal loss or misjudgment caused by bit errors is avoided. This provides high-fidelity, low-error data support for real-time early warning and disaster prevention of rockbursts, and is an important guarantee for realizing the integrated solution of "signal noise reduction - installation interference resistance - transmission error correction".

[0045] The noise reduction and error reduction method for mine rockburst monitoring sensor data in this invention significantly improves the signal-to-noise ratio and data transmission reliability of mine rockburst monitoring signals, effectively separates complex noise from valid signals, and reduces the error rate to [missing value]. Magnitude.

[0046] Example 2 A noise reduction and error reduction system for mine rockburst monitoring sensor data is characterized by comprising a software system for data processing and transmission and a hardware device for noise interference suppression, wherein the software system includes: The wavelet analysis and classification module is used to perform wavelet analysis decomposition on the vibration signals collected at the mine site to obtain high-frequency noise components and low-frequency effective signal components. The trained artificial intelligence model is then used to identify and classify the components, removing the high-frequency noise components and retaining the low-frequency effective signal components. The wavelet reconstruction and error correction coding module is used to reconstruct the retained low-frequency effective signal components using wavelets, generate the noise-reduced effective vibration signal, and encode the effective vibration signal using data error correction codes, and transmit the encoded data to the data receiving end. The error correction decoding and data output module is used to decode the encoded data at the data receiving end, identify and correct the errors generated during transmission through the verification rules of the data error correction code, and output error-free valid vibration monitoring data.

[0047] Furthermore, the wavelet analysis and classification module is also used for: The vibration signal acquired by the sensor is decomposed into noise components of different frequencies using the db4 wavelet basis function; A CNN-LSTM hybrid model is used to classify noise components of different frequencies and remove high-frequency noise components. The CNN-LSTM hybrid model includes a CNN model and an LSTM model. The CNN model is used to extract the frequency features of each noise component, and the LSTM model is used to extract the time series features of each noise component. The frequency features and time series features corresponding to the noise classification are input into a fully connected layer to obtain the classification result of the noise component. The frequency features include peak frequency and spectral energy distribution, and the time series features include vibration duration and amplitude variation trend.

[0048] The hardware device includes: A composite shielding layer sensor module is used to collect vibration signals at the mine site through a sensor with a composite shielding layer, wherein the composite shielding layer is composed of an outer layer of polyethylene insulating plastic and an inner layer of tin foil. The communication cable is used to transmit the effective vibration signal encoded by the wavelet reconstruction and error correction coding module to the data receiving end. The communication cable is a mine-use flame-retardant shielded cable with a composite shielding layer on the outside. A grounding electrode is set on the cable every 50-100m.

[0049] Specifically, in some implementations, encoded data is transmitted via communication cables with grounding electrodes set at predetermined intervals. The outer layer of these communication cables is wrapped with the same polyethylene insulating plastic and tin foil structure as the sensor's composite shielding layer. This technology is based on the synergistic effect of electromagnetic shielding and electrostatic grounding, aiming to effectively suppress strong electromagnetic interference (EMI) and electrostatic induction noise in the mine environment, thereby ensuring the integrity and reliability of data transmission. This step is a key component of the anti-interference installation structure of this invention, and together with the data error correction mechanism, constitutes both physical and logical protection against interference for the system.

[0050] In terms of specific operational methods, such as Figure 4 As shown, the communication cable is a mining-grade flame-retardant shielded cable. Its outer layer structure is consistent with the composite shielding layer of the sensor, consisting of a polyethylene insulating plastic layer and a tin foil shielding layer from the outside in. The tin foil thickness is... The thickness of the polyethylene insulating plastic is The overall shielding layer thickness is This structure utilizes the metallic conductivity of tin foil to reflect and absorb electromagnetic waves, achieving high shielding effectiveness. Effective attenuation Electromagnetic interference signals at the above frequencies. Meanwhile, polyethylene insulating plastic has excellent dielectric properties, with a breakdown voltage of... It can prevent the insulation of cables from failing due to the high-pressure environment underground, and at the same time protect the shielding layer from environmental factors such as moisture and dust.

[0051] Furthermore, to enhance the electrostatic discharge capability of the shielding layer, during the laying of communication cables along the tunnel, at each interval... Set up a grounding electrode. The grounding electrode should be of length [length missing]. Diameter is Galvanized steel pipes are vertically buried below the mine floor. This is to ensure good electrical contact with the ground. (Through cross-sectional area) The copper cable connects the cable shielding layer to the grounding electrode, and the grounding resistance... This allows for the rapid discharge of induced charges on the cable surface, reducing the interference of static electricity accumulation on the signal circuit.

[0052] This communication cable is suitable for rockburst monitoring systems in deep fully mechanized mining faces, especially in complex electromagnetic environments with high-voltage power supplies, mining equipment operation, and mechanical vibrations. Through periodic grounding, it effectively suppresses transient electromagnetic pulse interference caused by the start-up and shutdown of underground equipment and the operation of locomotives. Combined with RS (255,223) encoding technology, the data error rate is significantly reduced to [missing value]. This represents a 3-4 order of magnitude improvement over traditional transmission methods.

[0053] In terms of technical effectiveness, the communication cable significantly improves the anti-interference capability of the communication link through physical shielding and grounding design, providing a high-quality transmission channel for subsequent data decoding and ground pressure signal analysis. It is an important supporting link for realizing the integrated technical solution of "signal noise reduction - installation anti-interference - transmission error correction".

[0054] Furthermore, such as Figure 3 As shown, the thickness of the composite shielding layer is The thickness of the tin foil is The thickness of the polyethylene insulating plastic is The electromagnetic shielding effectiveness of the shielding layer Breakdown voltage .

[0055] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0056] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A method for noise reduction and error reduction of data from a mine rockburst monitoring sensor, characterized in that, include: S1, which collects vibration signals from the mine working face through sensors; S2, perform wavelet analysis decomposition on the vibration signal to obtain high-frequency noise components and low-frequency effective signal components, and use an artificial intelligence model to identify and classify the noise components, remove high-frequency noise components and retain low-frequency effective signal components. S3 performs wavelet reconstruction on the retained low-frequency effective signal components to generate a denoised effective vibration signal, and uses data error correction code to encode the effective vibration signal, and transmits the encoded data to the data receiving end. S4 decodes the encoded data at the data receiving end, identifies and corrects errors generated during transmission through the verification rules of the data error correction code, and outputs error-free and valid vibration monitoring data.

2. The method as described in claim 1, characterized in that, S2 also includes, S21, the vibration signal collected by the sensor is decomposed into noise components of different frequencies using the db4 wavelet basis function; S22 uses a CNN-LSTM hybrid model to classify noise components of different frequencies and remove high-frequency noise components.

3. The method as described in claim 2, characterized in that, S22 further includes using a CNN model to extract frequency features of each noise component, using an LSTM model to extract time series features of each noise component, and outputting the classification results of the noise components through a fully connected layer. The frequency features include peak frequency and spectral energy distribution, and the time series features include vibration duration and amplitude variation trend.

4. The method as described in claim 1, characterized in that, The data error correction code in S3 is an RS (255,223) code, where 255 is the length of the encoded data and 223 is the length of the original valid data.

5. The method as described in claim 1, characterized in that, The sensor is provided with a composite shielding layer, which includes an outer polyethylene insulating plastic layer and an inner tin foil layer.

6. The method as described in claim 1, characterized in that, The thickness of the composite shielding layer is The thickness of the tin foil is The thickness of the polyethylene insulating plastic is The electromagnetic shielding effectiveness of the shielding layer Breakdown voltage .

7. The method as described in claim 1, characterized in that, The frequency range corresponding to the high-frequency noise component is 125-1000Hz, and the frequency range corresponding to the low-frequency effective signal component is... .

8. A system for noise reduction and error reduction of mine rockburst monitoring sensor data, characterized in that, It includes a software system for data processing and transmission and a hardware device for noise immunity, wherein the software system includes: The wavelet analysis and classification module is used to perform wavelet analysis decomposition on the vibration signals collected at the mine site to obtain high-frequency noise components and low-frequency effective signal components. The trained artificial intelligence model is then used to identify and classify the components, removing the high-frequency noise components and retaining the low-frequency effective signal components. The wavelet reconstruction and error correction coding module is used to reconstruct the retained low-frequency effective signal components using wavelets, generate the noise-reduced effective vibration signal, and encode the effective vibration signal using data error correction codes, and transmit the encoded data to the data receiving end. The error correction decoding and data output module is used to decode the encoded data at the data receiving end, identify and correct the errors generated during transmission through the verification rules of the data error correction code, and output error-free valid vibration monitoring data.

9. The apparatus as claimed in claim 8, characterized in that, The wavelet analysis and classification module is also used for: The vibration signal acquired by the sensor is decomposed into noise components of different frequencies using the db4 wavelet basis function; A CNN-LSTM hybrid model is used to classify noise components of different frequencies and remove high-frequency noise components. The CNN-LSTM hybrid model includes a CNN model and an LSTM model. The CNN model is used to extract the frequency features of each noise component, and the LSTM model is used to extract the time series features of each noise component. The frequency features and time series features corresponding to the noise classification are input into a fully connected layer to obtain the classification result of the noise component. The frequency features include peak frequency and spectral energy distribution, and the time series features include vibration duration and amplitude variation trend.

10. The apparatus as claimed in claim 8, characterized in that, The hardware device includes, A composite shielding layer sensor module is used to collect vibration signals at the mine site through a sensor with a composite shielding layer, wherein the composite shielding layer is composed of an outer layer of polyethylene insulating plastic and an inner layer of tin foil. The communication cable is used to transmit the effective vibration signal encoded by the wavelet reconstruction and error correction coding module to the data receiving end. The communication cable is a mine flame-retardant shielded cable with a composite shielding layer on the outside. A grounding electrode is set on the cable every 50-100m. The thickness of the composite shielding layer is The thickness of the tin foil is The thickness of the polyethylene insulating plastic is The electromagnetic shielding effectiveness of the shielding layer Breakdown voltage .

Images