Micro-nano geotechnical robot geological disaster multi-source information fusion early warning system

Abstract

The invention discloses a micro-nano rock-soil robot geological disaster multi-source information fusion early warning system which is characterized by comprising a micro-nano rock-soil robot cluster, a wireless energy supply system, electromagnetic modified micro-nano particles, a data transmission and control module, an injection deployment system, a main control unit, an information fusion analysis module and an early warning module. The early warning system can realize multi-dimensional and multi-scale information monitoring of geological disasters, has high timeliness, accuracy and reliability, is suitable for application scenes of landslide, slope collapse, tunnel deformation, tailing dam safety monitoring and the like, and provides technical support for early warning and risk assessment of geological disasters.

Description

Technical Field

[0001] The present invention belongs to the technical field of the combination of geological information monitoring, disaster information monitoring and early warning, and machine learning, and particularly relates to a multi-source information fusion early warning system for geological disasters based on a wireless micro-nano multi-sensor cluster. Background Art

[0002] Geological disasters include collapses, landslides, debris flows, ground collapses, ground fissures, and land subsidence; specifically, geological disasters induced by coal mines, leakage of underground carbon dioxide storage, induced earthquakes in deep well wastewater injection projects, deterioration of surrounding rocks during deep salt cavern energy storage, deterioration of the physical and mechanical properties of geological bodies due to chemical reactions between energy storage media and geological bodies, micro-seismicity caused by underground fluid injection and production activities, earthquake-induced disasters, and instability of tailings dams, etc., which often have characteristics such as suddenness, concealment, and complex disaster chains. Realizing real-time monitoring and early warning of geological disasters is an important research direction in the current field of disaster prevention and mitigation.

[0003] Current geological disaster monitoring technologies mainly include remote sensing technology, surface sensor monitoring, and geophysical exploration technology. However, the application of these technologies in the monitoring of deep geological disasters has the following problems: only limited surface or shallow information can be obtained, and real-time monitoring of key parameters such as deep fracture structures, fluid pressures, and stress changes cannot be achieved; data transmission of traditional monitoring systems is easily interfered by complex geological environments, and signal reliability is poor; monitoring equipment is difficult to operate stably for a long time under extreme conditions, which limits the comprehensiveness and accuracy of geological disaster information collection.

[0004] The emergence of wireless sensor networks provides a new approach for geological disaster monitoring. By deploying distributed sensor nodes in the target monitoring area, dynamic data collection and monitoring of the geological environment can be realized. In the current nanotechnology for geological information monitoring, micro-nano sensors are put into the ground through drilling, but the micro-nano sensors remain on the inner wall of the drilling or are distributed around the drilling. Compared with the deep formation, only single-point information can be monitored. At the same time, the drilling will interfere with the surrounding geological environment, and the geological data monitored by this method will also be greatly affected.

[0005] Chinese Patent CN202410680445.2 discloses an in-situ monitoring system for oil and gas reservoir information based on a wireless underground sensor network, which provides a solution for data transmission and processing of the sensor network, but its application scope is limited and it cannot meet the deep real-time monitoring requirements of multi-source and multi-scale geological information.

[0006] Therefore, it is necessary to propose a multi-source information fusion early warning system for geological disasters based on a wireless micro-nano multi-sensor cluster to overcome the deficiencies of the prior art. Summary of the Invention

[0007] In view of the above technical problems, the present invention proposes a multi-source information fusion early warning system for geological disasters based on a wireless micro-nano multi-sensor cluster, which is used to solve the problems existing in the existing geological disaster monitoring systems, such as poor real-time performance, limited data transmission, and insufficient multi-source data fusion ability. Through the collaborative action of multi-functional modules, efficient, accurate, and intelligent geological disaster monitoring and early warning are realized.

[0008] The technical solution provided by the present invention is as follows:

[0009] A multi-source information fusion early warning system for geological disasters of micro-nano geotechnical robots, comprising: a micro-nano geotechnical robot cluster, a wireless power supply system, electromagnetic modified micro-nano particles, a data transmission and control module, an injection and deployment system, a main control unit, an information fusion and analysis module, and an early warning module;

[0010] The micro-nano geotechnical robot cluster is used to collect multi-source geological information of the target monitoring area and obtain multi-sensor cluster data;

[0011] The wireless power supply system forms a stable electromagnetic field in the target monitoring area through an electromagnetic excitation device to provide continuous energy for the micro-nano geotechnical robot cluster;

[0012] The electromagnetic modified micro-nano particles are used to enhance the electromagnetic properties of the fracture area, improve the wireless power supply efficiency, and optimize the visual modeling of the fracture distribution as a tracer combined with electromagnetic exploration means;

[0013] The injection and deployment system is used to accurately inject the micro-nano geotechnical robot cluster and the electromagnetic modified micro-nano particles into the target monitoring area;

[0014] The data transmission and control module includes a first communication module, a signal enhancement module, and a first control module; wherein, the first communication module is used to realize information exchange between the micro-nano geotechnical robot cluster and the main control unit through wireless communication, the signal enhancement module is used to enhance the wireless signal, and the second control module is used to control the operation of the wireless power supply system;

[0015] The main control unit is used to analyze and process the multi-sensor cluster data to obtain first data;

[0016] The information fusion and analysis module is used to perform spatio-temporal alignment and analysis on the first data to obtain second data, and monitor structural deformation, crack changes, chemical composition changes, and geological safety status;

[0017] The early warning module is based on a machine learning model, used to perform time series analysis on the second data and model spatio-temporal information by combining deep learning algorithms to predict disaster trends; at the same time, potential geological anomalies are identified through an anomaly detection algorithm to generate disaster risk assessment and early warning information.

[0018] In a possible implementation, the micro-nano geotechnical robot in the micro-nano geotechnical robot cluster includes a housing and the following components disposed within the housing:

[0019] A micro-nano sensor for monitoring changes in geological information in the target monitoring area;

[0020] A battery for supplying electrical energy;

[0021] A second communication module, including a transmitting antenna and a receiving antenna, for realizing information exchange with the main control unit;

[0022] A second control module for controlling the working states of the micro-nano sensor and the communication module, storing the collected geological information data and the corresponding time, and performing signal conditioning and data compensation on the micro-nano sensor.

[0023] Furthermore, the micro-nano sensor includes an inertial sensor and a monitoring sensor;

[0024] The monitoring sensor includes a temperature sensor, a stress sensor, a fluid pressure sensor, a microbial sensor, an ion concentration sensor, and a pH sensor.

[0025] Furthermore, the battery is an electret and is charged by the excitation of the energy supply unit.

[0026] Furthermore, the housing includes a waterproof layer from the inside out and a shell layer with surface modification treatment; the waterproof layer is made of a polymer coating; the method for obtaining the shell layer with surface modification treatment is: mixing the micro-nano geotechnical robot, negatively charged graphene, and a coupling agent at high speed to enhance the electrostatic repulsion between the micro-nano geotechnical robots.

[0027] In a possible implementation, the injection and deployment system includes an injection device, a fluid delivery system, and a pressure regulation unit;

[0028] The injection device is used to inject the micro-nano geotechnical robot cluster and the electromagnetic modified micro-nano particles into the target monitoring area through a borehole, a fracture, or a specific injection channel;

[0029] The fluid delivery system is used to input the micro-nano geotechnical robot cluster and the electromagnetic modified micro-nano particles into the target area;

[0030] The pressure regulation unit is used to adjust the injection pressure and dynamically regulate the injection flow rate according to the physical properties of the target formation.

[0031] In a possible implementation, the wireless power supply system uses a DC power supply or an AC power supply. The two poles of the power supply are respectively placed between the strata in the target monitoring area, and an electric field is excited in the target monitoring area at intervals less than the limit working duration of the micro-nano geotechnical robot to supply power to the micro-nano geotechnical robot cluster.

[0032] In a possible implementation, the electromagnetic modified micro-nano particles are obtained by coating a carbon-based conductive layer on the surfaces of ferrite particles, cobalt oxide particles and nickel oxide particles.

[0033] In a possible implementation, the information fusion analysis module includes a data processing unit, an anomaly detection unit, a feature extraction unit and a multi-modal data fusion unit;

[0034] The data processing unit is used to sort, correct and supplement the first data based on the time stamp and the sensor identifier, including: calibrating the time stamp of the first data by using the Kalman filter and the time interpolation algorithm to ensure the timing consistency of the asynchronous sensor data; dynamically adjusting the noise suppression strategy by combining the Kalman filter and the low-pass filter technology to remove the noise in the first data; predicting the missing data by using the machine learning algorithm and filling the missing data;

[0035] The anomaly detection unit is used to intelligently identify complex anomaly patterns by using the isolation forest and DBSCAN clustering methods to discover potential geological disaster signals;

[0036] The multi-modal data fusion unit is used to synthesize the data processed by the data processing unit and the feature data in the geological disaster signal to obtain the second data.

[0037] In a possible implementation, the early warning module includes a time series prediction unit, an adaptive early warning strategy unit, a remote monitoring and feedback unit;

[0038] The time series prediction unit is used to perform time series analysis based on the spatio-temporal information data in the second data by using the deep learning method to obtain a geological disaster prediction model; using the geological disaster prediction model to analyze the historical change trends of geological parameters such as ground stress and crack propagation, and predicting the occurrence trend and evolution path of geological disasters;

[0039] The adaptive early warning strategy unit is used to calculate the comprehensive risk index of geological disasters by combining multiple data sources including ground stress, temperature change, fluid pressure, and chemical composition fluctuation, and to evaluate the affected range of the crack propagation and formation instability areas; detecting sudden geological anomaly events by using the clustering algorithm and determining whether the anomaly events are triggered by external environmental factors;

[0040] The remote monitoring and feedback unit is used to set multi-level abnormal alarm thresholds and provide real-time monitoring data, early warning history records, visual risk maps, and remote access.

[0041] The beneficial effects of the present invention are as follows:

[0042] The multi-source geological information in-situ intelligent perception system based on micro-nano geotechnical robots provided by the present invention records the movement trajectories and collected geological information of the micro-nano geotechnical robots in the target monitoring area by circulating injection and extraction of micro-nano geotechnical robots, and the geological information data of the target monitoring area is output after data processing by the main control unit, information fusion analysis module, and early warning module. The system of the present invention effectively overcomes the problem of limited size of underground sensors by specializing the functions of micro-nano geotechnical robots and expands the application scenarios. The system can realize multi-dimensional and multi-scale information monitoring of geological disasters, has high timeliness, accuracy, and reliability, is applicable to application scenarios such as landslides, slope collapses, tunnel deformations, and tailings dam safety monitoring, and provides technical support for the early warning and risk assessment of geological disasters. Brief Description of the Drawings

[0043] The following introduces the prior art using drawings. The drawings in the following description are some examples of the present invention, and other drawings can be obtained based on these drawings without creative labor.

[0044] Figure 1 is a schematic structural diagram of a multi-source information fusion early warning system for geological disasters based on a micro-nano multi-sensor cluster;

[0045] Figure 2 is a schematic composition diagram of a micro-nano geotechnical robot and an electromagnetically modified micro-nano particle;

[0046] Figure 3 is a flowchart of data validity processing and abnormal data early warning work in an embodiment of a multi-source information fusion early warning system for geological disasters based on a micro-nano multi-sensor cluster;

[0047] Figure 4 is a schematic working diagram of an outlier identification and processing unit of a multi-source information fusion early warning system for geological disasters of a micro-nano multi-sensor cluster in an example of the present application;

[0048] Figure 5 A schematic working diagram of a multi-source information fusion early warning system for geological disasters of a micro-nano multi-sensor cluster in an example of the present application. Detailed Embodiments

[0049] The following clearly and completely describes the present application in combination with specific embodiments. Based on the embodiments in the present invention, all other embodiments obtained by those of ordinary skill in the art without creative labor belong to the scope of protection of the present invention.

[0050] As Figure 1 shown, it is a multi-source information fusion early warning system for geological disasters based on a micro-nano multi-sensor cluster, including:

[0051] A micro-nano geotechnical robot cluster, a wireless power supply system, electromagnetic modified micro-nano particles, a data transmission and control module, an injection and deployment system, a main control unit, an information fusion analysis module, and an early warning module;

[0052] The micro-nano geotechnical robot cluster is used to collect multi-source geological information of the target monitoring area and obtain multi-sensor cluster data;

[0053] The wireless power supply system forms a stable electromagnetic field in the target monitoring area through an electromagnetic excitation device to provide continuous energy for the micro-nano geotechnical robot cluster;

[0054] The electromagnetic modified micro-nano particles are used to enhance the electromagnetic performance of the fissure area, improve the wireless power supply efficiency, and optimize the visual modeling of the fissure distribution as a tracer combined with electromagnetic exploration means;

[0055] The injection and deployment system is used to accurately inject the micro-nano geotechnical robot cluster and electromagnetic modified micro-nano particles into the target monitoring area;

[0056] The data transmission and control module includes a first communication module, a signal enhancement module, and a first control module; among them, the first communication module is used to realize the information exchange between the micro-nano geotechnical robot cluster and the main control unit through wireless communication, the signal enhancement module is used to enhance the wireless signal, and the second control module is used to control the operation of the wireless power supply system;

[0057] The main control unit is used to verify the timestamp, sensor ID, and packet integrity of the multi-sensor cluster data to obtain the first data;

[0058] The information fusion analysis module is used to perform spatio-temporal alignment and analysis on the first data to obtain the second data, and monitor the structural deformation, crack change, chemical composition change (detect leakage), and geological safety status;

[0059] The early warning system is based on a machine learning model, used to perform time series analysis on the multi-sensor cluster data and model the spatio-temporal information combined with a deep learning algorithm to predict the disaster trend; at the same time, potential geological anomalies are identified through an anomaly detection algorithm to generate disaster risk assessment and early warning information.

[0060] In a possible implementation, see Figure 2 , the micro-nano geotechnical robots in the micro-nano geotechnical robot cluster 11 include a housing and the following placed in the housing:

[0061] A micro-nano sensor for monitoring changes in geological information in a target monitoring area;

[0062] A power supply for supplying electrical energy;

[0063] A second communication module, including a transmitting antenna and a receiving antenna, for realizing information exchange with the main control unit and supporting low-power communication protocols such as LoRa and UWB;

[0064] A second control module for controlling the working states of the micro-nano sensor and the communication module, storing the collected geological information data and the corresponding time, and performing signal conditioning and data compensation on the micro-nano sensor.

[0065] Further, the micro-nano sensor includes an inertial sensor and a monitoring sensor;

[0066] The monitoring sensor includes a temperature sensor, a stress sensor, a fluid pressure sensor, a microbial sensor, an ion concentration sensor, and a pH sensor.

[0067] Furthermore, the inertial sensor includes an accelerometer and a gyroscope for recording the running trajectory of the micro-nano geotechnical robot.

[0068] Further, both the transmitting antenna and the receiving antenna of the second communication module use embedded high-frequency micro-antennas.

[0069] It can be understood that the monitoring sensor is configured according to specific monitoring requirements, and thus the size of the micro-nano geotechnical robot can be ensured to meet the limitations of the crack size through high specialization of functions.

[0070] Further, the power supply is an electret, which is charged by the excitation of the energy supply unit.

[0071] Specifically, the material of the electret is silicon dioxide.

[0072] Further, the outer shell includes a waterproof layer from the inside out and a shell layer with surface modification treatment; the waterproof layer is made of a polymer coating or silica gel; the method for obtaining the shell layer with surface modification treatment is: mixing the outer shell, negatively charged graphene, and a coupling agent at high speed to increase the electrostatic repulsion force between micro-nano geotechnical robots, prevent agglomeration, and prevent adsorption on the surface of negatively charged rock cracks.

[0073] Specifically, the polymer coating includes polyimide, parylene, and epoxy resin.

[0074] Specifically, the shell layer is a carbon nanotube or a boron nitride nanotube.

[0075] Specifically, the coupling agent is 3-aminopropyltriethoxysilane.

[0076] It should be noted that considering that inertial sensors are prone to error accumulation during operation, which affects the discrimination of the motion trajectory of the micro-nano geotechnical robot, a trajectory correction algorithm is preset in the second control module.

[0077] To ensure the accurate positioning and long-term operation of the micro-nano geotechnical robot, the inertial sensor and the control module are always in working condition, and the monitoring sensor works at a certain time interval according to the preset value of the control module. According to the preset time interval, the ultimate working duration of the micro-nano geotechnical robot is measured in an indoor simulation environment.

[0078] In a possible implementation manner, the injection device in the injection deployment system includes an injection device, a fluid delivery system, and a pressure regulation unit, which are used to precisely control the distribution of the micro-nano geotechnical robot and the electromagnetic modified micro-nano particles in the target monitoring area, and ensure the uniformity, stability, and long-term operation ability of the robot network;

[0079] The injection device (such as a fluid pump) is used to inject the micro-nano geotechnical robot cluster and the electromagnetic modified micro-nano particles into the target monitoring area through a borehole, a fracture, or a specific injection channel;

[0080] The fluid delivery system (such as an injection-production well and a pipeline, etc.) is used to input the micro-nano geotechnical robot cluster and the electromagnetic modified micro-nano particles into the target area;

[0081] The pressure regulation unit is used to adjust the injection pressure and dynamically regulate the injection flow rate according to the physical properties of the target formation, and ensure that the robots and particles are distributed along the set path.

[0082] It should be noted that before injecting the micro-nano geotechnical robot cluster and the electromagnetic modified micro-nano particles, artificial fractures should be created by means of hydraulic fracturing, supercritical carbon dioxide fracturing, etc. to form a migration channel. The electromagnetic modified micro-nano particles should be injected underground prior to the micro-nano geotechnical robot cluster.

[0083] In a possible implementation manner, the power supply unit uses a DC power supply or an AC power supply, and the two poles of the power supply are respectively placed between the formations in the target monitoring area, and an electric field is excited in the target monitoring area at an interval less than the ultimate working duration of the micro-nano geotechnical robot to supply energy to the micro-nano geotechnical robot cluster.

[0084] Specifically, the two poles of the power supply are respectively placed inside the wellbore walls of the injection well and the production well located in the formation of the target monitoring area.

[0085] In a possible implementation manner, the electromagnetic modified micro-nano particles are formed by doping ferrite particles (Fe 3 O 4 ) and cobalt ferrite particles (CoFe 2 O4 ) The electromagnetic modified micro-nano particles are obtained by coating a conductive coating on the surface of nickel ferrite particles (NiFe 2 O 4 ). By changing the permittivity, permeability and conductivity in the fissures, the electromagnetic modified micro-nano particles improve the propagation quality of wireless signals, form an electromagnetic field regulation network, increase the propagation distance of signals in the fissures, and reduce signal attenuation.

[0086] Specifically, the conductive coating is aluminum zinc oxide, polyaniline or a carbon-based conductive coating (such as graphene).

[0087] In a possible implementation manner, the first communication module includes a multi-band antenna array and a dynamic frequency adjustment module. The multi-band antenna array is used to reflect and receive wireless signals to achieve signal interaction with the second communication module. The dynamic adjustment module is used to automatically select the best operating frequency band according to the real-time detected signal quality to avoid interference and improve signal stability. At the same time, it automatically adjusts the frequency and signal strength in a high-mineralization or high-humidity environment to ensure stable signal transmission.

[0088] In a possible implementation manner, the signal enhancement module includes a wireless repeater, a signal amplifier, and a signal preprocessing and error correction module. The signal enhancement module enhances the signal transmission ability in the deep underground environment by gradiently arranging wireless repeaters and signal amplifiers. The repeater gradually amplifies the signal to ensure stable data transmission, and the amplifier further increases the signal strength and reduces signal attenuation when passing through different geological layers. The signal preprocessing and error correction module uses signal processing algorithms such as Kalman filters and wavelet transforms to remove noise, correct errors and enhance signals to ensure high-quality data transmission.

[0089] In a possible implementation manner, the information fusion and analysis module includes a data processing unit, an anomaly detection unit, and a multi-modal data fusion unit;

[0090] The data processing unit is used to sort, correct and supplement the first data based on timestamps and sensor identifiers, including: preliminary screening, format conversion and data integrity check to avoid data packet loss or repeated transmission and ensure the reliability of subsequent analysis; calibrating the timestamps of the first data using Kalman filtering and time interpolation algorithms to ensure the temporal consistency of asynchronous sensor data, solving the time synchronization and spatial position calibration problems of different sensor data, and ensuring that the data is analyzed under a unified spatio-temporal reference; dynamically adjusting the noise suppression strategy by combining Kalman filtering and low-pass filtering techniques to remove noise in the first data and improve the accuracy of monitoring data; predicting missing data using machine learning algorithms (such as LSTM, random forest) and filling in the missing data;

[0091] The anomaly detection unit is used to intelligently identify complex anomaly patterns by using the Isolation Forest and DBSCAN clustering methods, discover potential geological disaster signals, identify and extract geological disaster feature data in the potential geological disaster signals, and improve the model analysis efficiency.

[0092] The multi-modal data fusion unit is used to integrate the data processed by the comprehensive data processing unit and the geological disaster feature data to obtain the second data, so as to improve the accuracy and integrity of subsequent data analysis.

[0093] In a possible implementation manner, the early warning module includes a time series prediction unit, an adaptive early warning strategy unit, and a remote monitoring and feedback unit.

[0094] The time series prediction unit is used to adopt a deep learning method (such as Transformer-GRU), perform time series modeling based on the spatio-temporal information data in the second data to obtain a geological disaster prediction model, and use the geological disaster prediction model to analyze the historical change trends of geological parameters such as ground stress and fracture propagation, and predict the occurrence trend and evolution path of geological disasters.

[0095] The adaptive early warning strategy unit is used to combine multiple data sources including ground stress, temperature change, fluid pressure, and chemical composition fluctuation, calculate the comprehensive risk index of geological disasters, and evaluate the affected range of fracture propagation and formation instability regions. Detect sudden geological anomaly events through clustering algorithms, such as rapid fracture propagation and sudden chemical leakage, and determine whether the anomaly events are triggered by external environmental factors (such as rainfall, earthquake, etc.).

[0096] The remote monitoring and feedback unit is used to set multi-level anomaly alarm thresholds to ensure that minor anomalies do not cause false alarms, while significant anomalies can trigger emergency warnings, and provide real-time monitoring data, early warning history records, and remote access.

[0097] Furthermore, the method for modeling by the time series prediction unit includes using a hybrid time series modeling method based on Transformer-GRU. This method combines the advantages of the Transformer model in capturing long sequence dependencies with the advantages of GRU (Gated Recurrent Unit) in processing non-linear time series data. Through this modeling method, the time series and evolution laws of geological disaster data can be captured more accurately, thereby improving the prediction accuracy.

[0098] Furthermore, the clustering algorithms in the adaptive early warning strategy unit include K-means, DBSCAN, and hierarchical clustering, etc. By performing clustering analysis on the abnormal change patterns in geological data.

[0099] The following uses specific embodiments to illustrate the micro-nano geotechnical robot:

[0100] Scenario 1: The micro-nano geotechnical robot is applied to the dry fracture environment at a depth of 100 m to 500 m underground. The fracture width is 50 - 500 μm, mainly composed of granite and basalt, and there is a lack of groundwater. The monitoring targets are ground stress changes, fracture propagation, and rock microseisms. The waterproof layer of the robot is coated with a polyimide (PI) coating, which enhances the stability of the robot in high-temperature environments. Surface modification uses 3-aminopropyltriethoxysilane (APTES) coupling agent to make the robot surface positively charged, thereby improving its dispersibility in fractures, preventing agglomeration, and increasing the electrostatic repulsion force by mixing negatively charged graphene at high speed to prevent the robot from being adsorbed by the fracture surface. To enhance the stability of signal propagation and power supply paths, magnetic particles (ferrite particles Fe 3 O 4 ) are incorporated. The introduction of magnetic particles improves the positioning and energy transmission efficiency of the robot in complex fractures. The robot is driven by an external alternating magnetic field, sliding or tumbling along the fracture surface. Combining with magnetic micro-nano particles further improves the movement flexibility of the robot in fractures and effectively avoids blockages in the fractures. The power supply method uses an electret wireless charging system, which combines magnetic particles to improve the stability of the power supply path and energy transmission efficiency, thereby reducing the energy loss caused by unstable electric fields between fractures. In terms of data collection and transmission, the built-in storage module of the robot can save long-term data and transmit the data to the ground step by step through the LoRa communication protocol. The magnetic micro-nano particles improve the electromagnetic environment during this process, enhancing the quality and stability of data transmission. Wireless relay nodes are arranged every 30 m - 50 m to form a multi-hop communication network to ensure that signals can effectively penetrate and stably transmit in fractures.

[0101] Scenario 2: The micro-nano geotechnical robot is applied to the high-stress fracture zone area, and the monitoring object is the seismic active zone. The geological characteristics are that the rock layer bears high stress (>10 MPa) and the fracture deformation rate is high. The monitoring targets are rock microseisms, fracture closure / propagation, and seismic precursor stress changes. The waterproof layer of the robot uses an epoxy resin coating, which enhances the wear resistance and impact resistance of the robot. Surface modification uses diisooctyl bis(acetylacetonato)titanium coupling agent to improve the corrosion resistance, and the surface is coated with a carbon nanotube coating to enhance the electromagnetic induction ability, thereby improving the wireless power supply efficiency. Combining with magnetic micro-nano particles (ferrite particles Fe 3 O 4) Further optimized the signal propagation and power supply path. The robot slowly moves and stays at a fixed point in the fracture zone through active vibration drive (piezoelectric vibration + magnetic field assistance), and uses magnetic particles to enhance the electromagnetic response ability, enabling it to operate stably in a high-stress environment. In terms of power supply, the robot is remotely powered by an electromagnetic field and combines an electret to store electrical energy. The addition of magnetic particles improves the power supply efficiency, enabling the robot to operate for a long time in a complex geological environment. In terms of data acquisition and transmission, the robot is equipped with highly sensitive strain sensors (for real-time monitoring of rock stratum stress changes), seismic wave monitoring sensors (for identifying microseismic signals), and temperature sensors (for monitoring local temperature rise in the fracture zone). Wireless communication uses a LoRa + UWB combined communication protocol to ensure efficient data transmission. Optimizing the electromagnetic environment with magnetic particles improves the stability of data transmission. At the same time, by deploying repeaters at intervals of 50 m, an efficient signal transmission link is formed.

[0102] For the data processing and outlier warning process, Figure 3Fully describes the logical process from data reception to early warning information release. This process consists of three main stages, including data reception and integrity check, data preprocessing and analysis, and early warning generation and release; In the data reception stage, the system first obtains multi-source information from the monitoring data transmitted by the micro-nano robots. These data include parameters such as temperature, in-situ stress, fluid pressure, and chemical composition. After being received by the wireless repeater, the data is transmitted to the main control unit. The main control unit ensures the validity of the input data by verifying the timestamp, sensor ID, and packet integrity, and eliminates invalid data or duplicate records. The data that has passed the inspection enters the information fusion analysis module, and the data processing unit performs format standardization, time alignment, and noise reduction on the data. Format standardization ensures that multi-source data can be analyzed uniformly, time alignment solves the problem of inconsistent sampling frequencies of different sensors, and noise reduction uses techniques such as Kalman filtering or wavelet transform to remove environmental noise. The multi-modal data fusion unit uses a deep learning model (such as multi-modal Transformer) to perform fusion analysis on multi-source data, integrating heterogeneous data such as temperature, stress, and chemical composition into a unified analysis framework. Through dynamic time warping (DTW) and multi-modal fusion technology, the system identifies the change patterns in the monitoring area. Subsequently, the time series prediction unit performs time series modeling based on deep learning algorithms (such as Transformer-GRU). Using the model analysis to predict the change trend of geological data, combined with 3D geological modeling technology to dynamically display key geological features such as fracture propagation and pressure change. The adaptive early warning strategy unit generates the risk value of each monitoring area according to the risk calculation model, and classifies the potential threats in the area into three levels: low, medium, and high through the risk grading algorithm. In the early warning generation stage, the adaptive early warning strategy unit creates early warning content according to the risk assessment results, including disaster type, risk level, impact range, and time prediction.

[0103] Figure 4 Shows the data processing unit's process for handling data outliers, from data reception, detection, correction to the final result output. In the outlier detection stage, the system uses machine learning algorithms (such as K-means, DBSCAN, and hierarchical clustering, etc.) to analyze the data. Based on upper and lower limit filtering, data points that exceed the physical parameter range are initially marked; at the same time, the density clustering algorithm (DBSCAN) is used to further identify potential outlier patterns, including sudden outliers or trend changes. After completing the outlier detection, the system enters the outlier correction stage. For isolated outliers, iterative correction is used for correction, and for missing data, it is complemented through a time series model to ensure the spatio-temporal continuity of the data. The corrected data is evaluated for quality in the final stage to ensure that the data after outlier processing has sufficient integrity and consistency. The finally processed data is sent to the downstream module as the input for subsequent geological analysis and disaster prediction.

[0104] As shown Figure 5 in the figure, it is a schematic diagram of the monitoring work of the geological disaster multi-source information fusion early warning system based on the micro-nano multi-sensor cluster provided by the present invention; the micro-nano robots are injected into the fissures or potential risk areas of the target area (such as landslide bodies, tunnel structures, etc.).

[0105] These robots are designed to have multi-functional sensing capabilities and can collect multi-source data of the target area, including:

[0106] Temperature change: The temperature distribution in the fissure is monitored in real time through a MEMS temperature sensor to identify whether there is a thermal anomaly, such as rock friction heating;

[0107] Ground stress: A piezoelectric sensor is used to monitor the stress distribution in the fissure area to identify local stress concentration areas in the rock mass; Fluid pressure: The pressure change of the fluid in the fissure is recorded through a micro pressure sensor to analyze whether there is a risk of hydraulic-induced fissure expansion;

[0108] Chemical leakage: Ion sensors and pH sensors are used to detect chemical composition changes in the fissure, such as groundwater pollution or gas leakage.

[0109] The robots are powered by a wireless power supply unit (such as an electret battery) to provide the energy required for long-term operation. The data collected by the micro-nano robots is transmitted through a built-in second communication module (supporting LoRa and UWB protocols). However, due to the complex signal transmission environment in the fissure area (such as rock absorption and scattering), the role of the relay device is crucial. The relay device sends the transmitted data to the relevant modules on the ground. The main tasks of the relevant modules on the ground are to store, preprocess, and analyze the monitoring data; the analyzed and processed data is input into the early warning module. The early warning module generates a risk map and early warning information and transmits the information to the relevant departments in various ways. For example, acoustic and optical alarm: For high-risk situations, trigger an alarm for emergency notification; Remote notification: Send the early warning information to the user through text messages, APP push, or cloud platform; Monitoring interface: Real-time display the dynamic geological model and risk distribution of the target area on the ground system display screen.

[0110] Furthermore, after long-term operation, the battery power in the wireless micro-nano sensor energy supply unit is exhausted. The main control unit makes an intelligent judgment, issues an instruction to the power supply system, radiates energy to the target monitoring area, and supplies power to the micro-nano geotechnical robot cluster to ensure the continuous operation of the micro-nano geotechnical robot cluster.

[0111] In specific implementation, the working process of the geological disaster multi-source information fusion early warning system based on the micro-nano multi-sensor cluster includes the following steps:

[0112] S701: Conduct on-site surveys of the target area, evaluate the geological environment, including fracture distribution, soil layer composition, and stress state, to provide data support for system deployment.

[0113] S702: Based on the geological survey data, determine the layout positions of the micro-nano geotechnical robot clusters, plan the drilling depths, and select the repeater installation points to ensure signal coverage of the entire monitoring area.

[0114] S703: Through the injection and production equipment, inject the micro-nano robot clusters and ferromagnetic nanoparticles into the target fracture area.

[0115] S704: Deploy the repeaters and wireless power supply systems. The repeaters are responsible for signal reception and forwarding, and the wireless power supply systems provide continuous energy for the micro-nano robots.

[0116] S705: Install the relevant modules on the ground, including data transmission and control modules, main control units, information fusion and analysis modules, and early warning modules.

[0117] S706: The main control unit receives the data transmitted from the micro-nano robot clusters and repeaters and conducts analysis and processing.

[0118] S707: Build a software platform to achieve real-time monitoring and data management.

[0119] S708: The micro-nano robot clusters collect multi-source geological data (temperature, pressure, chemical composition, etc.) from the target area and transmit them to the ground modules through the repeaters.

[0120] S709: The repeaters transmit the signals to the ground to complete the wireless relay of the data.

[0121] S710: The information fusion and analysis modules and early warning modules analyze the multi-source data, generate disaster early warning signals, and support scientific decision-making.

[0122] The above embodiments are only used to illustrate the technical solutions of the present application and are not restrictive. Although the present application has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the purpose and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention. The technologies, shapes, and structures not described in detail in the present invention are all well-known technologies.

Claims

1. The micro-nano geotechnical robot geological disaster multi-source information fusion early warning system is characterized by: It includes a micro-nano geotechnical robot cluster, a wireless energy supply system, electromagnetically modified micro-nano particles, a data transmission and control module, an injection deployment system, a main control unit, an information fusion analysis module, and an early warning module; The micro-nano geotechnical robot cluster is used to collect multi-source geological information of the target monitoring area and obtain multi-sensor cluster data; The wireless energy supply system forms a stable electromagnetic field in the target monitoring area through an electromagnetic excitation device, providing continuous energy for the micro-nano geotechnical robot cluster; The electromagnetically modified micro-nanoparticles are used to enhance the electromagnetic performance of the fracture area, improve the efficiency of wireless energy supply, and serve as tracers combined with electromagnetic exploration methods to optimize the visualization modeling of fracture distribution; The injection deployment system is used to accurately inject a cluster of micro-nano geotechnical robots and electromagnetically modified micro-nano particles into a target monitoring area; The data transmission and control module includes a first communication module, a signal enhancement module and a first control module; wherein the first communication module is used to realize information exchange between the micro-nano geotechnical robot cluster and the main control unit through wireless communication, the signal enhancement module is used to enhance the wireless signal, and the second control module is used to control the operation of the wireless energy supply system; The main control unit is used to analyze and process the multi-sensor cluster data to obtain first data; The information fusion analysis module is used to perform spatiotemporal alignment and analysis on the first data to obtain second data, and monitor structural deformation, crack changes, chemical composition changes and geological safety status; The early warning module is based on a machine learning model, which is used to perform time series analysis on the second data and model spatiotemporal information in combination with a deep learning algorithm to predict disaster trends; at the same time, it identifies potential geological anomalies through an anomaly detection algorithm to generate disaster risk assessment and early warning information.

2. The micro-nano geotechnical robot geological disaster multi-source information fusion early warning system according to claim 1 is characterized in that: The micro-nano geotechnical robot and the micro-nano geotechnical robot in the micro-nano geotechnical robot cluster include a shell and: Micro-nano sensors are used to monitor changes in geological information in the target monitoring area; Batteries, used to supply electrical energy; A second communication module, including a transmitting antenna and a receiving antenna, is used to exchange information with the main control unit; The second control module is used to control the working states of the micro-nano sensor and the communication module, store the collected geological information data and the corresponding time, and perform signal conditioning and data compensation on the micro-nano sensor.

3. The micro-nano geotechnical robot multi-source geological information in-situ intelligent perception system according to claim 2 is characterized in that: The micro-nano sensor includes an inertial sensor and a monitoring sensor; The monitoring sensors include a temperature sensor, a stress sensor, a fluid pressure sensor, a microbial sensor, an ion concentration sensor and a pH sensor.

4. The micro-nano geotechnical robot multi-source geological information in-situ intelligent perception system according to claim 2 is characterized in that: The battery is an electret and is charged by being excited by the energy supply unit.

5. The micro-nano geotechnical robot multi-source geological information in-situ intelligent perception system according to claim 2 is characterized in that: The shell includes a waterproof layer from the inside to the outside and a shell layer with surface modification; the waterproof layer is made of a polymer coating; the method for obtaining the shell layer with surface modification is: high-speed rotating and mixing the micro-nano geotechnical robot, negatively charged graphene and a coupling agent to enhance the electrostatic repulsion between the micro-nano geotechnical robots.

6. The micro-nano geotechnical robot geological disaster multi-source information fusion early warning system according to claim 1 is characterized in that: The injection deployment system includes an injection device, a fluid delivery system and a pressure regulation unit; The injection device is used to inject the micro-nano geotechnical robot cluster and electromagnetically modified micro-nano particles into the target monitoring area through a borehole, a crack or a specific injection channel; The fluid delivery system is used to input the micro-nano geotechnical robot cluster and electromagnetically modified micro-nano particles into the target area; The pressure control unit is used to adjust the injection pressure and dynamically control the injection flow rate according to the physical characteristics of the target formation.

7. The micro-nano geotechnical robot geological disaster multi-source information fusion early warning system according to claim 1 is characterized in that: The wireless energy supply system adopts a DC power supply or an AC power supply, and the two poles of the power supply are respectively placed between the strata in the target monitoring area. The electric field is excited in the target monitoring area at an interval less than the maximum working time of the micro-nano geotechnical robot to supply energy to the micro-nano geotechnical robot cluster.

8. The micro-nano geotechnical robot multi-source geological information in-situ intelligent perception system according to claim 1 is characterized in that: The electromagnetically modified micro-nano particles are obtained by coating a carbon-based conductive layer on the surface of ferrite particles, cobalt oxide particles and nickel oxide particles.

9. The micro-nano geotechnical robot multi-source geological information in-situ intelligent perception system according to claim 1 is characterized in that: The information fusion analysis module includes a data processing unit, an anomaly detection unit and a multimodal data fusion unit; The data processing unit is used to organize, correct and supplement the first data based on the timestamp and the sensor identification, including: preliminary screening, format conversion and data integrity check; The Kalman filter and time interpolation algorithm are used to calibrate the timestamp of the first data to ensure the timing consistency of the asynchronous sensor data; the Kalman filter and low-pass filter technology are combined to dynamically adjust the noise suppression strategy to remove the noise in the first data; the machine learning algorithm is used to predict the missing data and fill the missing data; The anomaly detection unit is used to intelligently identify complex anomaly patterns and discover potential geological disaster signals using isolation forest and DBSCAN clustering methods; identify and extract geological disaster feature data from potential geological disaster signals; The multimodal data fusion unit is used to integrate the data processed by the data processing unit and the potential geological disaster characteristic data to obtain the second data.

10. The micro-nano geotechnical robot multi-source geological information in-situ intelligent perception system according to claim 1 is characterized in that: The warning module includes a time series prediction unit, an adaptive warning strategy unit, and a remote monitoring and feedback unit; The time series prediction unit is used to adopt a deep learning method to perform time series modeling based on the spatiotemporal information data in the second data to obtain a geological disaster prediction model; the geological disaster prediction model is used to analyze the historical change trends of geological parameters such as ground stress and crack expansion, and predict the occurrence trend and evolution path of geological disasters; The adaptive early warning strategy unit is used to combine multiple data sources including ground stress, temperature changes, fluid pressure, and chemical composition fluctuations to calculate the comprehensive risk index of geological hazards and evaluate the affected scope of crack expansion and formation instability areas; Detect sudden geological anomalies through clustering algorithms to determine whether the anomalies are triggered by external environmental factors; The remote monitoring and feedback unit is used to set multi-level abnormal alarm thresholds and provide real-time monitoring data, warning history records, visual risk maps and remote access.

Images