A method for downhole multi-parameter environmental monitoring

Abstract

This invention relates to the field of mining technology and discloses a method for multi-parameter environmental monitoring underground. The method includes the following steps: real-time acquisition of equipment pose, vibration spectrum, and environmental parameters through multi-source sensors; spatiotemporal alignment, filtering and noise reduction, and feature extraction of the raw data; outputting a standardized state vector; fusing pose data, vibration signals, and environmental parameters under a unified spatiotemporal reference to construct a multi-dimensional feature matrix; and using principal component analysis to reduce dimensionality and remove redundant information, outputting the fused equipment state vector. This invention achieves real-time fusion of equipment pose and airflow dynamics, collaboratively controlling ventilation equipment. Based on coupled modeling of vibration spectrum and pose drift, it intelligently compensates for sensor reading errors, eliminates the interference of mechanical vibration on gas monitoring accuracy, and ensures the reliability of gas concentration detection. Through spatial correlation analysis and dynamic airflow optimization, it proactively eliminates sudden drops in wind speed caused by mine car passage, achieving continuous monitoring throughout the roadway.

Description

Technical Field

[0001] This invention relates to the field of mining, and more specifically, to a method for monitoring multi-parameter environmental conditions underground. Background Technology

[0002] In deep mining, the dynamic operations such as tunneling machine propulsion and transport vehicle movement cause airflow disturbances and equipment interference, leading to three core problems for traditional environmental monitoring systems:

[0003] The air volume at the tunnel face fluctuated by more than ±15% due to equipment displacement.

[0004] Equipment vibration caused sensor readings to drift (gas detection error reached ±0.15%).

[0005] When the transport vehicle passes by, the wind speed drops by 40% in some areas, creating a blind spot for monitoring.

[0006] Existing static monitoring models are inadequate to handle such dynamic disturbances, and there is an urgent need to develop monitoring methods that strongly couple equipment and environment. Summary of the Invention

[0007] This invention provides a downhole multi-parameter environmental monitoring method, which solves the technical problem that existing static monitoring models in related technologies are unable to cope with such dynamic interference.

[0008] This invention provides a downhole multi-parameter environmental monitoring method, comprising the following steps:

[0009] S100, Real-time Equipment Status Perception: Real-time acquisition of equipment pose, vibration spectrum and environmental parameters through multi-source sensors, spatiotemporal alignment, filtering and noise reduction and feature extraction of raw data, and output of standardized status vector.

[0010] S200, multi-source data fusion processing: fuses pose data, vibration signals and environmental parameters under a unified spatiotemporal reference to construct a multi-dimensional feature matrix; uses principal component analysis to reduce dimensionality and remove redundant information, and outputs the fused device state vector;

[0011] S300, Digital Twin Disturbance Modeling: Based on fused data, a vibration-sensitive model and an environmental disturbance model are established. The least squares method is used to fit the equations, the model accuracy is verified online, and the pose drift prediction equation is output.

[0012] S400, Dynamic Feedforward Compensation Control: Predicts the pose drift based on the disturbance model, generates control signals by combining the feedforward-PID composite controller, sends control commands to the actuator in real time, and verifies the compensation effect through actual measurement data.

[0013] S500, edge intelligent decision-making: evaluates compensation performance indicators, dynamically triggers model updates or parameter optimization, performs hierarchical responses based on comprehensive security situation, and synchronizes decisions to the digital twin platform;

[0014] S600, control command output: reconstruct control signals, perform encryption encoding and timestamp synchronization, transmit commands according to priority through real-time network, receive execution feedback and calculate control error and delay to form closed-loop verification.

[0015] Furthermore, S100 includes the following steps:

[0016] S110, Equipment pose acquisition: Real-time capture of the three-dimensional coordinates and attitude angles of the tunneling machine and transport vehicle through UWB base station and tag, LiDAR scanning area point cloud data, dynamic modeling of equipment surface deformation;

[0017] S120, Environmental parameter acquisition: Synchronously reads data from gas / temperature / humidity / wind speed sensors, and monitors the internal pressure of the air duct by a pressure sensor;

[0018] S130, Vibration signal acquisition: Deploy accelerometers in key parts of the equipment to acquire raw vibration signals from 0 to 500 Hz;

[0019] S140, Vibration spectrum calculation: Perform FFT transformation on the original acceleration signal to calculate the power spectral density;

[0020] S150, Data Encapsulation and Timestamp Marking: Pack pose / environment / spectrum data and add PTP1588v2 protocol timestamps.

[0021] Furthermore, S200 includes the following steps:

[0022] S210, PTP1588v2 time synchronization: Aligns the time base of all devices / sensors through a master-slave clock architecture, calculates clock offset and corrects timestamps;

[0023] S220, Equipment Coordinate System to Roadway Global Coordinate System Transformation: Transform the equipment pose to the roadway global coordinate system and fuse LiDAR point cloud data to correct positioning errors;

[0024] S230, Environmental Sensor Spatial Mapping: Calculate global coordinates based on sensor installation location and establish sensor-device spatial relationship matrix;

[0025] S240, Spatiotemporal correlation of vibration spectrum: Correlate the vibration spectrum density vector to the location of a specific device to establish a frequency-spatial distribution model;

[0026] S250, Spatiotemporal Index Construction: Establish an R* tree spatiotemporal index to achieve millisecond-level cross-modal data retrieval.

[0027] Furthermore, in S300, the following steps are included:

[0028] S310, Vibration-pose coupling modeling: Analyze the quantitative relationship between vibration spectrum and equipment pose drift, and establish a frequency sensitivity matrix;

[0029] S320, Environmental-Sensor Error Modeling: Quantifying the impact of environmental parameters on positioning / measurement errors and constructing a temperature, humidity and gas composite interference model;

[0030] S330, comprehensive prediction of pose drift: integrates vibration and environmental factors to predict equipment pose changes and constructs a state-space model;

[0031] S340, Perturbation parameter identification: Training sensitivity coefficients and perturbation coefficients based on historical data;

[0032] S350, Model Validation and Confidence Assessment: Calculate the predicted residuals and confidence intervals, and dynamically update the model parameters.

[0033] Furthermore, the calculation formula for vibration-pose coupling modeling is as follows:

[0034] Vibration-induced pose drift:

[0035] ;

[0036] ;

[0037] ;

[0038] in, Let be the pose drift vector caused by vibration. This is the frequency sensitivity matrix. The vibration spectral density vector. This represents the power spectral density value at the first frequency point. This is the power spectral density value at the second frequency point. Let K be the power spectral density value at the Kth frequency point. This is a transpose operation;

[0039] Frequency sensitivity matrix:

[0040] ;

[0041] in, Let be the sensitivity coefficient of the vibration in the j-th frequency band to the drift along the i-th axis. This represents the sensitivity coefficient of the first frequency band vibration to the x-axis drift. This represents the sensitivity coefficient of the second frequency band vibration to the x-axis drift. Let be the sensitivity coefficient of the K-th frequency band vibration to the x-axis drift. This represents the sensitivity coefficient of the vibration in the first frequency band to the y-axis drift. This represents the sensitivity coefficient of the second frequency band vibration to the y-axis drift. Let be the sensitivity coefficient of the K-th frequency band vibration to the y-axis drift. This represents the sensitivity coefficient of the first frequency band vibration to the z-axis drift. This represents the sensitivity coefficient of the second frequency band vibration to the z-axis drift. Let be the sensitivity coefficient of the K-th frequency band vibration to the z-axis drift. For the dimension of the spectrum vector, This refers to the device's three-dimensional coordinates in the global coordinate system. This represents the change in position components along the x, y, and z axes in the global coordinate system. For change operators.

[0042] Furthermore, the calculation formula for environment-sensor error modeling is as follows:

[0043] Comprehensive environmental error:

[0044] ;

[0045] ;

[0046] ;

[0047] in, The error is due to a combination of environmental factors. For temperature deviation, The current temperature. For reference temperature, For humidity deviation, The current humidity. For reference humidity, Gas / carbon monoxide concentration, The temperature interference coefficient is... Humidity interference coefficient, The gas concentration interference coefficient, This represents the interference coefficient for carbon monoxide concentration.

[0048] Furthermore, the calculation formula for pose drift comprehensive prediction is as follows:

[0049] State-space equations:

[0050] ;

[0051] ;

[0052] in, This is the device state vector. This is the state vector of the device at the next moment. This refers to the device's three-dimensional coordinates in the global coordinate system. The velocity components in the global coordinate system. Here is the state transition matrix. The vibration disturbance input matrix is... The input matrix is ​​the environmental disturbance. For Gaussian process noise, This is a discrete-time index.

[0053] Furthermore, in S400, the following steps are included:

[0054] S410, disturbance prediction calculation, predicts pose drift based on the state-space model of S330, and integrates vibration and environmental disturbance factors;

[0055] S420, Compensation Control Quantity Generation: Design a PID controller to generate compensation instructions, and introduce feedforward compensation to offset prediction drift;

[0056] S430, multi-objective collaborative optimization: resolves compensation conflicts among multiple devices and meets safety and energy consumption constraints;

[0057] S440, compensation command issuance: commands are issued to the device PLC via the OPCUA protocol. The command format is standardized and the transmission delay requirement is <20ms.

[0058] S450, Compensation Effect Verification: Compare the pose error before and after compensation, and calculate key performance indicators.

[0059] Furthermore, the S500 includes the following steps:

[0060] S510, Dynamic evaluation of compensation effect: Calculate key performance indicators based on the verification results of S450, and trigger model updates and parameter optimization thresholds;

[0061] S520, Online update of disturbance model: The vibration sensitivity coefficient and environmental disturbance coefficient are updated in real time using the recursive least squares method;

[0062] S530, Control Parameter Co-optimization: PID gain and feedforward gain are optimized through particle swarm optimization algorithm;

[0063] S540, Safety Situation Decision: Executes safety strategies based on integrated environmental and vibration data, and generates equipment-level / system-level emergency commands;

[0064] S550, Digital Twin Synchronization and Logs: Synchronizes decision results to the digital twin platform, generates encrypted decision logs and security requirements, log integrity is verified by SHA-256 hash, and synchronization latency is <50ms.

[0065] Furthermore, in S600, the following steps are included:

[0066] S610, Control Signal Reconstruction: The control signal is generated by integrating the optimized control parameters of S530 with the real-time pose error, and feedforward compensation is embedded.

[0067] S620, security instruction embedding: Based on the security decision overlay control signal of S540, generate device-level security linkage instructions;

[0068] S630, Signal Security Encoding: Encrypts and verifies the integrity of control signals, and adds timestamps for synchronization;

[0069] S640, Real-time Transmission Scheduling: Transmits encrypted commands via TSN and dynamically allocates transmission priorities;

[0070] S650, Execution Status Feedback: Receives execution confirmation signals from the device PLC and calculates control errors and delays.

[0071] The beneficial effects of this invention are as follows:

[0072] This invention integrates real-time equipment position and airflow dynamics to coordinate and regulate ventilation equipment, significantly mitigating sudden airflow changes caused by equipment movement and ensuring stable airflow at the tunneling face. Based on coupled modeling of vibration spectrum and position drift, it intelligently compensates for sensor reading errors, eliminates the interference of mechanical vibration on gas monitoring accuracy, and ensures the reliability of gas concentration detection. Through spatial correlation analysis and dynamic airflow optimization, it proactively eliminates sudden drops in wind speed caused by mine car passage, achieving continuous monitoring of the entire roadway and eliminating the risk of local data loss. Ultimately, it forms a closed loop of "perception-modeling-compensation-decision-making," improving the robustness and safety of underground environmental monitoring. Attached Figure Description

[0073] Figure 1 This is a flowchart of a downhole multi-parameter environmental monitoring method proposed in this invention;

[0074] Figure 2 This is a flowchart of the sub-steps of S100 in the downhole multi-parameter environmental monitoring method proposed in this invention;

[0075] Figure 3 This is a flowchart of the sub-steps of S200 in the downhole multi-parameter environmental monitoring method proposed in this invention;

[0076] Figure 4 This is a flowchart of the sub-steps of S300 in the downhole multi-parameter environmental monitoring method proposed in this invention;

[0077] Figure 5This is a flowchart of the sub-steps of S400 in the downhole multi-parameter environmental monitoring method proposed in this invention;

[0078] Figure 6 This is a flowchart of the sub-steps of the S500 method for downhole multi-parameter environmental monitoring proposed in this invention;

[0079] Figure 7 This is a flowchart of the sub-steps of S600 in the downhole multi-parameter environmental monitoring method proposed in this invention. Detailed Implementation

[0080] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and implement the subject matter described herein, and changes may be made to the function and arrangement of the elements discussed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the examples. Furthermore, features described in some examples may be combined in other examples.

[0081] like Figures 1-7 As shown, a downhole multi-parameter environmental monitoring method includes the following steps:

[0082] S100, Real-time Equipment Status Awareness: Real-time acquisition of equipment pose, vibration spectrum and environmental parameters (temperature, gas concentration, etc.) through multi-source sensors (IMU, LiDAR, environmental sensors), spatiotemporal alignment, filtering and noise reduction and feature extraction (such as vibration spectrum peak value) of raw data, and output of standardized status vector.

[0083] In one embodiment of the present invention, the following steps are specifically included:

[0084] S110, Equipment Pose Acquisition (UWB / LiDAR Positioning): Real-time capture of the three-dimensional coordinates and attitude angles of the tunneling machine and transport vehicle through UWB base stations and tags, point cloud data of the LiDAR scanning area, and dynamic modeling of equipment surface deformation.

[0085] formula:

[0086] Device pose vector:

[0087] ;

[0088] in, The coordinates are in the Cartesian coordinate system. Yaw (in rad) The pitch angle is measured in rad. The device pose state vector;

[0089] Positioning accuracy constraints:

[0090] ;

[0091] in, Let be the pose measurement error vector. For the Euclidean norm, This is the positioning accuracy threshold.

[0092] S120, Environmental parameter acquisition (multi-sensor fusion): synchronously reads data from gas / temperature and humidity / wind speed sensors, and monitors the internal pressure of the air duct by a pressure sensor;

[0093] formula:

[0094] Environmental parameter vector:

[0095] ;

[0096] in, For temperature, For humidity, This refers to the methane concentration. This refers to the carbon monoxide concentration. For wind speed, The air pressure in the ventilation duct. For time indexing, Let t be the environmental parameter state vector;

[0097] S130, Vibration signal acquisition (triaxial MEMS accelerometer): Accelerometers are deployed in key parts of the equipment to acquire raw vibration signals from 0-500Hz.

[0098] formula:

[0099] Acceleration time-domain signal:

[0100] ;

[0101] in, For time variables, It is a triaxial acceleration. Let τ be the acceleration vector at time τ;

[0102] Sampling Theorem:

[0103] ;

[0104] in, Sampling frequency, The highest frequency of the signal. The Nyquist sampling coefficients;

[0105] S140, Vibration spectrum calculation (edge ​​node preprocessing): Perform FFT transformation on the original acceleration signal to calculate the power spectral density (PSD).

[0106] formula:

[0107] Power spectral density function:

[0108] ;

[0109] in, For frequency, For the Hanning window function, For the number of segments, For each segment of points, It is a discrete-time acceleration signal. The imaginary unit, For power spectral density, For discrete-time indexing, For segmented indexes;

[0110] Output spectrum vector:

[0111] ;

[0112] in, For the k-th frequency point, , For the dimension of the spectrum vector, Vibration spectral density vector For frequency range, These are the power spectral densities from the 1st to the Kth, respectively.

[0113] S150, Data Encapsulation and Timestamp Marking: Pack pose / environment / spectrum data and add PTP1588v2 protocol timestamp;

[0114] Timestamp alignment:

[0115] ;

[0116] in, This includes a time offset calculated by the PTP1588 protocol, used to correct clock skew between different devices. This is the sensor's local timestamp, representing the local system time when the sensor acquired data. The timestamp after synchronization indicates the unified time base after clock synchronization;

[0117] Packet serialization:

[0118] ;

[0119] in, The data packet is in serialized JSON format. This is a JSON encoding function that converts a data structure into a JSON string. For the data packet timestamp field, the value is , This is a data field containing the pose vector. Environmental parameter vector and spectrum vector ;

[0120] Output: Structured data packets (including pose, environment, and spectrum data);

[0121] S200, multi-source data fusion processing: pose data, vibration signals, and environmental parameters are fused under a unified spatiotemporal reference to construct a multi-dimensional feature matrix; principal component analysis (PCA) is used to reduce dimensionality and remove redundant information, and the fused device state vector is output;

[0122] In one embodiment of the present invention, the following steps are specifically included:

[0123] S210, PTP1588v2 time synchronization: Aligns the time base of all devices / sensors through a master-slave clock architecture, calculates clock offset and corrects timestamps;

[0124] formula:

[0125] Clock offset calculation:

[0126] ;

[0127] in, The time when the master clock sends a Sync message. To receive Sync message time from the clock, To send the Delay_Req time from the clock, Receive Delay_Req time from the master clock. This is the clock offset, representing the time difference between the master and slave clocks;

[0128] Timestamp correction:

[0129] ;

[0130] in, This is the local original timestamp, the time recorded locally by the device. This is the synchronized timestamp, a unified time after offset correction. This is the clock offset;

[0131] S220, Equipment coordinate system to roadway global coordinate system transformation: Convert equipment pose Transform to the global coordinate system of the tunnel and integrate LiDAR point cloud data to correct positioning errors;

[0132] formula:

[0133] Homogeneous transformation matrix:

[0134] ;

[0135] in, This is a homogeneous transformation matrix used for coordinate system transformation. This is the yaw angle rotation matrix, representing the rotation of the equipment in the horizontal plane. This is the pitch angle rotation matrix, representing the vertical rotation of the equipment. Let be the translation vector, representing the displacement of the origin. Yaw angle, representing the angle of rotation of the equipment in the horizontal plane. The pitch angle represents the angle of inclination of the equipment relative to the horizontal plane. The coordinates of the tunnel reference point, It is a cosine function. It is a sine function. and Fill the matrix with constants;

[0136] Global coordinate calculation:

[0137] ;

[0138] ;

[0139] in, This is the device pose vector, which contains position and orientation information. This is a homogeneous transformation matrix used for coordinate system transformation. The global coordinate vector represents the transformed device location. This refers to the device's three-dimensional coordinates in the global coordinate system. It is the transpose symbol;

[0140] S230, Environmental Sensor Spatial Mapping: Calculate global coordinates based on sensor installation location and establish sensor-device spatial relationship matrix;

[0141] formula:

[0142] Sensor global coordinates:

[0143] ;

[0144] ;

[0145] in, This represents the sensor's position in the global coordinate system. This represents the device's position in the global coordinate system. The equipment rotation matrix is ​​determined by the yaw angle. and pitch angle Calculations show that This refers to the sensor's installation offset relative to the device's reference point;

[0146] Distance matrix:

[0147] ;

[0148] in, For sensor number index, For equipment number indexing, Given the Euclidean norm, calculate the distance between two points. Let i be the global coordinates of the i-th sensor. Let j be the global coordinates of the j-th device. This is a distance matrix, representing the distances between all sensors and the device;

[0149] S240, Spatiotemporal correlation of vibration spectrum: The vibration spectrum density vector is... Establish a frequency-spatial distribution model by associating it with the location of a specific device;

[0150] formula:

[0151] Spatial distribution of vibration energy:

[0152] ;

[0153] in, For spatial points Vibrational energy density at the location The global coordinates of the vibration source are given. The global coordinates of the sensor, The spatial attenuation coefficient of the k-th frequency component. For frequency Power spectral density at that point The dimension of the spectrum vector represents the number of frequency components. This refers to the three-dimensional spatial position in the global coordinate system.

[0154] S250, Spatiotemporal Index Construction: Establish an R* tree spatiotemporal index to achieve millisecond-level cross-modal data retrieval;

[0155] Index key generation:

[0156] ;

[0157] in, To synchronize timestamps, This refers to the device's three-dimensional coordinates in the global coordinate system. This is a four-dimensional index key used to uniquely identify a spatiotemporal data point in an R* tree;

[0158] Combine timestamps and global coordinates as a multidimensional index key;

[0159] R* tree insertion operation:

[0160] ;

[0161] in, To associate device / sensor metadata, It is a multidimensional index key, containing a timestamp and three-dimensional coordinates. This is the insertion method for R* trees, used to insert key-value pairs. Insert it into the appropriate position in the tree;

[0162] Range query optimization:

[0163] ;

[0164] in, To query a specific time point, To query the start time, The query end time. It is a spatial position vector. A spatial cubic region. This represents the total number of index nodes. For query time complexity, For range query conditions, time and space query ranges are defined to retrieve data points in the R* tree that meet the conditions;

[0165] S300, Digital Twin Disturbance Modeling: Based on fused data, a vibration-sensitive model (β parameter) and an environmental disturbance model (α parameter) are established. The least squares method is used to fit the equations, the model accuracy is verified online, and the pose drift prediction equation is output.

[0166] In one embodiment of the present invention, the following steps are specifically included:

[0167] S310, Vibration-pose coupling modeling: Analyze the quantitative relationship between vibration spectrum and equipment pose drift, and establish a frequency sensitivity matrix;

[0168] formula:

[0169] Vibration-induced pose drift:

[0170] ;

[0171] ;

[0172] ;

[0173] in, Let be the pose drift vector caused by vibration. This is the frequency sensitivity matrix. The vibration spectral density vector. This represents the power spectral density value at the first frequency point. This is the power spectral density value at the second frequency point. Let K be the power spectral density value at the Kth frequency point. This is a transpose operation;

[0174] Frequency sensitivity matrix:

[0175] ;

[0176] in, Let be the sensitivity coefficient of the vibration in the j-th frequency band to the drift along the i-th axis, for example, This represents the sensitivity coefficient of the first frequency band vibration to the x-axis drift. This represents the sensitivity coefficient of the second frequency band vibration to the x-axis drift. Let be the sensitivity coefficient of the K-th frequency band vibration to the x-axis drift. This represents the sensitivity coefficient of the vibration in the first frequency band to the y-axis drift. This represents the sensitivity coefficient of the second frequency band vibration to the y-axis drift. Let be the sensitivity coefficient of the K-th frequency band vibration to the y-axis drift. This represents the sensitivity coefficient of the first frequency band vibration to the z-axis drift. This represents the sensitivity coefficient of the second frequency band vibration to the z-axis drift. Let be the sensitivity coefficient of the K-th frequency band vibration to the z-axis drift. For the dimension of the spectrum vector, This refers to the device's three-dimensional coordinates in the global coordinate system. This represents the change in position components along the x, y, and z axes in the global coordinate system. For change operators;

[0177] S320, Environmental-Sensor Error Modeling: Quantifying the impact of environmental parameters on positioning / measurement errors and constructing a temperature, humidity and gas composite interference model;

[0178] formula:

[0179] Comprehensive environmental error:

[0180] ;

[0181] ;

[0182] ;

[0183] in, The error is due to a combination of environmental factors. For temperature deviation, The current temperature. For reference temperature, For humidity deviation, The current humidity. For reference humidity, Gas / carbon monoxide concentration, The temperature interference coefficient is... Humidity interference coefficient, The gas concentration interference coefficient, The interference coefficient for carbon monoxide concentration;

[0184] S330, comprehensive prediction of pose drift: integrates vibration and environmental factors to predict equipment pose changes and constructs a state-space model;

[0185] formula:

[0186] State-space equations:

[0187] ;

[0188] ;

[0189] in, This is the device state vector (position + velocity). Let be the state vector of the device at the next moment. This refers to the device's three-dimensional coordinates in the global coordinate system. The velocity components in the global coordinate system. This is the state transition matrix (describing the kinematic characteristics of the equipment). The vibration disturbance input matrix is... The input matrix is ​​the environmental disturbance. The noise is Gaussian process noise (mean 0, covariance Q, unit consistent with the state vector). For discrete-time indexing;

[0190] S340, Perturbation Parameter Identification (OLS-Ridge Regression): Training Sensitivity Coefficients Based on Historical Data and interference coefficient ;

[0191] formula:

[0192] Optimize the objective function:

[0193] ;

[0194] in, To measure pose drift, To account for actual environmental errors, For the spectral density matrix, For the environmental parameter matrix, Let be the vibration sensitivity coefficient vector. This is the vector of environmental interference coefficients. The regularization coefficient is . It is the Euclidean norm;

[0195] S350, Model Validation and Confidence Assessment: Calculate prediction residuals and confidence intervals, and dynamically update model parameters;

[0196] formula:

[0197] Predicted residuals:

[0198] ;

[0199] 95% confidence interval:

[0200] ;

[0201] in, Predict the residual at time t. For the average residual, The standard deviation of the residuals. To verify the sample size, To predict drift for the model, For actual drift measurement, This is the critical value for the standard normal distribution at a 95% confidence level. The confidence interval is 95%.

[0202] S400, Dynamic Feedforward Compensation Control: Predicts pose drift based on disturbance model, generates control signal by combining feedforward-PID composite controller, sends control commands to actuator in real time, and verifies compensation effect through measured data (calculates drift suppression rate η and residual energy ratio r).

[0203] In one embodiment of the present invention, the following steps are specifically included:

[0204] S410, disturbance prediction calculation, predicts pose drift based on the state-space model of S330, and integrates vibration and environmental disturbance factors;

[0205] formula:

[0206] Perturbation prediction equation:

[0207] ;

[0208] ;

[0209] ;

[0210] ;

[0211] in, To predict pose drift, The perturbation input matrix is... This is the vibration sensitivity coefficient matrix. The vibration spectrum, The environmental interference coefficient is... For the environmental parameter vector ( (Source: S230, units are ℃, %, vol%, ppm respectively). denoted as the predicted pose drift in the x, y, and z directions, respectively. This represents the power spectral density value at the first frequency point. This is the power spectral density value at the second frequency point. Let K be the power spectral density value at the Kth frequency point. For transpose operation, Let be the sensitivity coefficient of the vibration in the j-th frequency band to the drift along the i-th axis;

[0212] S420, Compensation Control Quantity Generation: Design a PID controller to generate compensation instructions, and introduce feedforward compensation to offset prediction drift;

[0213] formula:

[0214] PID-feedforward composite controller:

[0215] ;

[0216] ;

[0217] ;

[0218] in, To compensate for the control quantity, For pose tracking error, For reference trajectory, For global pose, This is the PID gain matrix. This is the feedforward compensation gain matrix. To control the cycle, These are the control quantities in the x, y, and z directions, respectively;

[0219] S430, multi-objective collaborative optimization: resolves compensation conflicts among multiple devices and meets safety and energy consumption constraints;

[0220] formula:

[0221] Optimize the objective function:

[0222] ;

[0223] Constraints:

[0224] ;

[0225] in, For the target compensation amount, Energy consumption weighting coefficient This is the minimum safe distance between equipment. The maximum control threshold. Let i and j be the position vectors of devices. This is the kth control component;

[0226] S440, Compensation Command Issuance: Commands are issued to the device PLC via the OPCUA protocol. The command format is standardized, and the transmission delay requirement is [missing information]. ,in This refers to the transmission delay time.

[0227] formula:

[0228] Command format:

[0229] ;

[0230] Transmission delay constraints:

[0231] ;

[0232] in, To standardize the instruction format, It is a unique identifier for the instruction. For command type enumeration values ​​(compensation / shutdown / ventilation, etc.). For control vectors, To synchronize timestamps, For PLC to receive timestamps, Send timestamps to the system. This refers to the transmission delay time.

[0233] OPCUA communication parameters:

[0234] ;

[0235] in, Command issuance frequency, representing the number of commands sent per second. For service quality levels, 1 indicates a "at least one" delivery guarantee. This is the communication timeout period; communication is considered to have failed if this timeout is exceeded.

[0236] S450, Compensation effect verification: Compare the pose error before and after compensation, and calculate key performance indicators;

[0237] formula:

[0238] Drift suppression rate:

[0239] ;

[0240] Residual energy ratio:

[0241] ;

[0242] in, For uncompensated measured drift, To compensate for the measured drift, Drift suppression rate (expressed as a percentage). The residual energy ratio;

[0243] S500, edge intelligent decision-making: evaluate compensation performance indicators (η,r), dynamically trigger model updates (RLS algorithm to optimize β,α) or parameter optimization (PSO to adjust PID gain), and execute graded responses (shutdown / ventilation) based on comprehensive safety situation (methane concentration, vibration peak), and synchronize the decision to the digital twin platform.

[0244] In one embodiment of the present invention, the following steps are specifically included:

[0245] S510, Dynamic evaluation of compensation effect: Calculate key performance indicators (KPIs) based on the verification results of S450, and trigger model updates and parameter optimization thresholds;

[0246] formula:

[0247] Drift suppression ratio to residual energy ratio:

[0248] ;

[0249] Where η is the drift suppression rate, representing the percentage of compensation effect, and r is the residual energy ratio, representing the ratio of position error to the reference trajectory. For uncompensated measured drift, To compensate for the measured drift, For device global coordinates, For reference trajectory coordinates, Let be the vector norm, representing the magnitude of the vector. is the Euclidean norm, representing the 2-norm of a vector;

[0250] Decision-making logic:

[0251] ;

[0252] S520, Online Update of Disturbance Model: The vibration sensitivity coefficient is updated in real time using the recursive least squares (RLS) method. and environmental interference coefficient ;

[0253] formula:

[0254] Recursive least squares update equation:

[0255] ;

[0256] ;

[0257] ;

[0258] ;

[0259] ;

[0260] in, This is the current parameter vector, which includes the vibration sensitivity coefficient and the environmental disturbance coefficient. This is the parameter vector from the previous time step. This is the Kalman gain matrix, used to update the parameter estimates. The characteristic matrix contains vibration and environmental parameters (from S230-S240). Let be the observation vector at the current moment. This is the forgetting factor, used to adjust the weight of historical data (default 0.98). Let be the current covariance matrix, representing the uncertainty of the parameter estimation. Estimate the covariance matrix of the parameters from the previous time step, where the initial values ​​are... , It is the identity matrix. For time step index, The vibration sensitivity coefficient, Environmental interference coefficient;

[0261] S530, Control Parameter Co-optimization: PID gain is optimized using Particle Swarm Optimization (PSO) algorithm. and feedforward gain ;

[0262] Optimization goal:

[0263] ;

[0264] ;

[0265] Constraints:

[0266] ;

[0267] in, The objective function value represents the control performance evaluation index. To track errors, To control the quantity, To control energy consumption weight, For PID gain constraint boundaries, The maximum control threshold. To optimize the time window length, The infinity norm represents the absolute value of the largest component of the vector. For the proportional, integral, and derivative gains of the PID controller;

[0268] S540, Safety Situation Decision: Executes safety strategies based on integrated environmental and vibration data, and generates equipment-level / system-level emergency commands;

[0269] Decision-making rules:

[0270] ;

[0271] in, This refers to the methane concentration. The peak value of the vibration spectrum. For wind speed, Let be the power spectral density at the k-th frequency point. Frequency point index;

[0272] S550, Digital Twin Synchronization and Logging: Synchronizes decision results to the digital twin platform, generating encrypted decision logs and security requirements. Log integrity is verified using SHA-256 hash checksums, and the synchronization latency is [not specified]. , This is the synchronization delay time;

[0273] Log hash verification:

[0274] ;

[0275] in, The generated hash value (256 bits). For the decision parameter set ( wait), To synchronize timestamps, For data concatenation, Use the SHA-256 hash function;

[0276] Incremental synchronization protocol:

[0277] ;

[0278] in, For incremental data packets, For parameter identifiers, For the new value of the parameter, The old value of the parameter. For parameter index

[0279] Synchronization period constraint:

[0280] ;

[0281] in, For synchronization period, For model update frequency, The maximum synchronization period threshold. The function is for finding the minimum value;

[0282] S600, control command output: reconstruct control signals (embedded with optimized parameters and security commands), perform encryption encoding (AES-256+CRC32 check) and timestamp synchronization, transmit commands according to priority through real-time network (TSN), receive execution feedback and calculate control error and delay to form closed-loop verification;

[0283] In one embodiment of the present invention, the following steps are specifically included:

[0284] S610, Control Signal Reconstruction: The control signal is generated by integrating the optimized control parameters of S530 with the real-time pose error, and feedforward compensation is embedded.

[0285] formula:

[0286] PID controller and feedforward compensation:

[0287] ;

[0288] in, To output the control vector, The optimized PID gain matrix (from S530). To optimize the feedforward gain matrix (from S530). For real-time pose error, To predict pose drift, To control the cycle, For the error integral term, This is the error differential term;

[0289] S620, security instruction embedding: Based on the security decision overlay control signal of S540, generate device-level security linkage instructions;

[0290] Logical AND parameters:

[0291] Safety control vector:

[0292] ;

[0293] in, For safety control vectors, Ventilation control gain (default 0.5). This refers to the methane concentration. For the original control vector, Zero control vector, used for emergency shutdown. To maximize the function, ensure that the ventilation control quantity is non-negative;

[0294] S630, Signal Security Encoding: Encrypts and verifies the integrity of control signals, and adds timestamps for synchronization;

[0295] Protocol and parameters:

[0296] Encryption and Verification:

[0297] ;

[0298] ;

[0299] in, This is the encrypted control vector (binary data). For dynamic session keys (updated every hour, 256 bits). For spatiotemporally aligned timestamps, For data concatenation, It uses the AES encryption function with a 256-bit key. This is a 32-bit cyclic redundancy check function;

[0300] S640, Real-time transmission scheduling: Transmits encrypted commands via TSN (Time-Sensitive Networking) and dynamically allocates transmission priorities;

[0301] Scheduling strategy:

[0302] ;

[0303] Among them, the transmission constraints, maximum delay, and bandwidth allocation are: ;

[0304] in, For the maximum transmission delay, To allocate bandwidth, For data packet length, Transmission priority (7 is the highest);

[0305] S650, Execution Status Feedback: Receives execution confirmation signals from the device PLC and calculates control errors and delays;

[0306] formula:

[0307] Control error and delay calculation:

[0308] ;

[0309] ;

[0310] ;

[0311] in, To control errors, this indicates the deviation in instruction execution. Transmission delay represents the time difference between instruction transmission and execution. For the control vectors issued, For the control quantities that have been confirmed by the equipment, For device response timestamps, To synchronize timestamps, The Euclidean norm is used to calculate the magnitude of the control error. These are the control variables that have been executed in the x, y, and z directions, respectively.

[0312] In one embodiment of the present invention, the following is an example of the operation of a downhole multi-parameter monitoring system;

[0313] The scene is described as follows: In a deep coal mine tunneling face, a tunneling machine is cutting the coal face, and transport vehicles are frequently going back and forth to transport coal slag;

[0314] At this time, equipment vibration, airflow disturbance, and gas release cause drastic fluctuations in environmental parameters, and the system initiates a dynamic compensation monitoring process:

[0315] 1. Real-time monitoring of equipment status

[0316] Tunnel boring machine (TBM) position acquisition: UWB positioning tags capture the TBM's position in real time (102.3m, 45.7m, -15.2m in the tunnel), and LiDAR scans the surface deformation of the equipment to detect the offset of the cutting head;

[0317] Environmental parameter synchronization: The gas sensor detected a sudden increase in CH4 concentration near the cut-off point, and the wind speed sensor showed local airflow stagnation;

[0318] Vibration signal processing: The accelerometer of the cutting section of the tunneling machine collects high-frequency vibrations, and 85Hz is identified as the dominant frequency through spectrum analysis.

[0319] 2. Multi-source data fusion processing

[0320] The spatiotemporal alignment engine unifies the tunneling machine pose, gas concentration, and vibration spectrum into a global coordinate system and associates them with the same timestamp;

[0321] The fusion data revealed that the cutting head position coincided with the high gas concentration area, and the 85Hz vibration intensity was positively correlated with the equipment's positional drift.

[0322] 3. Digital Twin Perturbation Modeling

[0323] Vibration-pose modeling: 85Hz vibration was identified as the main cause of pose drift, and the sensitivity coefficient β was marked as high risk;

[0324] Environmental interference analysis: Excessive gas concentration leads to LiDAR ranging error; dynamic correction of environmental interference coefficient α.

[0325] Model output warning: The cutting head pose will continue to deviate and may collide with the tunnel sidewall.

[0326] 4. Dynamic feedforward compensation control

[0327] Predict the orientation shift of the tunneling machine after 1 second, and generate a reverse compensation force using a feedforward-PID controller;

[0328] Multi-objective optimization module coordinates the ventilation system: increases local fan speed, dilutes methane, and stabilizes airflow;

[0329] The instruction is sent to the tunneling machine PLC: Adjust the angle of the hydraulic outrigger to counteract the positional drift.

[0330] 5. Edge Intelligent Decision Making

[0331] Post-compensation verification: The pose offset suppression rate reached 88%, but the residual energy ratio still exceeded the standard, triggering a model update.

[0332] The RLS algorithm optimizes the vibration sensitivity coefficient β, and the PSO adjusts the PID gain parameters.

[0333] Safety decision: Due to the continuous increase in gas concentration, the "enhanced ventilation" command is activated, and the standby fan is activated.

[0334] 6. Control command output

[0335] Reconstruct the control signal: embed the optimized PID gain and ventilation command;

[0336] AES-256 encrypted commands are transmitted to the device with high priority via the TSN network;

[0337] Execution feedback: The tunneling machine is in a stable position, the ventilation system reduces the gas concentration to a safe threshold, and the transmission delay is only 18ms.

[0338] The above examples demonstrate the following effects achieved through downhole multi-parameter environmental monitoring methods:

[0339] Risk mitigation: Equipment collision was successfully avoided, and the gas concentration decreased by 50%-80% within 10 seconds.

[0340] System Adaptation: After the model is updated online, the efficiency of subsequent vibration disturbance compensation is improved by 35%.

[0341] Panoramic visualization: The digital twin platform displays equipment trajectories, risk heat maps, and control command flows in real time.

[0342] The embodiments of the present invention have been described above, but the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention, all of which are within the protection scope of the present invention.

Claims

1. A method for downhole multi-parameter environmental monitoring, characterized in that, Includes the following steps: S100, Real-time Equipment Status Perception: Real-time acquisition of equipment pose, vibration spectrum and environmental parameters through multi-source sensors, spatiotemporal alignment, filtering and noise reduction and feature extraction of raw data, and output of standardized status vector. S100 includes the following steps: S110, Equipment pose acquisition: Real-time capture of the three-dimensional coordinates and attitude angles of the tunneling machine and transport vehicle through UWB base station and tag, LiDAR scanning area point cloud data, dynamic modeling of equipment surface deformation; S120, Environmental parameter acquisition: Synchronously reads data from gas / temperature / humidity / wind speed sensors, and monitors the internal pressure of the air duct by a pressure sensor; S130, Vibration signal acquisition: Deploy accelerometers in key parts of the equipment to acquire raw vibration signals from 0 to 500 Hz; S140, Vibration spectrum calculation: Perform FFT transformation on the original acceleration signal to calculate the power spectral density; S150, Data Encapsulation and Timestamp Marking: Pack pose / environment / spectrum data and add PTP1588v2 protocol timestamp; S200, multi-source data fusion processing: fuses pose data, vibration signals and environmental parameters under a unified spatiotemporal reference to construct a multi-dimensional feature matrix; uses principal component analysis to reduce dimensionality and remove redundant information, and outputs the fused device state vector; S300, Digital Twin Disturbance Modeling: Based on fused data, a vibration-sensitive model and an environmental disturbance model are established. The least squares method is used to fit the equations, the model accuracy is verified online, and the pose drift prediction equation is output. S400, Dynamic Feedforward Compensation Control: Predicts the pose drift based on the disturbance model, generates control signals by combining the feedforward-PID composite controller, sends control commands to the actuator in real time, and verifies the compensation effect through actual measurement data. S500, edge intelligent decision-making: evaluates compensation performance indicators, dynamically triggers model updates or parameter optimization, performs hierarchical responses based on comprehensive security situation, and synchronizes decisions to the digital twin platform; S600, control command output: reconstruct control signals, perform encryption encoding and timestamp synchronization, transmit commands according to priority through real-time network, receive execution feedback and calculate control error and delay to form closed-loop verification.

2. The downhole multi-parameter environmental monitoring method according to claim 1, characterized in that, S200 includes the following steps: S210, PTP1588v2 time synchronization: Aligns the time base of all devices / sensors through a master-slave clock architecture, calculates clock offset and corrects timestamps; S220, Equipment Coordinate System to Roadway Global Coordinate System Transformation: Transform the equipment pose to the roadway global coordinate system and fuse LiDAR point cloud data to correct positioning errors; S230, Environmental Sensor Spatial Mapping: Calculate global coordinates based on sensor installation location and establish sensor-device spatial relationship matrix; S240, Spatiotemporal correlation of vibration spectrum: Correlate the vibration spectrum density vector to the location of a specific device to establish a frequency-spatial distribution model; S250, Spatiotemporal Index Construction: Establish an R* tree spatiotemporal index to achieve millisecond-level cross-modal data retrieval.

3. The downhole multi-parameter environmental monitoring method according to claim 1, characterized in that, In S300, the following steps are included: S310, Vibration-pose coupling modeling: Analyze the quantitative relationship between vibration spectrum and equipment pose drift, and establish a frequency sensitivity matrix; S320, Environmental-Sensor Error Modeling: Quantifying the impact of environmental parameters on positioning / measurement errors and constructing a temperature, humidity and gas composite interference model; S330, comprehensive prediction of pose drift: integrates vibration and environmental factors to predict equipment pose changes and constructs a state-space model; S340, Perturbation parameter identification: Training sensitivity coefficients and perturbation coefficients based on historical data; S350, Model Validation and Confidence Assessment: Calculate the predicted residuals and confidence intervals, and dynamically update the model parameters.

4. The downhole multi-parameter environmental monitoring method according to claim 3, characterized in that, The calculation formula for vibration-pose coupling modeling is as follows: Vibration-induced pose drift: ； ； ； in, Let be the pose drift vector caused by vibration. This is the frequency sensitivity matrix. The vibration spectral density vector. This represents the power spectral density value at the first frequency point. This is the power spectral density value at the second frequency point. Let K be the power spectral density value at the Kth frequency point. This is a transpose operation; Frequency sensitivity matrix: ； in, Let be the sensitivity coefficient of the vibration in the j-th frequency band to the drift along the i-th axis. This represents the sensitivity coefficient of the first frequency band vibration to the x-axis drift. This represents the sensitivity coefficient of the second frequency band vibration to the x-axis drift. Let be the sensitivity coefficient of the K-th frequency band vibration to the x-axis drift. This represents the sensitivity coefficient of the vibration in the first frequency band to the y-axis drift. This represents the sensitivity coefficient of the second frequency band vibration to the y-axis drift. Let be the sensitivity coefficient of the K-th frequency band vibration to the y-axis drift. This represents the sensitivity coefficient of the first frequency band vibration to the z-axis drift. This represents the sensitivity coefficient of the second frequency band vibration to the z-axis drift. Let be the sensitivity coefficient of the K-th frequency band vibration to the z-axis drift. For the dimension of the spectrum vector, This refers to the device's three-dimensional coordinates in the global coordinate system. This represents the change in position components along the x, y, and z axes in the global coordinate system. For change operators.

5. The downhole multi-parameter environmental monitoring method according to claim 4, characterized in that, The calculation formula for environmental-sensor error modeling is as follows: Comprehensive environmental error: ； ； ； in, The error is due to a combination of environmental factors. For temperature deviation, The current temperature. For reference temperature, For humidity deviation, The current humidity. For reference humidity, Gas / carbon monoxide concentration, The temperature interference coefficient is... Humidity interference coefficient, The gas concentration interference coefficient, This represents the interference coefficient for carbon monoxide concentration.

6. The downhole multi-parameter environmental monitoring method according to claim 5, characterized in that, The formula for calculating pose drift prediction is as follows: State-space equations: ； ； in, This is the device state vector. This is the state vector of the device at the next moment. This refers to the device's three-dimensional coordinates in the global coordinate system. The velocity components in the global coordinate system. Here is the state transition matrix. The vibration disturbance input matrix is... The input matrix is ​​the environmental disturbance. For Gaussian process noise, This is a discrete-time index.

7. The downhole multi-parameter environmental monitoring method according to claim 6, characterized in that, In S400, the following steps are included: S410, disturbance prediction calculation, predicts pose drift based on the state-space model of S330, and integrates vibration and environmental disturbance factors; S420, Compensation Control Quantity Generation: Design a PID controller to generate compensation instructions, and introduce feedforward compensation to offset prediction drift; S430, multi-objective collaborative optimization: resolves compensation conflicts among multiple devices and meets safety and energy consumption constraints; S440, compensation command issuance: commands are issued to the device PLC via the OPCUA protocol. The command format is standardized and the transmission delay requirement is <20ms. S450, Compensation Effect Verification: Compare the pose error before and after compensation, and calculate key performance indicators.

8. The downhole multi-parameter environmental monitoring method according to claim 7, characterized in that, In S500, the following steps are included: S510, Dynamic evaluation of compensation effect: Calculate key performance indicators based on the verification results of S450, and trigger model updates and parameter optimization thresholds; S520, Online update of disturbance model: The vibration sensitivity coefficient and environmental disturbance coefficient are updated in real time using the recursive least squares method; S530, Control Parameter Co-optimization: PID gain and feedforward gain are optimized through particle swarm optimization algorithm; S540, Safety Situation Decision: Executes safety strategies based on integrated environmental and vibration data, and generates equipment-level / system-level emergency commands; S550, Digital Twin Synchronization and Logs: Synchronizes decision results to the digital twin platform, generates encrypted decision logs and security requirements, log integrity is verified by SHA-256 hash, and synchronization latency is <50ms.

9. A downhole multi-parameter environmental monitoring method according to claim 8, characterized in that, In S600, the following steps are included: S610, Control Signal Reconstruction: The control signal is generated by integrating the optimized control parameters of S530 with the real-time pose error, and feedforward compensation is embedded. S620, security instruction embedding: Based on the security decision overlay control signal of S540, generate device-level security linkage instructions; S630, Signal Security Encoding: Encrypts and verifies the integrity of control signals, and adds timestamps for synchronization; S640, Real-time Transmission Scheduling: Transmits encrypted commands via TSN and dynamically allocates transmission priorities; S650, Execution Status Feedback: Receives execution confirmation signals from the device PLC and calculates control errors and delays.

Images