CN-121977657-A - Groundwater environment monitoring system

Abstract

The invention belongs to the technical field of environment monitoring, and discloses a groundwater environment monitoring system which comprises a monitoring well body, a plurality of remotely controllable packers, an optical fiber array, a robot shuttle, an artificial intelligent model and a control system, wherein the packers are arranged along the depth direction of the monitoring well body, the optical fiber array can invert the seepage velocity field of a stratum around the well by applying controlled heat pulse and monitoring temperature response, the robot shuttle can move in the well and can be provided with an in-situ chemical analysis module or a geophysical imaging module, the artificial intelligent model is operated on a remote cloud platform, and can fuse multisource data, take the seepage field obtained by active detection and the stratum structure obtained by imaging as physical constraints, and can autonomously schedule the robot shuttle to execute in-situ chemical diagnosis of a minute level or scan the stratum among the wells. The invention realizes the leap from passive observation to active intelligent exploration, and remarkably improves the monitoring precision and timeliness.

Inventors

• Shu Qinfeng
• Zhou Meizhu
• LUO HAO
• LI RAN

Assignees

• 中国地质科学院探矿工艺研究所

Dates

Publication Date

20260505

Application Date

20260408

Claims (10)

1. 1. A groundwater environment monitoring system, comprising: Monitoring the well body; A plurality of remotely controllable packers are arranged along the depth direction of the monitoring well body and are used for dividing the monitoring well body into a plurality of independent monitoring intervals, and each monitoring interval is internally provided with a point type sensor for monitoring conventional water quality or hydrologic parameters; The optical fiber array is arranged along the depth direction of the monitoring well body and comprises sensing optical fibers and heating elements, wherein the heating elements are arranged along with the sensing optical fibers, the sensing optical fibers are used for acquiring distributed temperature and/or strain data, and the heating elements are used for applying controlled heat pulses to stratum around the monitoring well body; A robotic shuttle configured to be movable in a depth direction of the monitoring well; The edge calculation module is in communication connection with the packer, the point sensor, the optical fiber array and the robot shuttle; And the remote cloud platform is operated with an artificial intelligent model, and the artificial intelligent model is used for fusion analysis of the monitoring data of the point sensor and the optical fiber array and issuing a control instruction for the robot shuttle to the edge calculation module according to an analysis result.
2. 2. The system of claim 1, wherein the artificial intelligence model is further configured to reverse calculate a thermal physical parameter and a seepage velocity field of the formation surrounding the monitoring well along a depth profile based on the temperature profile monitored by the sensing fiber during and after the heating element is pulsed with heat.
3. 3. The system of claim 2, wherein the artificial intelligence model is a physical information neural network model that uses a percolation velocity field that is inverted according to the temperature profile as a physical constraint for improving accuracy of predictions of groundwater flow and solute transport.
4. 4. The system of claim 1, wherein the robotic shuttle carries a replaceable task module thereon.
5. 5. The system of claim 4, wherein the task module is an in-situ chemical analysis module comprising a microfluidic chip and a sampling interface for extracting trace amounts of water sample from the monitoring interval to the microfluidic chip to effect in-situ real-time detection of chemical components in the water sample.
6. 6. The system of claim 4, wherein the task module is a geophysical imaging module comprising a signal transmitting unit and/or a signal receiving unit.
7. 7. The system of claim 6, wherein the system comprises at least two of the monitoring wells, wherein one of the robotic shuttles within the monitoring well carries the signal emitting unit and the other of the robotic shuttles or stationary sensors within the monitoring well acts as the signal receiving unit, both configured to cooperate to perform cross-hole tomography to obtain structural images of the inter-well formations.
8. 8. The system of claim 1, wherein the packer is provided with a sampling valve that mates with a sampling interface of the robotic shuttle, the sampling valve for allowing the robotic shuttle to extract a water sample after docking.
9. 9. The system of claim 1, wherein the artificial intelligence model is further configured to autonomously generate a task instruction upon identifying a monitoring data anomaly or predicting a potential risk and schedule the robotic shuttle to move to a specified depth by the edge calculation module to perform an in situ chemical analysis or geophysical imaging task.
10. 10. The system of claim 1, further comprising a composite energy harvesting module and an intelligent power management unit, wherein the artificial intelligence model is further configured to predict future task energy consumption and to optimally schedule energy harvested by the composite energy harvesting module via the intelligent power management unit.

Description

Groundwater environment monitoring system Technical Field The invention belongs to the technical field of environmental monitoring, and particularly relates to a groundwater environment monitoring system. Background Groundwater is a vital strategic resource, and environmental monitoring has a central significance for water resource protection and pollution control. The existing underground water monitoring technology mainly relies on the arrangement of point sensors (such as pressure, temperature and pH meters) in a monitoring well and the utilization of a distributed optical fiber sensing technology, and data are transmitted to a cloud platform for analysis through a wireless network. Some systems have also attempted to introduce Artificial Intelligence (AI) models for data processing and trend prediction. However, in practicing the present invention, the inventors have found that the prior art system has a fundamental limitation in that it is a passive observation system in nature. It can only wait for environmental changes to occur and be passively captured by the sensor, thus leading to the following deep technical difficulties: 1. Whether point sensors or passive optical fibers, the measured physical quantity (e.g., temperature, pressure) is non-linear, non-unique, indirect relationship with the core hydrogeologic parameters (e.g., water cut, permeability coefficient, seepage velocity, formation structure). This results in artificial intelligence models, mostly black box models, lack of physical interpretability of the predicted results, and limited accuracy and reliability. 2. Once an anomaly is found, a detailed investigation or sampling analysis is usually performed by manually arriving at the scene, the response period is long, and transient or rapid change pollution events cannot be captured. Even the motion detector of U.S. patent application publication No. US20170044894A1 is limited in its function to passive point sensing along the cross-section and cannot provide diagnostic information. 3. The formation parameters on which artificial intelligence models rely are typically from one-time sampling while drilling, static and low resolution. When groundwater flow changes the formation structure (e.g., undercut forms a dominant channel), the model cannot update itself, resulting in prediction distortion. Disclosure of Invention The present invention aims to solve the above technical problems at least to some extent. Therefore, the invention aims to provide a groundwater environment monitoring system which can actively acquire key physical parameters, dynamically update a stratum model and realize quick response and diagnosis of abnormal events. In order to achieve the above purpose, the present invention provides the following technical solutions: A groundwater environment monitoring system comprises a monitoring well body, a plurality of remotely controllable packers, a robot shuttle, an edge computing module, a remote cloud platform and an artificial intelligent model, wherein the packers are arranged along the depth direction of the monitoring well body and are used for dividing the monitoring well body into a plurality of independent monitoring layers, a point type sensor for monitoring conventional water quality or hydrologic parameters is arranged in each monitoring layer, the optical fiber array is arranged along the depth direction of the monitoring well body and comprises a sensing optical fiber and a heating element which is arranged along the sensing optical fiber and is used for acquiring distributed temperature and/or strain data, the heating element is used for applying controlled heat pulses to stratum around the monitoring well body, the robot shuttle is configured to be capable of moving in the depth direction of the monitoring well body, the edge computing module is in communication connection with the packers, the point type sensor, the optical fiber array and the robot shuttle, and the remote cloud platform is operated with the artificial intelligent model which is used for fusion analysis of monitoring data of the point type sensor and the optical fiber array and is used for sending a robot control instruction to the robot shuttle according to an analysis result. Further, the artificial intelligence model is further configured to reverse calculate a thermophysical parameter and a seepage velocity field of the formation surrounding the monitoring well along the depth profile based on the temperature profile monitored by the sensing fiber during and after the heating element is pulsed. Further, the artificial intelligence model is a physical information neural network model, which uses a seepage velocity field which is inverted according to the temperature change curve as a physical constraint condition for improving the precision of prediction of underground water flow and solute transport. Further, a replaceable task module is mounted on the robot shuttle. Further, the task mod