Monitoring method and device for underground water seepage

Abstract

The embodiment of the invention provides a monitoring method and device for underground water seepage, and relates to the field of underground water seepage research, and the method specifically comprises the steps: obtaining a target historical water seepage data set of a target underground region; the target historical water seepage data set comprises a plurality of historical water seepage data subsets; performing feature selection on the target historical water seepage data set to generate a water seepage data feature set; acquiring historical water seepage state information corresponding to the water seepage data features, and training the initial machine learning model based on the water seepage data feature set and the historical water seepage state information to generate a target water seepage state prediction model; and inputting the real-time water seepage data set into the target water seepage state prediction model to obtain a water seepage state prediction result corresponding to the real-time water seepage data set. According to the invention, the accuracy and reliability of underground water seepage condition monitoring can be improved.

Description

Technical Field

[0001] The present disclosure relates to the field of underground seepage research, and particularly to a monitoring method and device for underground seepage. Background Art

[0002] In underground mining operations, the seepage problem is one of the key hidden dangers threatening safe production and stable operation. Long-term seepage will corrode equipment, shorten the service life of equipment, increase maintenance costs, and affect underground mining efficiency. Therefore, it is of great significance to accurately and comprehensively monitor the underground seepage situation.

[0003] Currently, the fiber Bragg grating sensor monitoring method is widely used in the field of underground seepage monitoring. Its working principle is to horizontally drill holes in the areas prone to seepage underground, horizontally bury the fiber Bragg grating sensors therein, seal the openings of the drill holes with waterproof foam, and connect them to the oscilloscope system outside the drill holes through cables. By using the chelation reaction of the coating layer with specific metal ions (such as iron ions), the concentration of specific metal ions inside the fiber Bragg grating section is increased, causing a change in the refractive index of the fiber Bragg grating area. By measuring this change in refractive index, the change in the concentration of metal ions in the drill hole is deduced, and then combined with the content of metal ions in the water quality water sample monitoring report, the seepage rate of the permeating water is further calculated.

[0004] However, the above monitoring method based on fiber Bragg grating sensors has certain limitations. Since it mainly focuses on calculating the seepage rate by monitoring the change in metal ion concentration, it ignores other data affecting the underground seepage situation, resulting in low accuracy of predicting underground seepage. Summary of the Invention

[0005] In view of this, the embodiments of the present application provide a monitoring method and device for underground seepage, and the present application can improve the accuracy and reliability of monitoring the underground seepage situation.

[0006] In a first aspect, the embodiments of the present application provide a monitoring method for underground seepage, which specifically includes: Obtain a target historical seepage data set of a target underground area; the target historical seepage data set includes a plurality of historical seepage data subsets; the historical seepage data subsets include: a subset of seepage refractive index change data sequences, a subset of seepage pH sequences, a subset of seepage dissolved oxygen concentration sequences, and a subset of seepage pressure change data sequences; Perform feature selection on the target historical seepage data set to generate a seepage data feature set; Obtain the historical seepage state information corresponding to the seepage data feature set, and train an initial machine learning model based on the seepage data feature set and the historical seepage state information to generate a target seepage state prediction model; Input the real-time water seepage data set into the target water seepage state prediction model to obtain the water seepage state prediction result corresponding to the real-time water seepage data set.

[0007] As an optional implementation manner of an embodiment of the present application, the obtaining of the target historical water seepage data set of the target underground area includes: Obtain the data collected by the fiber Bragg grating sensor, pH sensor, dissolved oxygen sensor, and pressure sensor set for the target underground area; Clean the data in the initial historical water seepage data set to obtain the target historical water seepage data set.

[0008] As an optional implementation manner of an embodiment of the present application, the feature extraction of the historical water seepage data set to generate a water seepage data feature set includes: Use a target feature selection algorithm to perform feature selection on the data in the target historical water seepage data set to generate the water seepage data feature set.

[0009] As an optional implementation manner of an embodiment of the present application, the using of the target feature selection algorithm to perform feature selection on the data in the target historical water seepage data set to generate the water seepage data feature includes: Use the target feature selection algorithm to screen out the redundant features in the water seepage refractive index change data sequence subset, the water seepage pH sequence subset, the water seepage dissolved oxygen concentration sequence subset, and the water seepage pressure change data sequence subset; Delete the redundant features in the water seepage refractive index change data sequence subset, the water seepage pH sequence subset, the water seepage dissolved oxygen concentration sequence subset, and the water seepage pressure change data sequence subset to generate the water seepage data feature.

[0010] As an optional implementation manner of an embodiment of the present application, the method further includes: Use the Network Time Protocol to add timestamps to the data collected by the fiber Bragg grating sensor, the pH sensor, the dissolved oxygen sensor, and the pressure sensor; Obtain the real-time water seepage data set under the same timestamp, and perform data fusion analysis on the real-time water seepage data set based on the data fusion algorithm to obtain the operating states corresponding to the fiber Bragg grating sensor, the pH sensor, the dissolved oxygen sensor, and the pressure sensor respectively; Combine the operating states corresponding to the fiber Bragg grating sensor, the pH sensor, the dissolved oxygen sensor, and the pressure sensor respectively to perform abnormal diagnosis analysis on each sensor to obtain an abnormal diagnosis result; the abnormal diagnosis result is used to indicate that a target sensor is abnormal.

[0011] As an alternative implementation manner of an embodiment of the present application, the method further includes: Periodically calibrate the fiber grating sensor, the pH sensor, the dissolved oxygen sensor, and the pressure sensor.

[0012] As an alternative implementation manner of an embodiment of the present application, after inputting the real-time water seepage data set into the target water seepage state prediction model and obtaining the water seepage state prediction result corresponding to the real-time water seepage data set, the method further includes: Obtain the current water seepage risk level and the predicted value of the water seepage volume based on the water seepage state prediction result, and generate a warning prompt message based on the water seepage risk level and the predicted value of the water seepage volume.

[0013] In a second aspect, an embodiment of the present application provides a monitoring device for underground water seepage, including: An acquisition unit, configured to acquire a target historical water seepage data set of a target underground area; the target historical water seepage data set includes a plurality of historical water seepage data subsets; the historical water seepage data subsets include: a water seepage refractive index change data sequence subset, a water seepage pH sequence subset, a water seepage dissolved oxygen concentration sequence subset, and a water seepage pressure change data sequence subset; A selection unit, configured to perform feature selection on the target historical water seepage data set to generate a water seepage data feature set; A generation unit, configured to obtain historical water seepage state information corresponding to the water seepage data feature set, and train an initial machine learning model based on the water seepage data feature set and the historical water seepage state information to generate a target water seepage state prediction model; A prediction unit, configured to input the real-time water seepage data set into the target water seepage state prediction model to obtain the water seepage state prediction result corresponding to the real-time water seepage data set.

[0014] As an alternative implementation manner of an embodiment of the present application, the acquisition unit is specifically configured to acquire data collected by a fiber grating sensor, a pH sensor, a dissolved oxygen sensor, and a pressure sensor set for the target underground area; perform data cleaning on the data in the initial historical water seepage data set to obtain the target historical water seepage data set.

[0015] As an alternative implementation manner of an embodiment of the present application, the selection unit is specifically configured to perform feature selection on the data in the target historical water seepage data set by using a target feature selection algorithm to generate the water seepage data feature set.

[0016] As an alternative implementation manner of the embodiment of the present application, the selection unit is specifically configured to use the target feature selection algorithm to filter out redundant features in the water seepage refractive index change data sequence subset, the water seepage pH sequence subset, the water seepage dissolved oxygen concentration sequence subset, and the water seepage pressure change data sequence subset; delete the redundant features in the water seepage refractive index change data sequence subset, the water seepage pH sequence subset, the water seepage dissolved oxygen concentration sequence subset, and the water seepage pressure change data sequence subset to generate the water seepage data features.

[0017] As an alternative implementation manner of the embodiment of the present application, the acquisition unit is further configured to use the Network Time Protocol to add timestamps to the data collected by the fiber Bragg grating sensor, the pH sensor, the dissolved oxygen sensor, and the pressure sensor; obtain a real-time water seepage data set under the same timestamp, and perform data fusion analysis on the real-time water seepage data set based on a data fusion algorithm to obtain the operating states corresponding to the fiber Bragg grating sensor, the pH sensor, the dissolved oxygen sensor, and the pressure sensor respectively; combine the operating states corresponding to the fiber Bragg grating sensor, the pH sensor, the dissolved oxygen sensor, and the pressure sensor respectively to perform abnormal diagnosis analysis on each sensor to obtain an abnormal diagnosis result; the abnormal diagnosis result is used to indicate that an abnormal condition occurs in the target sensor.

[0018] As an alternative implementation manner of the embodiment of the present application, the monitoring device for underground water seepage further includes a calibration unit, which is specifically configured to periodically calibrate the fiber Bragg grating sensor, the pH sensor, the dissolved oxygen sensor, and the pressure sensor.

[0019] As an alternative implementation manner of the embodiment of the present application, the prediction unit is further configured to obtain the current water seepage risk level and the predicted value of the water seepage volume based on the water seepage state prediction result, and generate a warning prompt message based on the water seepage risk level and the predicted value of the water seepage volume.

[0020] In a third aspect, an embodiment of the present application provides an electronic device, including: a memory and a processor, where the memory is used to store a computer program; the processor is configured to, when executing the computer program, enable the electronic device to implement the monitoring method for underground water seepage according to any one of the above embodiments.

[0021] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a computing device, the computing device is enabled to implement the monitoring method for underground water seepage according to any one of the above embodiments.

[0022] The monitoring method for underground water seepage provided in the embodiment of the present application obtains a target historical water seepage data set for a target underground area; the target historical water seepage data set includes multiple historical water seepage data subsets; the historical water seepage data subsets include: a water seepage refractive index change data sequence subset, a water seepage pH sequence subset, a water seepage dissolved oxygen concentration sequence subset, and a water seepage pressure change data sequence subset; thereby, more comprehensive data on various factors affecting underground water seepage can be obtained, covering various key factors affecting underground water seepage. Compared with traditional monitoring methods that rely only on a single data set, this method can more comprehensively reflect the actual water seepage situation, provide rich information for subsequent model training, effectively reduce prediction bias, and improve the accuracy of water seepage status prediction.

[0023] Then, based on these data, a seepage data feature set is generated to provide more sufficient information for subsequent model training, so that the target seepage status prediction model can mine the complex nonlinear relationship between the data, and then more accurately capture the seepage law, effectively improve the accuracy of the prediction of underground seepage status, and reduce the prediction deviation caused by single data; and by obtaining the seepage status prediction results corresponding to the real-time seepage data set, it is possible to accurately predict the seepage status, which is helpful to timely warn when the seepage risk exceeds the safety range.

[0024] Ultimately, inputting the real-time water seepage dataset into the target model to obtain prediction results will enable operation and maintenance personnel to grasp the water seepage dynamics in a timely manner, and thus make scientific decisions quickly based on the water seepage status prediction results, effectively reducing the risk of safety accidents caused by water seepage, ensuring production safety, and reducing economic losses. BRIEF DESCRIPTION OF THE DRAWINGS The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0025] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0026] Figure 1 This is a flowchart of one of the steps of the underground water seepage monitoring method provided in an embodiment of the present application; Figure 2 This is a second flow chart of the steps of the method for monitoring underground water seepage provided in an embodiment of the present application; Figure 3 This is a flowchart of the third step of the method for monitoring underground water seepage provided in an embodiment of the present application; Figure 4The structural schematic diagram of the monitoring device for underground water seepage provided by the embodiment of the present application; Figure 5 The hardware structural schematic diagram of the electronic device provided by the embodiment of the present application. Detailed implementation manners

[0027] In order to more clearly understand the above objects, features and advantages of the present disclosure, the solutions of the present disclosure will be further described below. It should be noted that, without conflict, the embodiments of the present disclosure and the features in the embodiments may be combined with each other.

[0028] Many specific details are set forth in the following description in order to fully understand the present disclosure, but the present disclosure may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only a part of the embodiments of the present disclosure, rather than all the embodiments.

[0029] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design solution described as "exemplary" or "for example" in the embodiments of the present application should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Exactly speaking, the use of words such as "exemplary" or "for example" is intended to present relevant concepts in a specific manner. In addition, in the description of the embodiments of the present application, unless otherwise specified, the meaning of "a plurality of" refers to two or more.

[0030] It should be noted that, in this article, the term "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitations, the element defined by the statement "comprising one..." does not exclude the existence of additional identical elements in the process, method, article or device comprising the element.

[0031] The embodiment of the present application provides a monitoring method for underground water seepage. Referring to Figure 1 as shown, the monitoring method for underground water seepage includes the following steps S101-S104: S101. Obtain the target historical water seepage data set of the target underground area.

[0032] Among them, the target historical water seepage data set includes a plurality of historical water seepage data subsets; the historical water seepage data subsets include: a water seepage refractive index change data sequence subset, a water seepage pH sequence subset, a water seepage dissolved oxygen concentration sequence subset, and a water seepage pressure change data sequence subset.

[0033] During underground mining, for example, in the mining scenario of an underground mine, the original stratum structure will inevitably be damaged. Due to the damage to the integrity of the original rock, channels for groundwater migration are formed, making it easier for groundwater to flow into the mine construction site. At the same time, surface water bodies such as rivers and lakes in the vicinity will become recharge sources for groundwater under certain conditions. If there is a hydraulic connection between the current location of the underground mine and these surface water bodies, such as being connected through a permeable rock formation or a water-conducting channel, surface water may leak into the mine location, affecting subsequent normal production work.

[0034] Moreover, minerals and chemical substances in the underground seepage water will corrode construction equipment such as ventilators and drainage pumps, shortening the service life of the equipment, increasing maintenance costs, reducing equipment performance, affecting normal operation, increasing the mining difficulty, raising the mining cost, and reducing the mining efficiency.

[0035] Therefore, the embodiments of the present application provide the monitoring method for underground seepage water to further predict the seepage situation, and thus potentially detect latent seepage conditions in advance, reduce the risk of equipment damage, and reasonably arrange the mining plan.

[0036] Specifically, the target historical seepage water data set contains multiple historical seepage water data subsets; among them, the seepage water refractive index change data sequence subset is detected and obtained by a fiber Bragg grating sensor; the fiber Bragg grating sensor utilizes the chelation reaction of a coating layer with specific metal ions to increase the concentration of specific metal ions inside the fiber Bragg grating section, resulting in a change in the refractive index of the fiber Bragg grating area. This is because the change in metal ion concentration is often directly related to seepage water. When seepage occurs, groundwater carries various metal ions into the borehole, changing the metal ion concentration. By monitoring the change in the refractive index of the fiber Bragg grating area, it is possible to indirectly understand whether seepage occurs, the amount of seepage water, and the duration of seepage. When the amount of seepage water is large, more metal ions enter the borehole, resulting in a more obvious change in the refractive index, thus helping the staff to judge the severity of seepage.

[0037] The seepage water pH value sequence subset is detected and obtained by a pH sensor, which can be a glass electrode type pH sensor. It generates a potential difference based on the different hydrogen ion activities inside and outside the glass membrane and determines the pH value of the solution by measuring the potential difference; the ion-selective electrode type pH sensor has a selective response to specific ions and reflects the pH value by detecting the ion activity; it can directly measure the pH value in the seepage water, providing an important basis for water quality analysis; and it helps to judge whether the seepage water has a corrosive effect on the relevant equipment at the mining construction site.

[0038] The subset of the seepage dissolved oxygen concentration sequence is obtained by detection with a dissolved oxygen sensor, which can be an electrochemical dissolved oxygen sensor. The dissolved oxygen concentration in water is measured by detecting the current or potential change generated by the chemical reaction between the dissolved oxygen and the electrode, and then the dissolved oxygen concentration in the seepage is monitored in real time to reflect the ecological status and redox state of the water body.

[0039] The subset of the seepage pressure change data sequence is obtained by detection with a pressure sensor. Based on the physical changes (such as resistance and capacitance changes) generated by the pressure-sensitive element under the action of pressure, the pressure signal is converted into an electrical signal for measurement. The seepage pressure change can be monitored in real time, which helps to judge the source and flow direction of the seepage water and measure the impact of the water pressure on the mining construction site.

[0040] Then, the fiber Bragg grating sensor is used to monitor the refractive index change caused by the change of metal ion concentration in the borehole, indirectly judge the seepage situation, and further more accurately evaluate the seepage volume and seepage rate in combination with other sensor data. Specifically, the pH value measured by the pH sensor can reflect the acid-base characteristics of the seepage water and help judge its corrosiveness to the relevant equipment at the mining construction site. Acidic seepage water may accelerate the corrosion of metal equipment, while alkaline seepage water may affect the stability of the ore. The dissolved oxygen concentration detected by the dissolved oxygen sensor reflects the redox state of the water body, provides a basis for analyzing the chemical properties and biological activity of the seepage water, and the change of the dissolved oxygen content may imply the occurrence of microbial activities or chemical reactions, affecting the environment where the mining construction site is located. The seepage pressure change monitored by the pressure sensor can reveal the flow direction and water source of the seepage water. A sudden increase in pressure may mean a large amount of seepage water pouring in, which helps to judge the potential water inrush risk.

[0041] S102. Select features from the target historical seepage data set to generate a seepage data feature set.

[0042] In the embodiments of the present application, the target historical water seepage dataset contains a large amount of data. However, not all data is equally important for predicting the water seepage state, so feature selection is required. Features such as wavelength drift amount and drift rate are extracted from the subset of the water seepage refractive index change data sequence, and these features can reflect the key change trends measured by the fiber grating sensor. For the subset of the water seepage pH sequence, features such as the change trend and fluctuation amplitude of the pH value are extracted to show the dynamic change law of the pH value. Features such as the mean value, variance, and change rate of the dissolved oxygen concentration are extracted from the subset of the water seepage dissolved oxygen concentration sequence, and these features help to analyze the stability and change of the dissolved oxygen. Features such as the amplitude and frequency of the pressure change are extracted from the subset of the water seepage pressure change data sequence to reflect the change characteristics of the pressure. By using feature selection algorithms such as correlation analysis, principal component analysis (PCA), information gain method, and genetic algorithm, a feature subset meaningful for evaluating the water seepage state is screened out, redundant or irrelevant features are removed, and finally a water seepage data feature set is generated, enabling subsequent model training to focus on key information.

[0043] S103. Obtain the historical water seepage state information corresponding to the water seepage data feature set, and train the initial machine learning model based on the water seepage data feature set and the historical water seepage state information to generate a target water seepage state prediction model.

[0044] Specifically, the historical water seepage state information refers to the actual state when water seepage occurred in the past, such as the water seepage risk level, the amount of water seepage, and other information. The previously generated water seepage data features are combined with the corresponding historical water seepage state information and used as training data to be input into the initial machine learning model.

[0045] In some embodiments, common machine learning algorithms such as neural networks and decision trees can learn the complex non-linear relationship between the water seepage data features and water seepage disasters from a large amount of historical data through the above machine learning algorithms. During the training process, the model continuously adjusts its own parameters to optimize the prediction ability of the water seepage state. After multiple rounds of training, when the model shows good prediction performance on the training data, a target water seepage state prediction model is generated, and this model has the ability to predict the water seepage state according to the input new data.

[0046] It should be noted that the target seepage state prediction model obtained in the embodiments of the present application can be deployed in edge devices through an edge computer and run on a mine gateway. Compared with deploying the target seepage state prediction model in the cloud, it can reduce the latency and bandwidth occupation caused by cloud transmission, and reduce the inference response time. At the same time, the timestamps of the above-mentioned multiple data acquisition sensors can be synchronized through the Network Time Protocol (NTP), and millisecond-level data fusion and model inference can be achieved in combination with a streaming computing framework (such as Apache Flink) to ensure low-latency processing, achieve early warning timeliness, and generate a hierarchical early warning mechanism according to the seepage risk level and the predicted value of the seepage volume.

[0047] S104. Input the real-time seepage water data set into the target seepage state prediction model to obtain the seepage state prediction result corresponding to the real-time seepage water data set.

[0048] Furthermore, after the target seepage state prediction model is generated, a real-time seepage water data set is collected, that is, a data set obtained by real-time collection through various sensors (fiber Bragg grating sensors, pH sensors, dissolved oxygen sensors, pressure sensors, etc.) currently deployed in the target underground area. These real-time data are input into the trained target seepage state prediction model. The model will analyze and predict these real-time data according to the previously learned rules, and output the corresponding seepage state prediction results, such as the current seepage risk level, the predicted seepage volume, etc. Furthermore, the management personnel can timely understand the seepage situation according to the seepage state prediction results, so as to take corresponding measures to ensure safety. For example, when a high seepage risk is predicted, drainage operations can be arranged in advance or protective measures can be strengthened.

[0049] The monitoring method for underground seepage provided by the embodiments of the present application obtains a target historical seepage water data set of a target underground area; the target historical seepage water data set includes multiple historical seepage water data subsets; the historical seepage water data subsets include: a seepage refractive index change data sequence subset, a seepage pH sequence subset, a seepage dissolved oxygen concentration sequence subset, and a seepage pressure change data sequence subset; thus, more comprehensive relevant data on various influencing factors of underground seepage can be obtained, covering a variety of key factors affecting underground seepage. Compared with the traditional monitoring method that only relies on a single data, it can more comprehensively reflect the actual situation of seepage, provide rich information for subsequent model training, effectively reduce the prediction deviation, and improve the accuracy of seepage state prediction.

[0050] Then, based on these data, a seepage data feature set is generated to provide more sufficient information for subsequent model training, enabling the target seepage state prediction model to mine complex non-linear relationships between data, and thus being able to capture seepage patterns more accurately, effectively improving the accuracy of underground seepage state prediction and reducing prediction deviations caused by single data; moreover, by obtaining the seepage state prediction results corresponding to the real-time seepage data set, accurate prediction of the seepage state can be achieved, which helps to give early warnings in a timely manner when the seepage risk exceeds the safe range.

[0051] Finally, inputting the real-time seepage data set into the target model to obtain the prediction results can enable the operation and maintenance personnel to timely master the seepage dynamics, and thus, based on the seepage state prediction results, make scientific decisions quickly, effectively reducing the risk of safety accidents caused by seepage, ensuring production safety, and reducing economic losses. As an extension and refinement of the above embodiments, referring to Figure 2 as shown, the monitoring method for underground seepage provided by the embodiments of the present application further includes the following steps S201 to S206: S201. Obtain the data collected by the fiber Bragg grating sensor, pH sensor, dissolved oxygen sensor, and pressure sensor set for the target underground area.

[0052] It should be noted that it is very important to reasonably deploy the positions of the fiber Bragg grating sensor, pH sensor, dissolved oxygen sensor, and pressure sensor in the target underground area.

[0053] Specifically, the underground environment is complex, and the performance of each sensor needs to be considered; among them, when setting the fiber Bragg grating sensor, high-quality optical filters and optical isolators need to be used to ensure that only light of a specific wavelength enters the fiber Bragg grating sensor; this is because in underground mining, there are various external light sources, such as the light emitted by lighting equipment and other optical signal sources, and these lights will interfere with the signals of the fiber Bragg grating sensor, affecting the measurement accuracy. By using high-quality optical filters, specific wavelength light can be accurately selected according to the working wavelength characteristics of the fiber Bragg grating sensor, blocking other wavelength lights from entering the sensor and reducing the interference of stray light. The optical isolator can prevent reflected light, scattered light, etc. from entering the sensor reversely, further ensuring that only the required forward optical signal can be effectively received by the sensor, thereby reducing the interference of external light sources, ensuring that the optical signal received by the sensor is true and reliable, and improving the measurement accuracy. At the same time, a general-purpose coating layer can also be adopted in the fiber Bragg grating sensor, which can then undergo chelation reactions with various metal ions. No matter what combination of iron ions, copper ions, zinc ions, etc. exists in the seepage water, it can react with them, enabling the fiber Bragg grating sensor to work effectively in complex water quality environments in different positions and different regions, broadening the application range of the sensor.

[0054] The pH sensor is required to be corrosion-resistant and highly accurate, with a measurement range covering a pH value range of 2 - 12. Since the pH of the seepage water may fluctuate within this range, only by meeting this range can accurate measurements be made. If the measurement range is insufficient, it will lead to inaccurate data and be unable to reflect the real situation. The dissolved oxygen sensor requires high stability and anti-interference ability because the underground environment is complex, with many factors such as electromagnetic interference, which will affect the measurement accuracy of the sensor. At the same time, the electrode material of the sensor should be highly durable and anti-pollution because there are many impurities in the groundwater. If the electrode is easily polluted, it will reduce the service life and measurement accuracy of the sensor. The pressure sensor is required to be corrosion-resistant and highly accurate. The measuring range should be reasonably selected according to the actual depth of the groundwater, and it should be able to withstand an overload impact of at least twice the measuring range pressure to ensure stable operation in a complex water pressure environment.

[0055] At the same time, the underground environment is complex, with power equipment, communication equipment, etc. These devices will generate strong electromagnetic fields. The electromagnetic field will interfere with the signal transmission and processing of the sensor, and then cause signal distortion. The originally accurate measurement signal becomes distorted and cannot truly reflect the physical quantity being measured. It may also cause misjudgment, further affecting model training. Therefore, electromagnetic shielding technology can be adopted to install a shielding layer for the sensor and transmission cable. The shielding layer is usually made of metal materials with high magnetic permeability, such as copper, aluminum, or metal alloys. When an external electromagnetic field exists, these metal materials can guide the magnetic force lines of the electromagnetic field to flow along the surface of the shielding layer without penetrating the shielding layer and entering the interior of the sensor. For the sensor, installing a shielding layer can protect the generation, transmission, and processing processes of the internal optical signal, avoid the influence of electromagnetic interference on the optical signal, and ensure that the optical signal can accurately reflect the physical quantity being monitored, such as the change in the refractive index of the fiber grating area caused by the change in the concentration of metal ions. For the transmission cable, the shielding layer can prevent the external electromagnetic field from coupling to the optical signal inside the cable during the transmission process, ensure the accuracy of the measurement data during transmission, and enable the data collected from the sensor to be transmitted to the subsequent processing system completely and accurately.

[0056] The underground space is limited. Therefore, the installation space of the above-mentioned multiple sensors needs to be as compact as possible to facilitate installation and layout. A unified packaging shell can also be used to integrate the fiber Bragg grating sensor, pH sensor, dissolved oxygen sensor, and pressure sensor. The material of the packaging shell is selected as high-strength, corrosion-resistant, and waterproof stainless steel or high-strength engineering plastics, which can provide physical protection for the sensors and resist the harsh underground environment.

[0057] S202. Clean the data in the initial historical seepage water dataset to obtain the target historical seepage water dataset.

[0058] In the embodiments of the present application, due to various factors that may interfere with the sensors during data acquisition, such as electromagnetic interference in different environments, measurement errors of the sensors themselves, signal loss during data transmission, etc., there may be problems such as noise data, missing values, and outliers in the initial historical seepage water dataset. Furthermore, this will affect the accuracy and reliability of subsequent data analysis and model training. Therefore, it is necessary to clean the data in the initial dataset.

[0059] For the missing values in the initial dataset, different methods can be used for processing. If the missing values are few, the data records containing the missing values can be directly deleted; if the missing values are many, interpolation methods such as linear interpolation and polynomial interpolation can be used to estimate the missing values based on the values of adjacent data points; statistical methods such as filling the missing values with the mean, median, or mode can also be used.

[0060] For the outliers in the initial dataset, it is necessary to first identify the outliers in the data by setting a reasonable threshold range. For example, the range of normal data can be determined based on the mean and standard deviation of the data, and the data outside this range is an outlier. For the identified outliers, they can be corrected or deleted according to the actual situation. If the outliers are caused by sensor failures or data transmission errors, they are usually selected to be deleted; if the outliers may reflect real special situations, further analysis and verification are required before processing.

[0061] For the noise data in the initial dataset, filtering techniques such as moving average filtering and median filtering can be used to remove the noise interference in the data. These filtering methods can smooth the data curve and make the data more truly reflect the actual situation of groundwater.

[0062] After the above data cleaning operations, the noise data, missing values, and outliers in the initial historical seepage water dataset are effectively processed, and a target historical seepage water dataset with higher quality and accuracy is obtained. This dataset will be used as the basic data for subsequent feature selection, model training, etc., providing reliable support for accurately predicting the seepage state.

[0063] S203. Use the target feature selection algorithm to perform feature selection on the data in the target historical seepage water dataset to generate the seepage water data feature set.

[0064] In the embodiments of the present application, the specific implementation method of using the target feature selection algorithm to perform feature selection on the data in the target historical seepage water dataset to generate the seepage water data feature set can refer to the following steps 1 and 2: Step 1: Use the target feature selection algorithm to screen out redundant features in the subset of the water seepage refractive index change data sequence, the subset of the water seepage pH sequence, the subset of the water seepage dissolved oxygen concentration sequence, and the subset of the water seepage pressure change data sequence.

[0065] Specifically, the subset of the water seepage refractive index change data sequence, the subset of the water seepage pH sequence, the subset of the water seepage dissolved oxygen concentration sequence, and the subset of the water seepage pressure change data sequence are respectively from fiber Bragg grating sensors, pH sensors, dissolved oxygen sensors, and pressure sensors. These data sequence subsets contain a large amount of information about underground water seepage. However, there are usually redundant features in the original data, that is, the information carried by some features is largely repetitive. These redundant features not only increase the complexity of data processing but may also interfere with the subsequent data analysis model's learning of the true pattern.

[0066] Therefore, it is necessary to screen out redundant features in the target historical water seepage dataset based on the target feature selection algorithm. Among them, the target feature selection algorithm can be correlation analysis, principal component analysis (PCA), information gain method, genetic algorithm, etc., which are used to find redundant features in the data.

[0067] Step 2: Delete the redundant features in the subset of the water seepage refractive index change data sequence, the subset of the water seepage pH sequence, the subset of the water seepage dissolved oxygen concentration sequence, and the subset of the water seepage pressure change data sequence to generate the water seepage data features.

[0068] After using the target feature selection algorithm to determine the redundant features in each data sequence subset, these redundant features are deleted from the corresponding data sequence subsets. After the operation of deleting redundant features, the generated water seepage data features are more concise and effective. These features remove the noise and redundant information in the data, retain the key information closely related to the water seepage state, and can more accurately reflect the actual situation of water seepage; it provides high-quality data support for subsequent establishment of an accurate water seepage model, prediction of the water seepage risk level, analysis of the water seepage volume, etc., and helps to improve the accuracy and efficiency of data analysis.

[0069] S204: Obtain the historical water seepage state information corresponding to the water seepage data feature set, and train the initial machine learning model based on the water seepage data feature set and the historical water seepage state information to generate the target water seepage state prediction model.

[0070] The description of this step can refer to the description of step S103 above and will not be elaborated here.

[0071] S205: Input the real-time water seepage dataset into the target water seepage state prediction model to obtain the water seepage state prediction result corresponding to the real-time water seepage dataset.

[0072] For the description of this step, reference can be made to the description of step S104 above, which will not be elaborated here.

[0073] S206. Obtain the current water seepage risk level and the predicted value of the water seepage volume based on the water seepage state prediction result, and generate a warning message based on the water seepage risk level and the predicted value of the water seepage volume.

[0074] After obtaining the water seepage risk level and the predicted value of the water seepage volume, the system will generate a warning message according to the preset rules. If the water seepage risk level reaches a high level, such as serious accidents that may cause roadway collapse, water inrush, etc., the warning message will remind the staff in a prominent way, informing them of the severity of the risk and the areas that may be affected. At the same time, combined with the predicted value of the water seepage volume, specific data will be provided in the message, enabling the staff to clearly obtain the change in the water volume of the water seepage. So that the staff can quickly respond according to the warning message and take corresponding measures, such as strengthening drainage, etc., thereby effectively reducing the harm that the water seepage may bring and ensuring production safety.

[0075] As an extension and refinement of the above embodiments, with reference to Figure 3 as shown, the monitoring method for underground water seepage provided by the embodiments of the present application further includes the following steps S301 to S303: S301. Use the Network Time Protocol to add timestamps to the data collected by the fiber Bragg grating sensor, the pH sensor, the dissolved oxygen sensor, and the pressure sensor.

[0076] Specifically, using the Network Time Protocol (NTP) to add timestamps to the data collected by the fiber Bragg grating sensor, the pH sensor, the dissolved oxygen sensor, and the pressure sensor is to ensure the consistency of the data collected by different sensors in terms of time. If the data times are not synchronized, it will be impossible to accurately correlate various water seepage parameters at the same moment during subsequent analysis, affecting the comprehensive judgment of the water seepage situation.

[0077] Among them, NTP is a network protocol that enables each device in the network to synchronize with an accurate time source. Through NTP, each sensor synchronizes the time of the collected data with the standard time and adds a time mark. Furthermore, the collection time of the data collected by each sensor can be determined based on the timestamp, providing an accurate time reference for subsequent data fusion and analysis.

[0078] S302. Obtain the real-time water seepage data set under the same timestamp, and perform data fusion analysis on the real-time water seepage data set based on the data fusion algorithm to obtain the operating states corresponding to the fiber Bragg grating sensor, the pH sensor, the dissolved oxygen sensor, and the pressure sensor respectively.

[0079] In the embodiment of the present application, after obtaining the real-time water seepage data set at the same time stamp, analysis is performed based on the data fusion algorithm, which can integrate the data of different sensors to obtain the operating status corresponding to each sensor.

[0080] The data fusion algorithm can perform real-time comparative analysis on the data of each sensor and can quickly detect when the data of a certain sensor is significantly inconsistent with the data of other sensors. For example, under normal circumstances, the change trends of the data of the pH sensor, the dissolved oxygen sensor, and the pressure sensor are interrelated. If the data of the pH sensor suddenly deviates abnormally, the algorithm can capture it in time to avoid misjudging the water seepage situation due to incorrect data output by a single sensor failure, ensuring the reliability of the overall monitoring data.

[0081] S303. Combine the operating statuses corresponding to the fiber Bragg grating sensor, the pH sensor, the dissolved oxygen sensor, and the pressure sensor respectively to perform abnormal diagnosis and analysis on each sensor to obtain an abnormal diagnosis result; the abnormal diagnosis result is used to indicate that an abnormal situation occurs in the target sensor.

[0082] Performing abnormal diagnosis and analysis based on the operating statuses of each sensor can promptly detect the sensors that have abnormal situations. By comparing the current operating status of each sensor with the normal operating status, preset diagnosis rules and algorithms are used to determine whether the sensor is abnormal. If the measurement data of the fiber Bragg grating sensor fluctuates beyond the normal range and does not match the change trend of the data of other sensors, it may be determined as abnormal. The abnormal diagnosis result clearly points out the target sensor with an abnormal situation, which is crucial for ensuring the reliability of the monitoring system. Once a sensor is found to be abnormal, the staff can take timely measures, such as inspection, repair, or replacement of the sensor, to avoid incorrect monitoring data caused by sensor failure and affecting the judgment and decision-making of the water seepage situation.

[0083] Based on the same inventive concept, as an implementation of the above method, the embodiment of the present application further provides a monitoring device for underground water seepage. This embodiment corresponds to the foregoing method embodiment. For the convenience of reading, the details in the foregoing method embodiment will not be described one by one in this embodiment, but it should be clear that a monitoring device for underground water seepage in this embodiment can correspondingly implement all the contents in the foregoing method embodiment.

[0084] The embodiment of the present application provides a monitoring device for underground water seepage. Figure 4 As shown in the structural schematic diagram of the monitoring device for underground water seepage, Figure 4 as shown, the monitoring device 400 for underground water seepage includes: An acquisition unit 401, configured to acquire a target historical water seepage data set of a target underground area; the target historical water seepage data set includes a plurality of historical water seepage data subsets; the historical water seepage data subsets include: a water seepage refractive index change data sequence subset, a water seepage pH sequence subset, a water seepage dissolved oxygen concentration sequence subset, and a water seepage pressure change data sequence subset; A selection unit 402, configured to perform feature selection on the target historical water seepage data set to generate a water seepage data feature set; A generation unit 403, configured to acquire historical water seepage state information corresponding to the water seepage data feature set, and train an initial machine learning model based on the water seepage data feature set and the historical water seepage state information to generate a target water seepage state prediction model; A prediction unit 404, configured to input a real-time water seepage data set into the target water seepage state prediction model to obtain a water seepage state prediction result corresponding to the real-time water seepage data set.

[0085] As an optional implementation manner of an embodiment of the present application, the acquisition unit 401 is specifically configured to acquire data collected by a fiber Bragg grating sensor, a pH sensor, a dissolved oxygen sensor, and a pressure sensor arranged for the target underground area; perform data cleaning on the data in the initial historical water seepage data set to obtain a target historical water seepage data set.

[0086] As an optional implementation manner of an embodiment of the present application, the selection unit 402 is specifically configured to perform feature selection on the data in the target historical water seepage data set by using a target feature selection algorithm to generate the water seepage data feature set.

[0087] As an optional implementation manner of an embodiment of the present application, the selection unit is specifically configured to use the target feature selection algorithm to filter out redundant features in the water seepage refractive index change data sequence subset, the water seepage pH sequence subset, the water seepage dissolved oxygen concentration sequence subset, and the water seepage pressure change data sequence subset; delete the redundant features in the water seepage refractive index change data sequence subset, the water seepage pH sequence subset, the water seepage dissolved oxygen concentration sequence subset, and the water seepage pressure change data sequence subset to generate the water seepage data feature.

[0088] As an alternative implementation manner of an embodiment of the present application, the obtaining unit 401 is further configured to add time stamps to the data collected by the fiber Bragg grating sensor, the pH sensor, the dissolved oxygen sensor, and the pressure sensor by using the Network Time Protocol; obtain a real-time water seepage data set under the same time stamp, and perform data fusion analysis on the real-time water seepage data set based on a data fusion algorithm to obtain the operating states corresponding to the fiber Bragg grating sensor, the pH sensor, the dissolved oxygen sensor, and the pressure sensor respectively; combine the operating states corresponding to the fiber Bragg grating sensor, the pH sensor, the dissolved oxygen sensor, and the pressure sensor respectively, and perform abnormal diagnosis analysis on each sensor to obtain an abnormal diagnosis result; the abnormal diagnosis result is used to indicate that a target sensor is abnormal.

[0089] As an alternative implementation manner of an embodiment of the present application, the monitoring device for underground water seepage further includes a calibration unit, which is specifically configured to periodically calibrate the fiber Bragg grating sensor, the pH sensor, the dissolved oxygen sensor, and the pressure sensor.

[0090] As an alternative implementation manner of an embodiment of the present application, the prediction unit 404 is further configured to obtain the current water seepage risk level and the predicted value of the water seepage volume based on the water seepage state prediction result, and generate a warning prompt message based on the water seepage risk level and the predicted value of the water seepage volume.

[0091] Based on the same inventive concept, an embodiment of the present disclosure also provides an electronic device. Figure 5 The structural schematic diagram of the electronic device provided by the embodiment of the present disclosure is as Figure 5 shown. The electronic device provided by this embodiment includes: a memory 501 and a processor 502. The memory 501 is used to store a computer program; the processor 502 is configured to execute the monitoring method for underground water seepage provided by the above embodiment when executing the computer program.

[0092] Based on the same inventive concept, an embodiment of the present application also provides a computer-readable storage medium. A computer program is stored on the computer-readable storage medium. When the computer program is executed by a processor, the computing device is enabled to implement the monitoring method for underground water seepage provided by the above embodiment.

[0093] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code.

[0094] The processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0095] The memory may include non-permanent memory in the form of computer-readable media, random access memory (RAM) and / or non-volatile memory such as read-only memory (ROM) or flash RAM. The memory is an example of computer-readable media.

[0096] Computer-readable media includes both permanent and non-permanent, removable and non-removable storage media. The storage media can implement information storage by any method or technology, and the information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, Phase Change Memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read-Only Memory (ROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), flash memory or other memory technologies, Compact Disc Read-Only Memory (CD-ROM), Digital Versatile Disc (DVD) or other optical storage, magnetic cassette tapes, disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.

[0097] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A monitoring method for underground water seepage, characterized in that, Specifically include: Obtain the target historical seepage water dataset of the target underground area; The target historical seepage water dataset includes multiple historical seepage water data subsets; The historical seepage water data subsets include: a subset of seepage water refractive index change data sequences, a subset of seepage water pH sequences, a subset of seepage water dissolved oxygen concentration sequences, and a subset of seepage water pressure change data sequences; Perform feature selection on the target historical seepage water dataset to generate a seepage water data feature set; Obtain the historical seepage water state information corresponding to the seepage water data feature set, and train the initial machine learning model based on the seepage water data feature set and the historical seepage water state information to generate a target seepage water state prediction model; Input the real-time seepage water dataset into the target seepage water state prediction model to obtain the seepage water state prediction result corresponding to the real-time seepage water dataset.

2. The method according to claim 1, wherein The obtaining of the target historical seepage water dataset of the target underground area includes: Obtain the data collected by the fiber Bragg grating sensor, pH sensor, dissolved oxygen sensor, and pressure sensor set for the target underground area to generate an initial historical seepage water dataset; Perform data cleaning on the data in the initial historical seepage water dataset to obtain the target historical seepage water dataset.

3. The method according to claim 1, characterized in that, The performing of feature selection on the target historical seepage water dataset to generate a seepage water data feature set includes: Use the target feature selection algorithm to perform feature selection on the data in the target historical seepage water dataset to generate the seepage water data feature set.

4. The method according to claim 3, wherein The using of the target feature selection algorithm to perform feature selection on the data in the target historical seepage water dataset to generate the seepage water data features includes: Use the target feature selection algorithm to screen out the redundant features in the subset of seepage water refractive index change data sequences, the subset of seepage water pH sequences, the subset of seepage water dissolved oxygen concentration sequences, and the subset of seepage water pressure change data sequences; Delete the redundant features in the subset of seepage water refractive index change data sequences, the subset of seepage water pH sequences, the subset of seepage water dissolved oxygen concentration sequences, and the subset of seepage water pressure change data sequences to generate the seepage water data features.

5. The method according to claim 2, wherein The method further includes: Use the Network Time Protocol to add timestamps to the data collected by the fiber Bragg grating sensor, the pH sensor, the dissolved oxygen sensor, and the pressure sensor; Obtain the real-time seepage water dataset at the same timestamp, and perform data fusion analysis on the real-time seepage water dataset based on the data fusion algorithm to obtain the operating states corresponding to the fiber Bragg grating sensor, the pH sensor, the dissolved oxygen sensor, and the pressure sensor respectively; Combine the operating states corresponding to the fiber Bragg grating sensor, the pH sensor, the dissolved oxygen sensor, and the pressure sensor respectively to perform abnormal diagnosis analysis on each sensor to obtain an abnormal diagnosis result; the abnormal diagnosis result is used to prompt the occurrence of a fault in the target sensor.

6. The method according to claim 2, wherein The method further includes: Periodically calibrate the fiber Bragg grating sensor, pH sensor, dissolved oxygen sensor, and pressure sensor.

7. The method according to claim 1, characterized in that After inputting the real-time water seepage data set into the target water seepage state prediction model and obtaining the water seepage state prediction result corresponding to the real-time water seepage data set, the method further includes: Obtaining the current water seepage risk level and the predicted value of the water seepage volume based on the water seepage state prediction result, and generating a warning message based on the water seepage risk level and the predicted value of the water seepage volume.

8. A monitoring device for underground water seepage, characterized in that, Including: An acquisition unit for acquiring a target historical water seepage data set of a target underground area; The target historical water seepage data set includes a plurality of historical water seepage data subsets; The historical water seepage data subset includes: a water seepage refractive index change data sequence subset, a water seepage pH sequence subset, a water seepage dissolved oxygen concentration sequence subset, and a water seepage pressure change data sequence subset; A selection unit for performing feature selection on the target historical water seepage data set to generate a water seepage data feature set; A generation unit for obtaining the historical water seepage state information corresponding to the water seepage data feature set, and training an initial machine learning model based on the water seepage data feature set and the historical water seepage state information to generate a target water seepage state prediction model; A prediction unit for inputting the real-time water seepage data set into the target water seepage state prediction model to obtain the water seepage state prediction result corresponding to the real-time water seepage data set.

9. An electronic device, characterized in that, Including: A memory and a processor, the memory is used for storing a computer program; the processor is used for, when executing the computer program, enabling the electronic device to implement the monitoring method for underground water seepage according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, and when the computer program is executed by a computing device, the computing device is enabled to implement the monitoring method for underground water seepage according to any one of claims 1-7.

Images