Surface collapse monitoring system based on multi-parameter distributed optical fiber sensing

Abstract

The invention relates to a surface collapse monitoring system based on multi-parameter distributed optical fiber sensing, and the system comprises a monitoring module which comprises a plurality of fully distributed optical fiber sensors laid in a monitoring area, and an optical signal transmission and demodulation module which comprises a continuous spectrum laser, an optical switch, an optical amplifier, an optical filter, and a demodulation module. The data processing and analysis center comprises a data acquisition module and a collapse model modeling module, and the data acquisition module is used for performing acquisition and digital conversion on the optical signal demodulated by the demodulation module; the collapse model modeling module constructs a model according to the monitoring data and can predict the development trend of surface collapse; and the early warning and communication module communicates with a remote monitoring center. The monitoring system provided by the invention has the advantages of multi-parameter monitoring, high monitoring precision, suitability for large-scale laying, real-time response and the like, and can comprehensively and accurately reflect the real condition of surface collapse.

Description

Technical Field

[0001] The present invention relates to the technical field of ground subsidence monitoring, and in particular to a ground subsidence monitoring system based on multi-parameter distributed optical fiber sensing. Background Art

[0002] As a highly destructive geological disaster, ground collapse has a complex and diverse formation mechanism, which is usually closely related to a variety of geological and human activities. From the perspective of geological factors, in areas with developed karst landforms, the existence of underground caves causes the stability of the overlying rock strata to be destroyed, which can easily cause ground collapse under the action of its own gravity or external loads. In mining areas, large-scale underground mining operations, such as coal and metal mining, will form large areas of goaf. As the scope of the goaf continues to expand, the overlying rock strata lose support, gradually deform, crack and collapse, leading to ground collapse. In addition, excessive groundwater extraction will cause a sharp drop in the groundwater level, change the stress state of the rock and soil, and cause the stratum structure that was originally in a balanced state to become unbalanced, which can also easily induce ground collapse.

[0003] Leveling is a more traditional method of monitoring ground subsidence. Its principle is to infer the ground settlement by measuring the height difference between different monitoring points. This method relies on manual point-by-point operation, which is not only inefficient, but also difficult to fully reflect the overall situation of ground subsidence when monitoring a large area due to the limited number of monitoring points. It is easy to cause monitoring blind spots and miss early signs of subsidence.

[0004] GPS measurement uses satellite positioning technology to obtain the three-dimensional coordinate information of the monitoring point, and judges the displacement of the ground through the change of coordinates. However, this method is seriously affected by satellite signal obstruction. In areas with high-rise buildings in cities or complex terrain such as mountain canyons, satellite signals are easily blocked, resulting in reduced positioning accuracy or even failure to work properly. In addition, GPS measurement also has the problem of sparse monitoring points. Its monitoring results can only reflect the situation of discrete monitoring points and cannot provide continuous ground deformation information. In some actual cases, such as when monitoring ground subsidence around mining areas in mountainous areas, the accuracy of GPS measurement is difficult to meet the requirements because the mountains block satellite signals, and it is impossible to accurately capture the subtle deformation in the early stage of ground subsidence, so it is impossible to issue early warnings.

[0005] Most of the existing ground collapse monitoring solutions based on distributed fiber optic sensing technology focus only on the monitoring of a single physical quantity, such as only using the strain characteristics of the optical fiber to monitor the deformation of the rock and soil, or only focusing on the impact of temperature changes on the optical fiber. This single parameter monitoring method has obvious shortcomings, because the ground collapse process is a complex geomechanical process, involving the interaction of multiple physical quantities such as the strain of the rock and soil, temperature changes, and acoustic signals generated by the fracture. Single parameter monitoring cannot fully and accurately reflect the actual situation of the ground collapse, and is prone to misjudgment or omission. Summary of the invention

[0006] The purpose of the present invention is to provide a ground collapse monitoring system based on multi-parameter distributed optical fiber sensing to achieve full coverage of ground collapse disasters from potential risk identification, real-time dynamic monitoring to accurate prediction and early warning and scientific decision support.

[0007] To achieve the above object, the present invention adopts the following technical solution:

[0008] A ground collapse monitoring system based on multi-parameter distributed optical fiber sensing, comprising:

[0009] A monitoring module, comprising a plurality of fully distributed optical fiber sensors laid in the monitoring area, wherein the fully distributed optical fiber sensors include a strain sensing core, a temperature sensing core and an acoustic sensing core;

[0010] An optical signal transmission and demodulation module, including a continuous spectrum laser, an optical switch, an optical amplifier, an optical filter and a demodulation module. The continuous spectrum laser is used to provide the required optical excitation to the fully distributed optical fiber sensor, and the optical switch is used to switch and control multiple fully distributed optical fiber sensors.

[0011] The data processing and analysis center includes a data acquisition module and a collapse model building module. The data acquisition module is used to collect and digitally convert the optical signal demodulated by the demodulation module; the collapse model building module builds a model based on the data monitored in real time by the distributed optical fiber sensor network and can predict the development trend of ground collapse;

[0012] The early warning and communication module is used to judge the data and communicate with the remote monitoring center when the alarm is triggered.

[0013] Furthermore, before the fully distributed fiber optic sensors are laid, geological surveys are conducted in the monitoring area to obtain data on underground strata structure, rock properties, groundwater distribution, and potential geological disaster risk points. These data are integrated into a three-dimensional geographic information system platform, and optimization algorithms are used to plan the optimal laying path and network topology of the fully distributed optical fiber.

[0014] Furthermore, during the laying process of fully distributed fiber optic sensors, fully automated laying equipment and intelligent construction management systems are used. The laying equipment can measure parameters such as the tension, curvature and burial depth of the optical fiber in real time, and automatically adjust the laying parameters according to preset standards; the intelligent construction management system is used to conduct all-round monitoring and management of the entire laying process, including the collection and analysis of information on equipment operating status, material supply, and construction progress.

[0015] Furthermore, the optical signal transmission and demodulation module starts optical signal emission and data collection according to a preset sampling frequency and time interval. The sampling frequency can be dynamically adjusted according to the risk level and geological activity of the monitored area. The sampling frequency can be increased in high-risk areas or periods of frequent geological activities; in low-risk areas or periods of geological stability, the sampling frequency can be reduced to reduce the amount of data and system energy consumption. After the optical signal is transmitted in the fully distributed optical fiber sensor and interacts with the rock and soil, the returned optical signal is amplified by the optical amplifier and filtered by the optical filter before entering the demodulation module for demodulation. The demodulated multi-parameter data is converted into digital signals by the data acquisition module and then transmitted through the network communication interface according to the set communication protocol.

[0016] Furthermore, the data transmission adopts a hybrid network architecture, including a fiber-optic wired network and a 5G wireless network. In areas near the monitoring site or with fiber-optic access conditions, the fiber-optic wired network is preferentially used for data transmission to ensure high-speed, stable and reliable data transmission. For some mobile monitoring equipment or monitoring points far away from the fiber-optic network, the 5G wireless network is used for data transmission to achieve real-time data return.

[0017] Furthermore, after receiving the data, the data processing and analysis center first stores the data in a distributed storage system, then starts a data preprocessing program, and processes the data using an outlier detection and data cleaning algorithm based on artificial intelligence; at the same time, the data is normalized and standardized; and then temperature compensation, acoustic signal processing and positioning, and ground collapse model prediction are performed.

[0018] Furthermore, the ground collapse model calculates and predicts based on the processed multi-parameter data to obtain the development trend of ground collapse, and stores the results in the result database. At the same time, data visualization technology is used to display the monitoring data, analysis results and prediction trends in the form of intuitive charts, maps or animations.

[0019] Furthermore, the early warning and communication module monitors the result database of the data processing and analysis center in real time. When the monitoring data exceeds the preset early warning threshold, the early warning program is immediately started. The early warning program generates detailed early warning information based on the relevant information of the early warning type, level and monitoring area, including the precise location, time, scale, possible impact range and response measures of the collapse event, and then sends the early warning information to relevant personnel and departments through multi-mode fusion communication technology.

[0020] Furthermore, the warning threshold is set using a dynamic threshold setting method based on big data analysis and machine learning. A large amount of historical data on ground collapse under different geological conditions, environmental monitoring data, and fully distributed fiber optic sensor monitoring data are collected, and training is performed using learning algorithms such as random forests and support vector machines to establish a warning threshold model.

[0021] Furthermore, the ground collapse model adopts a hybrid model based on the combination of finite element analysis and deep learning. First, the geomechanical model of the monitoring area is established using finite element analysis software to simulate the process of ground collapse under different working conditions, generate a large amount of training data, and then input these data into a deep neural network for training. The structure of the neural network adopts an architecture that combines a multi-layer long short-term memory network with a fully connected layer. It can learn the spatiotemporal evolution laws of the ground collapse process. By continuously optimizing the weights and thresholds of the neural network, the model can accurately predict the development trend of ground collapse based on the data monitored in real time by the fully distributed optical fiber sensor, including the expansion direction, speed and possible affected range of the collapse area.

[0022] Beneficial effects of the present invention:

[0023] (1) Ultra-high precision monitoring

[0024] The present invention uses a fully distributed optical fiber sensor, which can be detected in time at the embryonic stage of ground collapse, that is, when the internal structure of the rock and soil mass begins to undergo very slight changes, making it possible to take preventive measures in advance. It can not only accurately reflect the subtle changes in the underground temperature field caused by ground collapse, but also effectively compensate for the measurement errors of other physical quantities caused by temperature fluctuations. By using super-resolution voiceprint recognition based on deep neural networks and multi-dimensional arrival time difference precise positioning technology, it is possible to parse out weak collapse-related acoustic signals and accurately determine their source location. It can accurately identify specific acoustic features closely related to ground collapse in the complex geological environment noise background.

[0025] (2) Seamless coverage over a large area

[0026] The fully distributed optical fiber sensor used in the present invention has super flexibility and adaptability. Combined with the laying robot, it can achieve uninterrupted and seamless large-scale laying under various complex terrains and harsh environmental conditions.

[0027] (3) Super strong real-time response

[0028] Data collection and transmission are extremely fast. At the moment of geological disasters, the system can synchronously collect multi-parameter information on fully distributed optical fibers at a high sampling frequency and quickly transmit the data to the data processing and analysis center. The early warning signal is triggered in time, which wins extremely valuable golden time for emergency response and effectively reduces the risk of casualties and property losses.

[0029] (4) Comprehensive analysis and accurate prediction

[0030] Based on the ultra-large-scale data processing platform that integrates quantum computing and artificial intelligence and the multi-scale finite element-deep learning coupling model, the comprehensive analysis and prediction capabilities of ground collapse have been improved. Through in-depth mining and learning of the massive multi-parameter data collected by the fully distributed optical fiber for a long time, it is not only possible to accurately predict the development trend of ground collapse, including detailed information such as the expansion direction, speed, depth of the collapse area, and the scope of secondary disasters that may be caused, but also to dynamically adjust the parameters and boundary conditions of the prediction model according to real-time monitoring data, so as to achieve adaptive and accurate prediction of ground collapse processes under complex and changeable geological environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a schematic diagram of the layout principle of the ground collapse monitoring system based on multi-parameter distributed optical fiber sensing of the present invention.

[0032] The names corresponding to the marks in the figure are:

[0033] 1. Monitoring module; 2. Data transmission module; 3. Data processing and analysis center; 4. Data receiving module; 5. Fully distributed optical fiber sensor; 6. Quaternary system on the surface; 7. Bedrock. DETAILED DESCRIPTION

[0034] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0035] Example:

[0036] like Figure 1 As shown, a ground collapse monitoring system based on multi-parameter distributed optical fiber sensing includes a monitoring module, an optical signal transmission and demodulation module, a data processing and analysis center, and an early warning and communication module.

[0037] (i) Monitoring module 1, including multiple fully distributed optical fiber sensors 5 laid in the monitoring area, the fully distributed optical fiber sensors 5 including strain sensing cores, temperature sensing cores and acoustic sensing cores; the core of the fully distributed optical fiber sensor 5 is provided with an anti-interference coating. Nano-scale temperature-sensitive materials are embedded in the cladding, and their thermo-optical effect is used to achieve high-precision monitoring of underground temperature changes, effectively avoiding the temperature and strain measurement errors caused by differences in the installation positions of traditional separate sensors. A layer of intelligent acoustic sensing film is coated on the surface of the optical fiber, which is composed of piezoelectric polymers and micro-nano acoustic resonators. It can efficiently convert the acoustic wave signals generated by the rupture of the rock and soil body into electrical signals, and achieve wide-band and high-sensitivity detection of acoustic signals by coupling and modulating with the optical signals transmitted in the optical fiber. The frequency response range is from 1Hz to 10kHz, and the characteristic acoustic frequencies generated by different types of ground collapse can be accurately identified. This multi-structure fusion of fully distributed optical fiber realizes the synchronous and high-precision monitoring of strain, temperature and acoustic signals on one optical fiber, greatly simplifies the architecture of the monitoring system, and improves the reliability and accuracy of monitoring.

[0038] (ii) An optical signal transmission and demodulation module, including a continuous spectrum laser, an optical switch, an optical amplifier, an optical filter and a demodulation module. The continuous spectrum laser is used to provide the required optical excitation to the fully distributed optical fiber sensor, and the optical switch is used to switch and control multiple fully distributed optical fiber sensors.

[0039] The optical signal transmission and demodulation module starts optical signal transmission and data collection according to the preset sampling frequency and time interval. The sampling frequency can be dynamically adjusted according to the risk level and geological activity of the monitored area. The sampling frequency can be increased in high-risk areas or periods of frequent geological activities; in low-risk areas or periods of geological stability, the sampling frequency can be reduced to reduce the amount of data and system energy consumption. After the optical signal is transmitted in the fully distributed optical fiber sensor and interacts with the rock and soil, the returned optical signal is amplified by the optical amplifier and filtered by the optical filter before entering the demodulation module for demodulation. The demodulated multi-parameter data is converted into digital signals through the data acquisition module and then transmitted through the network communication interface according to the set communication protocol.

[0040] By deploying distributed optical amplifiers and intelligent optical switch nodes along the optical fiber, a dynamically adjustable optical transmission network is constructed. The system uses artificial intelligence algorithms to automatically optimize the gain distribution of optical amplifiers and the connection paths of optical switches based on the real-time monitoring of optical fiber loss, optical signal strength, and risk level of each monitoring area. For example, in areas far from the light source and with high optical fiber loss, the intelligent algorithm will increase the gain of the optical amplifier in that area, and adjust the optical switch so that the optical signal is preferentially transmitted along the path with lower loss, ensuring that the optical signal can reach each monitoring point with stable intensity. When a geological disaster suddenly causes damage to some optical fibers, the system can quickly detect the fault point and bypass the damaged area by reconfiguring the optical switch nodes to ensure that monitoring work in other normal areas continues, realizing adaptive reconstruction and fault-tolerant operation of the optical transmission network, and effectively improving the reliability and stability of the fully distributed optical fiber sensor system in complex environments.

[0041] The super computing power of quantum computing and machine learning algorithms are used to demodulate and analyze the complex optical signals transmitted back by the fully distributed optical fiber. The quantum machine learning model can quickly identify the characteristics of weak quantum state changes in the optical signal caused by strain, temperature and acoustic changes, and accurately calculate the corresponding physical quantity values ​​by representing and training the quantum states of a large number of optical signal samples. Compared with the traditional demodulation method based on classical computing, this demodulation method based on quantum machine learning has higher demodulation accuracy and speed, especially when processing massive, high-dimensional optical signal data. For example, when demodulating large-area, multi-node fully distributed optical fiber monitoring data, the demodulation speed can be increased by dozens of times, and the accuracy can be improved by about 30%, providing strong technical support for the real-time, high-precision monitoring of ground collapse.

[0042] The above data transmission adopts a hybrid network architecture, including fiber-optic wired network and 5G wireless network. In areas near the monitoring site or with fiber-optic access conditions, fiber-optic wired network is preferred for data transmission to ensure high-speed, stable and reliable data transmission; for some mobile monitoring equipment or monitoring points far away from the fiber-optic network, 5G wireless network is used for data transmission to achieve real-time data return. During the data transmission process, data reliability assurance technology based on error correction coding and retransmission mechanism is adopted, such as using low-density parity check code (LDPC) for data encoding. When the receiving end detects a data error, it automatically sends a retransmission request to the sending end to ensure the integrity and accuracy of the data, and the data transmission error rate is less than 0.0005%.

[0043] (iii) Data processing and analysis center 3, including a data acquisition module and a collapse model building module. The data acquisition module is used to collect and digitally convert the optical signal demodulated by the demodulation module; the collapse model building module is used to build a model based on the data monitored in real time by the distributed optical fiber sensor network and can predict the development trend of ground collapse.

[0044] After receiving the data through the data receiving module 5, the data processing and analysis center 3 first stores the data in a distributed storage system. The distributed storage system adopts a distributed file system based on a cloud storage architecture, which has high scalability, high availability and high data redundancy, and can meet the needs of long-term and large-scale data storage of the fully distributed optical fiber sensor network. Then the data preprocessing program is started, and the data is processed using an outlier detection and data cleaning algorithm based on artificial intelligence; for example, the data is reconstructed using an autoencoder neural network, and data points with large reconstruction errors are identified as outliers and corrected or eliminated. At the same time, the data is normalized and standardized to improve the quality and availability of the data. Then temperature compensation, acoustic signal processing and positioning, and ground collapse model prediction are performed. The temperature compensation algorithm performs real-time temperature compensation on the strain data on the fully distributed optical fiber based on the all-optical thermal effect physical model and deep learning algorithm mentioned above. The acoustic signal processing and positioning algorithm uses voiceprint recognition and arrival time difference positioning technology based on deep learning to determine the location and type of the collapse event. The ground collapse model calculates and predicts based on the processed multi-parameter data, obtains the development trend of the ground collapse, and stores the results in the result database. At the same time, data visualization technology, such as the Web-based three-dimensional visualization platform, is used to display monitoring data, analysis results and forecast trends in the form of intuitive charts, maps and animations, making it easier for managers to view and analyze, and providing strong support for decision-making.

[0045] The ground collapse model adopts a hybrid model based on finite element analysis and deep learning. First, the finite element analysis software is used to establish a geomechanical model of the monitoring area, simulate the process of ground collapse under different working conditions, generate a large amount of training data, and then input these data into a deep neural network for training. The structure of the neural network adopts an architecture that combines a multi-layer long short-term memory network with a fully connected layer. It can learn the spatiotemporal evolution laws of the ground collapse process. By continuously optimizing the weights and thresholds of the neural network, the model can accurately predict the development trend of ground collapse based on the real-time monitoring data of the fully distributed optical fiber sensor, including the expansion direction and speed of the collapse area and the possible affected range.

[0046] (iv) Early warning and communication module, which is used to judge the data and communicate with the remote monitoring center when an alarm is triggered.

[0047] The early warning and communication module monitors the result database of the data processing and analysis center in real time. When the monitoring data exceeds the preset early warning threshold, the early warning program is immediately started. The early warning program generates detailed early warning information based on the warning type, level and relevant information of the monitoring area, including the precise location, time, scale, possible impact range and response measures of the collapse event, and then sends the early warning information to relevant personnel and departments, such as the local government emergency management department, surrounding residents, and related enterprises through multi-mode fusion communication technology. After receiving the early warning information, the relevant personnel take corresponding protective measures according to the pre-made emergency plan and the specific content of the early warning information. For example, the surrounding residents evacuate according to the designated evacuation routes and shelters, and the relevant enterprises stop production operations in the dangerous areas and protect or transfer important equipment and facilities. The management department activates the disaster emergency command system and organizes rescue forces, such as fire fighting, medical, engineering rescue teams, etc., to quickly rush to the scene for rescue and emergency work. At the same time, satellite remote sensing, drone aerial photography and other technologies are used to monitor and evaluate the collapse site in real time, and the rescue strategy and resource allocation plan are adjusted in time according to the on-site situation to minimize the casualties and property losses caused by ground collapse.

[0048] The warning threshold is set using a dynamic threshold setting method based on big data analysis and machine learning. A large amount of historical data on ground collapse under different geological conditions, environmental monitoring data, and fully distributed fiber optic sensor monitoring data are collected, and training is performed using learning algorithms such as random forests and support vector machines to establish a warning threshold model.

[0049] When the system is initialized, each module is fully self-tested and calibrated. Each device has built-in intelligent sensors and diagnostic chips, which can automatically detect the hardware status of the device, the software operation status, the transmission performance of the optical signal and other parameters, and transmit the test results to the data processing and analysis center through the network. The optical signal transmission and demodulation module is accurately calibrated using standard light sources and calibration equipment to ensure the accuracy of the emission, transmission, demodulation and other links of the optical signal. The software system of the data processing and analysis center is initialized and configured, including the creation of the database, the setting of algorithm parameters, and the pre-training of the model, so that the system can quickly enter the normal operation state.

[0050] The above is the structure and principle description of each functional module of the monitoring system of the present invention. The present invention also optimizes the design of the laying of the fully distributed optical fiber sensor, as follows:

[0051] Before laying the fully distributed optical fiber sensor, geological surveys are conducted in the monitoring area, such as 3D geological radar, high-precision seismic exploration, and drone aerial mapping, to obtain data on underground stratum structure, rock characteristics, groundwater distribution, and potential geological disaster risk points. These data are integrated into the 3D Geographic Information System (GIS) platform, and the optimization algorithm is used to plan the optimal laying path and network topology of the fully distributed optical fiber. The optimization algorithm can be a combination of genetic algorithm and particle swarm optimization algorithm.

[0052] During the laying process of the fully distributed optical fiber sensor, fully automated laying equipment and intelligent construction management system are used. The laying equipment can measure the tension, curvature and burial depth of the optical fiber in real time, and automatically adjust the laying parameters according to the preset standards. The laying equipment uses a laying robot. On steep slopes in mountainous areas, the laying robot can use its anchoring device to firmly fix the optical fiber in the rock and soil. In river crossing areas, waterproof and sealed pipes are used to encapsulate the optical fiber for deep burial. The intelligent construction management system is used to monitor and manage the entire laying process in all directions, including the collection and analysis of information on equipment operation status, material supply, and construction progress.

[0053] Advantages of the ground subsidence monitoring system based on multi-parameter distributed optical fiber sensing of the present invention:

[0054] (1) Ultra-high precision monitoring

[0055] In the strain measurement dimension, the present invention uses a fully distributed optical fiber sensor, combined with a demodulation technology based on the principle of quantum optical interference, to capture subtle changes at the nano-strain level. For example, in the monitoring of micro-stress adjustments in underground rock and soil, the strain changes corresponding to an elongation of only one millionth of the diameter of a hair can be accurately detected. This allows for timely detection of ground collapse in the early stages, when the internal structure of the rock and soil begins to undergo very slight changes, making it possible to take preventive measures in advance. Traditional leveling and GPS measurements simply cannot achieve such microscopic perception accuracy, and ordinary distributed optical fiber sensing technology can only reach the micro-strain level, which is far less than the ultra-high accuracy of this solution.

[0056] In the dimension of temperature monitoring, ultra-fine measurement with a temperature resolution better than 0.01°C is achieved through all-fiber thermo-optical effect quantum control technology and an optical frequency comb calibration algorithm based on artificial intelligence optimization. It can not only accurately reflect the subtle changes in the underground temperature field caused by ground collapse, but also effectively compensate for the measurement errors of other physical quantities caused by temperature fluctuations. In contrast, the resolution of traditional temperature sensors is generally between 0.1°C and 1°C, which is difficult to meet the stringent requirements of high-precision ground collapse monitoring for temperature monitoring.

[0057] In the acoustic signal monitoring dimension, the use of super-resolution voiceprint recognition based on deep neural networks and multi-dimensional arrival time difference precise positioning technology can parse the collapse-related acoustic signals with a frequency as low as 0.05Hz and extremely weak sound intensity, and accurately determine its source location, with the positioning error controlled within 0.1 meters. It can accurately identify specific acoustic features closely related to ground collapse in the complex geological environment noise background.

[0058] (2) Seamless coverage over a large area

[0059] The fully distributed optical fiber used in this invention has super flexibility and adaptability. Combined with advanced all-terrain laying robots, it can achieve uninterrupted and seamless large-scale laying in various complex terrains and harsh environmental conditions. For example, in a large-scale watershed monitoring project spanning multiple geological and geomorphological units and covering an area of ​​thousands of square kilometers, only one set of fully distributed optical fiber sensing monitoring system is needed to achieve comprehensive coverage of the entire area from the source to the estuary, and the spacing between monitoring points can be flexibly adjusted to the centimeter level according to the geological risk level, truly achieving refined monitoring of every inch of land.

[0060] (3) Super strong real-time response

[0061] With the help of ultra-high-speed, multi-channel synchronous acquisition photon counting data acquisition cards and a hybrid data transmission network based on 5G slicing technology and satellite communication low-latency links, data acquisition and transmission are extremely fast. At the moment of a geological disaster, the system can synchronously collect multi-parameter information on the fully distributed optical fiber at a sampling frequency of millions of times per second, and transmit the data to the data processing and analysis center within milliseconds. For example, in the chain reaction of ground collapse caused by an earthquake, when the seismic wave just triggers the internal structural changes of the rock and soil, the system can capture the changes in related physical quantities, complete data acquisition, transmission and preliminary analysis within 1 second, and then trigger the early warning signal. It has won extremely precious golden time for disaster warning and emergency response, effectively reducing the risk of casualties and property losses.

[0062] (4) Extreme anti-interference stability

[0063] The fully distributed optical fiber's unique fully enclosed optical transmission structure and anti-electromagnetic interference and corrosion-resistant material coating enable it to maintain stable optical signal transmission and physical quantity perception performance in the face of ultra-strong electromagnetic pulse interference, strong chemical corrosion environment and extreme physical impact. For example, when monitoring ground collapse in the surrounding areas of military facilities that are attacked by high-intensity electromagnetic pulses, the fully distributed optical fiber sensor monitoring system of the present invention can work normally in harsh environments with electromagnetic interference intensity as high as several thousand volts / meter, and continuously and accurately monitor the ground status.

[0064] (5) Comprehensive analysis and accurate prediction

[0065] Based on the ultra-large-scale data processing platform that integrates quantum computing and artificial intelligence and the multi-scale finite element-deep learning coupling model, the comprehensive analysis and prediction capabilities of ground collapse have been improved. Through deep mining and learning of the massive multi-parameter data collected by the fully distributed optical fiber for a long time, it is not only possible to accurately predict the development trend of ground collapse, including detailed information such as the expansion direction, speed, depth of the collapse area, and the scope of secondary disasters that may be caused, with the prediction error controlled within 5%, but also able to dynamically adjust the parameters and boundary conditions of the prediction model according to real-time monitoring data, so as to achieve adaptive and accurate prediction of the ground collapse process under complex and changeable geological environments. For example, when dealing with ground collapse disasters caused by the coupling of rapid changes in groundwater levels and seismic activities, it is possible to analyze in real time the complex nonlinear relationship between the propagation characteristics of seismic waves, the fluctuation law of groundwater levels, and the mechanical response of rock and soil bodies, and accurately predict the time, location and scale of collapse based on this, providing a scientific basis for the formulation of targeted disaster prevention and control strategies.

[0066] The ground subsidence monitoring system based on multi-parameter distributed optical fiber sensing of the present invention can also be expanded:

[0067] Combine the fully distributed fiber optic sensor monitoring system with multi-source remote sensing technology (such as satellite optical remote sensing, radar remote sensing and drone low-altitude remote sensing) to build a ground collapse monitoring network that integrates the sky and the earth. Satellite optical remote sensing can provide large-area surface image data. By comparing and analyzing images from different periods, it can monitor the overall deformation trend of the ground, changes in vegetation coverage, and land use conditions at a macro level. This information helps to determine the possible impact range of ground collapse and regional ecological changes. Radar remote sensing can penetrate clouds and vegetation to obtain high-precision three-dimensional terrain information on the surface, accurately measure small elevation changes on the ground, and complement the underground monitoring data of fully distributed fiber optic sensors to further improve the accuracy and spatial resolution of ground collapse monitoring. Drone low-altitude remote sensing can conduct flexible and rapid local monitoring of key monitoring areas or sudden collapse events, obtain high-resolution aerial images and videos, and clearly display the detailed features of the collapse area, the damage to surrounding buildings, and the real-time status of the rescue site. By establishing a data fusion model, the fully distributed fiber optic sensor data and multi-source remote sensing data are deeply integrated, and the intrinsic connections between the data are mined using artificial intelligence algorithms to achieve all-round, multi-scale monitoring and analysis of ground collapse, providing richer and more accurate information support for disaster assessment, emergency rescue and post-disaster recovery.

[0068] Develop a multi-source data collaborative processing and analysis system based on a cloud platform. The system stores fully distributed fiber optic sensor data, multi-source remote sensing data, and other related geographic information data (such as geological structure maps, groundwater level data, etc.) in a centralized manner on a cloud server, and uses the powerful computing power and distributed storage advantages of cloud computing to achieve rapid data processing, analysis, and sharing. Users from different regions and departments can access the cloud platform through the Internet, obtain relevant data and analysis results according to their own permissions, and work together. For example, geological disaster prevention and control departments can view ground collapse monitoring data and analysis reports in real time on the cloud platform and formulate disaster prevention and control strategies; urban planning departments can use these data to assess the impact of ground collapse on urban construction and optimize urban planning plans; scientific research institutions can conduct ground collapse mechanism research and monitoring technology innovation based on cloud platform data. This multi-source data collaborative processing and analysis system based on a cloud platform breaks the data island, promotes information exchange and cooperation between various fields, and improves the overall efficiency and level of ground collapse monitoring and prevention work.

[0069] The fully distributed optical fiber sensing monitoring system can also be incorporated into the smart city infrastructure monitoring system, interconnected with other intelligent sensing devices in the city (such as intelligent traffic sensors, intelligent water and electricity meters, environmental monitoring sensors, etc.), and realize the integration and sharing of multi-field data in the city. The fully distributed optical fiber sensor network can not only monitor ground collapse, but also monitor the operating status of urban underground pipe networks (such as water supply pipes, gas pipes, sewage pipes, etc.) in real time. By monitoring the strain changes of the rock and soil around the pipeline, safety hazards such as pipeline leakage and rupture can be discovered in time, and the fault point can be accurately located, providing efficient and accurate technical means for the maintenance and management of urban pipe networks. At the same time, the ground collapse monitoring data is combined with the intelligent transportation system. When it is detected that a certain area may collapse, early warning information is sent to the surrounding traffic facilities (such as traffic lights, electronic display screens, vehicle navigation systems, etc.) in time to adjust the traffic flow and guide vehicles to avoid dangerous areas to ensure the safety and smoothness of urban traffic. In addition, it works in collaboration with the urban environmental monitoring system, using the fully distributed fiber optic sensor network to monitor underground temperature, humidity and other environmental parameters, analyzes environmental issues such as the urban heat island effect and the spread of groundwater pollution, and provides data support for urban environmental protection and sustainable development.

[0070] The present invention is not limited to the above-mentioned optimal implementation mode. Anyone can derive other various forms of products under the inspiration of the present invention. However, no matter what changes are made in the shape or structure, all technical solutions that are the same or similar to those of the present application fall within the protection scope of the present invention.

Claims

1. A ground subsidence monitoring system based on multi-parameter distributed optical fiber sensing, characterized in that: include: A monitoring module, comprising a plurality of fully distributed optical fiber sensors laid in the monitoring area, wherein the fully distributed optical fiber sensors include a strain sensing core, a temperature sensing core and an acoustic sensing core; An optical signal transmission and demodulation module, including a continuous spectrum laser, an optical switch, an optical amplifier, an optical filter and a demodulation module. The continuous spectrum laser is used to provide the required optical excitation to the fully distributed optical fiber sensor, and the optical switch is used to switch and control multiple fully distributed optical fiber sensors. The data processing and analysis center includes a data acquisition module and a collapse model building module. The data acquisition module is used to collect and digitally convert the optical signal demodulated by the demodulation module; the collapse model building module builds a model based on the data monitored in real time by the distributed optical fiber sensor network and can predict the development trend of ground collapse; The early warning and communication module is used to judge the data and communicate with the remote monitoring center when the alarm is triggered.

2. The ground subsidence monitoring system based on multi-parameter distributed optical fiber sensing according to claim 1 is characterized in that: Before laying the fully distributed fiber optic sensors, a geological survey is conducted in the monitoring area to obtain data on the underground stratum structure, rock properties, groundwater distribution, and potential geological disaster risk points. These data are integrated into the three-dimensional geographic information system platform, and the optimization algorithm is used to plan the optimal laying path and network topology of the fully distributed optical fiber.

3. The ground subsidence monitoring system based on multi-parameter distributed optical fiber sensing according to claim 2 is characterized in that: During the laying process of fully distributed fiber optic sensors, fully automated laying equipment and intelligent construction management systems are used. The laying equipment can measure the tension, curvature and burial depth parameters of the optical fiber in real time, and automatically adjust the laying parameters according to preset standards; the intelligent construction management system is used to conduct all-round monitoring and management of the entire laying process, including the collection and analysis of information on equipment operating status, material supply, and construction progress.

4. The ground subsidence monitoring system based on multi-parameter distributed optical fiber sensing according to claim 1 is characterized in that: The optical signal transmission and demodulation module starts optical signal emission and data collection according to the preset sampling frequency and time interval. The sampling frequency can be dynamically adjusted according to the risk level and geological activity of the monitoring area. The sampling frequency can be increased in high-risk areas or periods of frequent geological activities; in low-risk areas or periods of geological stability, the sampling frequency can be reduced to reduce the amount of data and system energy consumption. After the optical signal is transmitted in the fully distributed optical fiber sensor and interacts with the rock and soil, the returned optical signal is amplified by the optical amplifier and filtered by the optical filter before entering the demodulation module for demodulation. The demodulated multi-parameter data is converted into digital signals by the data acquisition module and then transmitted through the network communication interface according to the set communication protocol.

5. The ground subsidence monitoring system based on multi-parameter distributed optical fiber sensing according to claim 4 is characterized in that: The data transmission adopts a hybrid network architecture, including fiber-optic wired network and 5G wireless network. In areas near the monitoring site or with fiber-optic access conditions, fiber-optic wired network is preferentially used for data transmission to ensure high-speed, stable and reliable data transmission. For some mobile monitoring equipment or monitoring points far away from the fiber-optic network, 5G wireless network is used for data transmission to realize real-time data return.

6. The ground subsidence monitoring system based on multi-parameter distributed optical fiber sensing according to claim 1 is characterized in that: After receiving the data, the data processing and analysis center first stores the data in the distributed storage system, then starts the data preprocessing program, and processes the data using an outlier detection and data cleaning algorithm based on artificial intelligence; at the same time, the data is normalized and standardized; Then temperature compensation, acoustic signal processing and positioning, and ground collapse model prediction are carried out.

7. The ground subsidence monitoring system based on multi-parameter distributed optical fiber sensing according to claim 6 is characterized in that: The ground collapse model calculates and predicts based on the processed multi-parameter data, obtains the development trend of ground collapse, and stores the results in the result database. At the same time, it uses data visualization technology to display the monitoring data, analysis results and prediction trends in the form of intuitive charts, maps or animations.

8. The ground subsidence monitoring system based on multi-parameter distributed optical fiber sensing according to claim 7 is characterized in that: The early warning and communication module monitors the result database of the data processing and analysis center in real time. When the monitoring data exceeds the preset early warning threshold, the early warning program is immediately started. The early warning program generates detailed early warning information based on the early warning type, level and relevant information of the monitoring area, including the precise location, time, scale, possible impact range and response measures of the collapse event, and then sends the early warning information to relevant personnel and departments through multi-mode fusion communication technology.

9. The ground subsidence monitoring system based on multi-parameter distributed optical fiber sensing according to claim 8 is characterized in that: The warning threshold is set using a dynamic threshold setting method based on big data analysis and machine learning. A large amount of historical data on ground collapse under different geological conditions, environmental monitoring data, and fully distributed fiber optic sensor monitoring data are collected, and training is performed using learning algorithms such as random forests and support vector machines to establish a warning threshold model.

10. The ground subsidence monitoring system based on multi-parameter distributed optical fiber sensing according to claim 7, characterized in that: The ground collapse model adopts a hybrid model based on the combination of finite element analysis and deep learning. First, the geomechanical model of the monitoring area is established using finite element analysis software to simulate the process of ground collapse under different working conditions, generate a large amount of training data, and then input these data into a deep neural network for training. The structure of the neural network adopts an architecture that combines a multi-layer long short-term memory network with a fully connected layer, which can learn the spatiotemporal evolution laws of the ground collapse process. By continuously optimizing the weights and thresholds of the neural network, the model can accurately predict the development trend of ground collapse based on the data monitored in real time by the fully distributed optical fiber sensor, including the expansion direction and speed of the collapse area and the possible affected range.

Images