Automatic monitoring system and method for abnormal fluctuation data of underground water level

Abstract

The invention relates to the technical field of groundwater environment monitoring, in particular to an automatic monitoring system and method for groundwater level abnormal fluctuation data, the system comprises a cloud platform and a monitoring station, a monitoring module is arranged in a monitoring well, a sensor array is integrated in the monitoring module, and the monitoring well is further provided with a processing module for cleaning the sensor array; according to the invention, the monitoring module is arranged in the monitoring well, the sensor array integrated in the monitoring module is used for data acquisition, when one sensor has abnormal data, normal data still exist under the same time sequence, and the reason of abnormal fluctuation data of the sensor array is considered. Through fusion of other normal data, normal data capable of reflecting normal characterization under the time sequence is obtained, and compared with a conventional homogenization method, prediction processing and the like, the method has more obvious data characterization.

Description

Technical Field

[0001] The present invention relates to the technical field of groundwater environment monitoring, and in particular to an automatic monitoring system and method for abnormal groundwater level fluctuation data. Background Art

[0002] Groundwater resources are part of the earth's water resources. They are closely related to atmospheric precipitation resources and surface water resources, and they transform into each other. They play an important role in social economy and ecological environment. The protection of groundwater resources directly affects the production and life of the people. The lack of water resources and the intensification of the contradiction between supply and demand, especially the lagging management, will lead to the aggravation of the over-exploitation of groundwater resources and the continuous decline of groundwater levels. Many wells in well irrigation areas have been abandoned, and the water output has dropped significantly, which has caused a series of ecological and environmental problems such as ground subsidence and land desertification. Groundwater monitoring is an important basis for understanding and mastering the dynamic characteristics of groundwater, formulating reasonable development and utilization and effective protection measures, and preventing and reducing groundwater pollution and related ecological and environmental problems.

[0003] In the process of groundwater level data monitoring, transmission and storage, some abnormal data will inevitably be generated due to the influence of environment, monitoring instruments and human factors. In the overall range, the extreme data far away from most of the data are statistical abnormal data. The main causes of groundwater level abnormal data are statistical anomalies and environmental anomalies. For statistical anomalies, since the groundwater level belongs to a time series, its changes are continuous. When the monitor array exists in the monitoring environment for a long time, the groundwater environment will cause different degrees of physical impact on the monitor array, thereby affecting the monitoring accuracy. For statistical anomalies, the monitoring algorithm and model are limited by the data. At one point, the conventional solutions are elimination, regression analysis, filtering technology, etc., but there is a problem of missing corresponding data in the time series.

[0004] Therefore, there is an urgent need for an automatic monitoring system and method for abnormal groundwater level fluctuation data, which can reduce the abnormal data caused by environmental factors through physical means, and reduce the impact of missing corresponding time series data on monitoring accuracy by processing abnormal data. Summary of the invention

[0005] The present invention aims to provide an automatic monitoring system for abnormal groundwater level fluctuation data, so as to realize automatic monitoring of groundwater data through automated monitoring stations, and to clean the monitor array through the monitoring stations to maintain monitoring accuracy; and also to provide an automatic monitoring method for abnormal groundwater level fluctuation data, so as to reduce the influence of the missing corresponding data of the time series on the monitoring accuracy by processing the abnormal data.

[0006] The present invention is achieved through the following technical solutions:.

[0007] An automatic monitoring system for abnormal groundwater level fluctuation data comprises a cloud platform and a plurality of monitoring stations with monitoring wells, wherein a monitoring module is arranged in the monitoring well, wherein a sensor array for monitoring groundwater data is integrated in the monitoring module, and the monitoring well is also provided with a processing module for cleaning the sensor array; the monitoring station comprises: a control module connected to the processing module signal, wherein the control module is used to control the position of the monitoring module in the monitoring well and also to control the processing process of the sensor array; a fusion module connected to the sensor array signal, used to determine and eliminate abnormal groundwater level fluctuation data, and to obtain fused data by fusing data collected by the sensor array, wherein the fused data corresponds to the time series of the eliminated abnormal fluctuation data; and a communication module connected to the fusion module and the cloud platform signal, and used to realize information interaction.

[0008] The present invention sets a monitoring module in the monitoring well and uses the sensor array integrated in the monitoring module to collect data. The sensor array in this embodiment is preferably different types of sensors of the same type for collection. For example, when monitoring water level changes, for water level sensors, it is preferred to use four types of sensors, namely, pressure water level gauge, float water level gauge, capacitive water level gauge, and ultrasonic water level gauge, which are arranged in order to form an array for collection and have the same sampling interval. When abnormal data appears in one of the sensors, normal data still exists in the same time series. Taking into account the reasons for the abnormal fluctuation data in the sensor array, in the present invention, the remaining normal data are fused to obtain normal data that can reflect the normal characterization in the time series. Compared with conventional homogenization methods, predictive processing, etc., the data characterization is more obvious.

[0009] An automatic monitoring method for abnormal groundwater level fluctuation data comprises the following steps: step 1, data acquisition, lowering a monitoring module to the water surface of a monitoring well through a lifting unit, and controlling the monitoring module to a set position of an acquisition mode; step 2, data processing, after a fusion module receives the data collected by a sensor array in step 1, it determines and removes abnormal groundwater level fluctuation data, and obtains fused data by fusing data collected by at least three other groups of sensor arrays, wherein the fused data corresponds to a time series of the removed abnormal fluctuation data; step 3, data interaction, sending the fused data in step 2 and the collected normal data to a cloud platform through a communication module.

[0010] The present invention is achieved by

[0011] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0012] 1. The present invention sets a monitoring module in the monitoring well and uses the sensor array integrated in the monitoring module to collect data. The sensor array in this embodiment is preferably different types of sensors of the same type for collection. For example, when monitoring water level changes, for water level sensors, it is preferred to use pressure water level gauges, float water level gauges, capacitive water level gauges, and ultrasonic water level gauges. These four sensors are arranged in order to form an array for collection, and have the same sampling interval. When one of the sensors has abnormal data, normal data still exists in the same time series. Considering the reason for the abnormal fluctuation data in the sensor array, in the present invention, the remaining normal data are fused to obtain normal data that can reflect the normal characteristics of the time series. Compared with conventional homogenization methods, prediction processing, etc., the data representation is more obvious.

[0013] 2. The judgment unit in the present invention includes a decomposition subunit, a calculation subunit, a fusion subunit and a judgment subunit. First, the decomposition subunit uses the ensemble empirical mode decomposition method to perform noise reduction processing on the data set collected by the sensor array. EEMD is an adaptive signal processing method that can effectively extract trends and periodic components from complex data while reducing the impact of noise. In this way, the noise in the original data is effectively separated, providing a clearer data basis for subsequent anomaly detection;

[0014] 3. In the present invention, a method of combining data from multiple sensors to obtain more accurate and comprehensive information than a single sensor. In the context of groundwater level monitoring, due to environmental factors and limitations of the sensor itself, the data of a single sensor may be interfered with or produce errors. Data fusion technology can reduce such interference and errors by integrating data from multiple sensors, and improve the accuracy and reliability of monitoring results. The fusion unit evaluates the central tendency and dispersion of data by calculating information functions of different sensor data, such as mean or variance. Then, it uses these statistics to weight and integrate the data of each sensor, and finally obtains a comprehensive data representation, which can better reflect the actual situation of the groundwater level. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of this application, and do not constitute a limitation of the embodiments of the present invention. In the drawings:

[0016] Figure 1 It is a structural schematic diagram of the system of the present invention;

[0017] Figure 2 It is a schematic diagram of the internal structure of the monitoring station of the present invention;

[0018] Figure 3 This is a schematic diagram of the cross-sectional structure of the monitoring well;

[0019] Figure 4 It is a schematic diagram of the coordination status of the lifting unit, the monitoring module and the closure;

[0020] Figure 5 It is a schematic diagram of the matching status of the monitoring module and the closure;

[0021] Figure 6 The structural diagram of the monitoring module

[0022] Figure 7 It is a schematic diagram of the cross-sectional structure of the monitoring module;

[0023] Figure 8 It is a schematic diagram of the internal structure of the monitoring module;

[0024] Fig. 9 is another schematic diagram of the internal structure of the monitoring module;

[0025] Fig.10 is a schematic diagram of the internal structure of the closure;

[0026] Fig.11 Flow chart of the method of the present invention.

[0027] Marks and corresponding parts names in the attached drawings:

[0028] 1- monitoring station, 2- monitoring well, 3- monitoring module, 4- processing module;

[0029] 31-sensor array, 32-monitoring base, 33-submersible, 34-slideway, 35-limiting plate, 36-limiting slot, 37-fixing claw;

[0030] 311-water level sensor, 312-electromagnet;

[0031] 41-processing base, 42-lifting unit, 43-cleaning unit, 44-closing member, 45-water outlet;

[0032] 431-closing cylinder, 432-cleaning nozzle;

[0033] 441 - closed base, 442 - closed disc, 443 - closed groove, 444 - closed leaf, 445 - limit rod. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical scheme and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments and drawings. The schematic implementation modes and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention. It should be noted that the present invention is already in the actual development and use stage.

[0035] In order to make the technical solution of the present invention clearer, the professional terms involved in the present invention are explained here;

[0036] In the present invention, the term "cloud platform" refers to a remote server or service network used to store, process and analyze data from monitoring stations. The cloud platform allows centralized management and access to data, enabling remote monitoring and data sharing of the monitoring system, including data storage systems, computing resources and network services to support functions such as big data processing, real-time analysis and early warning systems;

[0037] In the present invention, the term "sensor array" refers to different types of sensors that are evenly arranged according to a certain layout and interval to achieve comprehensive monitoring of the environment in the monitoring well. The sensor array can provide a continuous and real-time data stream to provide basic data for monitoring and analysis of groundwater levels; in the present invention, sensors include temperature sensors, water level sensors, etc., and water level sensors include pressure water level gauges, float water level gauges, capacitive water level gauges, ultrasonic water level gauges, bubble water level gauges, etc.;

[0038] In the present invention, the term "control module" refers to a component used to control the position of the monitoring module in the monitoring well and the processing process of the sensor array. It communicates with the processing module through signal connection and is responsible for the lifting, lowering, cleaning and other operations of the monitoring module to ensure that the sensor array can accurately and effectively collect data. In the present invention, the control module is preferably an STM32 single-chip microcomputer, and the communication module preferably adopts the industrial-grade China Mobile 4G serial port module ML302.

[0039] In the present invention, the term "time series" refers to the time period corresponding to each data, and each data is usually collected within a fixed time interval;

[0040] In the present invention, the term “Raida’s criterion” refers to a statistical method for outlier detection;

[0041] In the present invention, the term "isolation forest algorithm" refers to a machine learning algorithm for anomaly detection, which identifies anomalies in data by constructing multiple isolated trees. In the present invention, the isolation forest algorithm is used to calculate the path length of each tree, and the anomaly score value is obtained by normalizing these path lengths, thereby assisting in identifying abnormal fluctuations in groundwater level data;

[0042] In the present invention, the term "information function" refers to a mathematical function used to describe the characteristics of sensor data, such as a mean function or a variance function, which can provide information about the distribution and variability of data and is used to weight and integrate data from different sensors during the data fusion process to obtain more accurate and reliable monitoring results;

[0043] As for the other necessary components of the monitoring station, it is preferred to be equipped with an off-grid photovoltaic power supply system and a battery for energy storage. The other structures will not be described in detail here.

[0044] Embodiment 1:

[0045] Please refer to the attached Figures 1 to 10 An automatic monitoring system for abnormal groundwater level fluctuation data includes a cloud platform and a plurality of monitoring stations 1 with monitoring wells 2. The monitoring wells 2 are provided with monitoring modules 3, and the monitoring modules 3 are integrated with a sensor array 31 for monitoring groundwater data. The monitoring wells 2 are also provided with a processing module 4 for cleaning the sensor array 31.

[0046] The monitoring station 1 comprises:

[0047] A control module connected to the processing module 4 by signals, the control module is used to control the position of the monitoring module 3 in the monitoring well 2, and is also used to control the processing process of the sensor array 31;

[0048] A fusion module, connected to the sensor array 31 signal, used to determine and remove abnormal fluctuation data of the groundwater level, and obtain fused data by fusing the data collected by the sensor array 31, wherein the fused data corresponds to the time series of the removed abnormal fluctuation data;

[0049] The communication module is connected with the fusion module and the cloud platform signal to realize information interaction.

[0050] It should be noted that the main causes of abnormal groundwater level data are statistical anomalies and environmental anomalies. For statistical anomalies, since the groundwater level belongs to a time series and its changes are continuous, when the monitor array exists in the monitoring environment for a long time, the groundwater environment will cause different degrees of physical impact on the monitor array. Specifically, these include corrosion problems, biological contamination, chemical deposition, electrolyte contamination and other problems, which will cause the sensor to produce abnormal data; for statistical anomalies, the monitoring algorithms and models that are limited by the data are used. At one point, the conventional solutions are elimination, regression analysis, filtering technology, etc., but there is a problem of missing corresponding data in the time series. Specifically, taking the removal of interpolation as an example, most of the processing methods are to remove outliers. There is also a method of using the data near the outliers for prediction processing to estimate the corresponding data, but in this method, the source of the predicted data still belongs to the same data set, and the credibility of the "normal data" attached to the outliers is questionable. Therefore, there is still uncertainty.

[0051] Based on the above problems, the applicant has proposed an automatic monitoring system for abnormal groundwater level fluctuation data, by setting a monitoring module 3 in the monitoring well 2, and using the sensor array 31 integrated in the monitoring module 3 to collect data. The sensor array 31 in this embodiment is preferably different types of sensors of the same type for collection. For example, when monitoring water level changes, for the water level sensor 311, it is preferred to use four types of sensors, namely, a pressure water level gauge, a float water level gauge, a capacitive water level gauge, and an ultrasonic water level gauge, and arrange them in an orderly manner to form an array for collection, and have the same sampling interval. When abnormal data appears in one of the sensors, normal data still exists in the same time series. Taking into account the reasons for the abnormal fluctuation data in the sensor array 31, in the present invention, the remaining normal data are fused to obtain normal data that can reflect the normal representation of the time series. Compared with conventional homogenization methods, predictive processing, etc., the data representation is more obvious.

[0052] For the present invention, firstly, using different types of sensors of the same type for data collection can improve the reliability and accuracy of the data. In groundwater level monitoring, due to environmental factors such as corrosion, biological contamination, chemical deposition and electrolyte contamination, a single sensor may generate abnormal data. By integrating multiple types of sensors, such as pressure water level gauges, float water level gauges, capacitive water level gauges and ultrasonic water level gauges, the system monitors the same parameter from multiple angles, thereby reducing the impact of single sensor failure or error on the overall monitoring results. The setting of this multi-sensor array 31 improves the redundancy and robustness of the monitoring data and ensures the continuity and integrity of the data.

[0053] Secondly, when a sensor has abnormal data, the system will not simply eliminate this data, but through data fusion technology, use the data of other normal sensors in the same time series to fuse them to obtain normal data that can represent the time series. This fusion process not only retains the integrity of the time series data, but also improves the representativeness of the data. In this way, the system can reduce the information loss caused by the elimination of abnormal data, and avoid the uncertainty that may be caused by using data in the same data set for predictive processing.

[0054] During the working process, the fusion module first receives the data collected by the sensor array 31, and then uses the ensemble empirical mode decomposition in the judgment unit to perform noise reduction on the data set. Next, the calculation subunit uses the isolation forest algorithm and the local outlier factor (LOF) algorithm to calculate the outlier score value and the local outlier factor, and normalizes these values. The fusion subunit performs weighted fusion on these normalized values ​​to obtain the fusion score. Finally, the judgment subunit uses the Raida criterion to judge the fusion score and identify outliers.

[0055] When an abnormal value is detected, the fusion unit will call at least three sets of normal data in the same time series for fusion. This process involves information functions, such as mean function or variance function, and the maximum and minimum AD in adjacent time series of different sensors. Through this fusion algorithm, the system can integrate data from multiple sensors to obtain a more accurate and reliable normal data value, thereby improving the accuracy and efficiency of groundwater level monitoring.

[0056] Embodiment 2:

[0057] This embodiment only describes the parts that are different from Embodiment 1. Specifically, the fusion module includes a judgment unit, and the judgment unit includes:

[0058] A decomposition subunit, connected to the sensor array signal, uses ensemble empirical mode decomposition to reduce noise on the data set collected by the sensor array;

[0059] The calculation subunit is connected to the decomposition subunit signal, and uses the isolation forest algorithm to construct multiple isolated trees, and calculates the path length of each isolated tree, and normalizes the average path length of the calculation result to obtain the abnormal score value; the LOF algorithm is also used to calculate the local abnormal factor of each sample, and the calculated local abnormal factor is normalized;

[0060] The fusion subunit is connected to the calculation subunit signal, and performs weighted fusion of the abnormal score value calculated in the calculation subunit and the normalized local outlier factor value to obtain a fusion score;

[0061] The judging subunit is connected to the fusion subunit signal and uses the Raida criterion to judge the fusion score. The judging process is as follows: the mean and standard deviation of the fusion score are calculated, and the fusion score that deviates from the mean by more than 3 times the standard deviation is recorded as an outlier.

[0062] The normalized calculation in the calculation subunit satisfies: Among them, LOF nor is the normalized local outlier factor value of the i-th sample; LOF(i) is the local outlier factor value of the i-th sample;

[0063] The fusion process in the fusion subunit satisfies: F(S, LOF nor )=ωα×S+ωβ×LOF nor ,

[0064] Among them, S is the anomaly score of the isolation forest; α and β are fusion weights; ω is the neighborhood size;

[0065] The determination process of the determination subunit satisfies: Here, σ is the standard deviation.

[0066] It should be noted that the judgment unit in the present invention includes a decomposition subunit, a calculation subunit, a fusion subunit and a judgment subunit. First, the decomposition subunit uses the ensemble empirical mode decomposition method to perform noise reduction processing on the data set collected by the sensor array. EEMD is an adaptive signal processing method that can effectively extract trends and periodic components from complex data while reducing the impact of noise. In this way, the noise in the original data is effectively separated, providing a clearer data basis for subsequent anomaly detection.

[0067] Next, the computing subunit uses the isolation forest algorithm to construct multiple isolated trees and calculates the path length of each tree. Isolation forest is an efficient anomaly detection algorithm that "isolates" observations by randomly selecting features and cutoff values. Since the number of anomalies is small, they are easily isolated, so the path length is short. The computing subunit also uses the local anomaly factor algorithm to calculate the local anomaly factor of each sample; the LOF algorithm identifies anomalies by evaluating the density difference between a sample and its neighboring samples. Samples with large density differences are considered abnormal. The calculation results of these two algorithms are normalized to obtain anomaly score values ​​and local outlier factor values. Then, the fusion subunit performs weighted fusion on the anomaly score value obtained in the computing subunit and the local outlier factor value obtained by normalization to obtain a fusion score. This step combines the advantages of the two algorithms. Through weighted fusion, the abnormality of data points can be more comprehensively evaluated, which improves the accuracy of anomaly detection.

[0068] Finally, the judgment subunit uses the Raida criterion to judge the fusion score. The Raida criterion is a statistical anomaly detection method. By calculating the mean and standard deviation of the fusion score, the fusion score that deviates from the mean by more than 3 times the standard deviation is recorded as an outlier. The above method is simple and intuitive, easy to implement, and can effectively identify outliers.

[0069] For the judgment unit, its working principle is based on statistical learning and machine learning algorithms. Through multi-step data processing and feature extraction, it comprehensively evaluates the abnormality of the data. This method can not only handle statistical anomalies, but also identify environmental anomalies, such as abnormal data caused by physical damage to the sensor or environmental changes.

[0070] Compared with the prior art, the judgment unit in this embodiment has the following advantages:

[0071] Compared with other machine learning algorithms, such as support vector machines or neural networks, the isolation forest algorithm has higher computational efficiency and can quickly identify anomalies on large data sets; EEMD, as an adaptive signal processing method, can process nonlinear and non-stationary data and has strong adaptability; by combining the isolation forest and LOF algorithms, the system can evaluate the abnormality of data from different angles, thereby improving the accuracy of anomaly detection; the Raida criterion is based on statistical principles, has good robustness to noise and outliers, and can stably identify anomalies; by weighted fusion of the results of different algorithms, the system can integrate multiple information, thereby improving the comprehensiveness and reliability of anomaly detection; the system of the present invention is designed for real-time monitoring of abnormal groundwater level fluctuation data, and can respond to and handle abnormal situations in a timely manner, which is of great significance for the protection and management of groundwater resources.

[0072] Embodiment 3:

[0073] This embodiment only describes the parts that are different from the second embodiment. Specifically, the fusion module includes a fusion unit connected to the determination subunit signal, the sensor array includes a number of water level sensors that are evenly spaced and used to collect groundwater levels in monitoring wells, and when an abnormal value appears in the water level sensor, the fusion unit calls at least three groups of normal data in the same time series for fusion, and the fusion process satisfies:

[0074] f i (x) = [a min +(a max -a min ) i (x)]A i (x)+[b min +(b max -b min )b i (x)]B i (x)+[c min +(c max -c min )c i (x)]C i (x);

[0075] Among them, A i (x), B i (x), C i (x) are information functions of three sets of normal data, including mean function or variance function;

[0076] a max 、a min They are the maximum and minimum AD values ​​of a water level sensor in adjacent time series respectively;

[0077] b max 、b minare the maximum and minimum AD values ​​of another water level sensor in adjacent time series respectively;

[0078] b max 、b min They are the maximum and minimum AD values ​​of another water level sensor in adjacent time series;

[0079] a i (x), b i (x), c i (x) are the AD value curves of the three water level sensors;

[0080] f i (x) is the normal data after fusion.

[0081] It should be noted that the data fusion process of the fusion unit starts with the data collected by the sensor array. In this embodiment, these data include readings from different types of sensors, such as pressure water level gauges, float water level gauges, capacitive water level gauges, and ultrasonic water level gauges. When an abnormal value appears in a sensor, the fusion unit will call at least three sets of normal data in the same time series for fusion. This process involves information functions, such as mean functions or variance functions, and the maximum and minimum AD values ​​of different sensors in adjacent time series. Specifically, the fusion unit will calculate the AD value curves of the three water level sensors. These curves represent the maximum and minimum values ​​of different sensors in adjacent time series. Through these values, the fusion unit can construct a comprehensive data curve that integrates the data characteristics of each sensor, thereby obtaining a more accurate and reliable normal data value.

[0082] The working principle of the fusion unit is based on data fusion technology, which combines data from multiple sensors to obtain more accurate and comprehensive information than a single sensor. In the context of groundwater level monitoring, due to environmental factors and the limitations of the sensor itself, the data of a single sensor may be disturbed or produce errors. Data fusion technology can reduce such interference and errors by integrating data from multiple sensors, and improve the accuracy and reliability of monitoring results. The fusion unit evaluates the central tendency and dispersion of the data by calculating information functions of different sensor data, such as mean or variance. Then, it uses these statistics to weight and integrate the data of each sensor, and finally obtains a comprehensive data representation, which can better reflect the true situation of the groundwater level.

[0083] Compared with the prediction method in the prior art, the fusion unit in this embodiment has the following advantages:

[0084] Traditional prediction methods may use other data in the same data set for prediction after removing outliers, which may lead to deviations in the prediction results. The fusion unit retains the integrity of the time series data by fusing data from multiple sensors and avoids information loss caused by removing outliers.

[0085] The fusion unit does not rely solely on the data from a single sensor, but integrates the data characteristics of multiple sensors, which makes the fused data more representative and can more accurately reflect the real changes in groundwater levels.

[0086] Since the fusion unit uses the normal data of multiple sensors in the same time series for fusion, rather than relying on the data near the outliers for prediction, the uncertainty of the prediction results is reduced.

[0087] The data fusion process of the fusion unit is resistant to the impact of single sensor failure or error, because even if a sensor fails, the system can still rely on the data of other sensors for accurate monitoring.

[0088] Embodiment 4:

[0089] This embodiment only describes the parts that are different from the embodiment 1. Specifically, the processing module 4 includes:

[0090] A processing base 41 is arranged at the wellhead of the monitoring well 2;

[0091] A lifting unit 42, which is disposed in the processing base 41 and is used to change the position of the monitoring module 3 on the water level;

[0092] A cleaning unit 43, disposed in the processing base 41 and used for cleaning the sensor array 31;

[0093] The cleaning unit 43 comprises:

[0094] A sealing cylinder 431, wherein the diameter of the sealing cylinder 431 decreases from top to bottom, and a sealing member 44 is disposed at the upper end, and the lower end is fixed to the bottom of the processing base 41;

[0095] A cleaning nozzle 432 disposed on the inner wall of the processing base 41, wherein the cleaning nozzle 432 and the sealing member 44 are both connected to the control module signal;

[0096] When the lifting unit 42 lifts the monitoring module 3 to the upper end of the sealing member 44, the control module controls the sealing member 44 to close the upper opening of the sealing cylinder 431, and controls the cleaning nozzle 432 to clean the monitoring module 3. A water outlet 45 is also provided at the bottom end of the processing base 41.

[0097] It should be noted that the processing base 41 is a rectangular cover structure as a whole and is arranged at the wellhead position of the monitoring well 2. It can play a relatively sealing role, keep the internal environment of the monitoring well 2 relatively independent, and not be affected by external environmental factors. In this embodiment, the lifting mechanism is preferably a winch, which also has a motor connected to the control module signal. An electromagnet 312 is provided at the end of its winch rope, and it is detachably connected to the upper end of the monitoring module 3 through the electromagnet 312 to realize the movement of the monitoring module 3 on the water surface. The use environment of the cleaning unit 43 is suitable for situations where the water quality environment in the monitoring well 2 is more complicated, such as when there are more halogens and the chemical deposition is more serious. Specifically, the closing cylinder 431 in this embodiment is preferably an inverted funnel shape, and the lower end is fixedly connected to the bottom of the processing base 41, and the upper end is fixedly set with the closing member 44, and the end of the processing base 41 is also provided with an outlet 45 for sewage discharge. When the cleaning unit 43 is working, the control module controls the lifting unit 42 to lift the monitoring module 3 to above the closing piece 44, and starts the closing piece 44 to close the upper end of the closing cylinder 431 to prevent the sewage after cleaning from flowing into the monitoring well 2, and then starts the cleaning nozzle 432 to clean the monitoring module 3. It can be understood that the cleaning nozzle 432 can be adjusted in angle and controlled by the control module. The cleaning water source of the cleaning nozzle 432 is preferably stored pure water.

[0098] The monitoring module 3 includes a monitoring base 32, in which a submersible 33 for changing the height of the monitoring module 3 in the groundwater of the monitoring well 2 is arranged. The sensor array 31 is arranged below the monitoring base 32. A plurality of slide grooves 34 are opened in the monitoring base 32. A limit plate 35 is arranged in the monitoring base 32. The limit plate 35 is connected to a motor connected to the control module signal through a gear. A plurality of limit grooves 36 are arranged on the limit plate 35. A fixing claw 37 is arranged in the slide groove 34 and the fixing claw 37 cooperates with the limiting groove 36 through a pin. When the limiting disk 35 rotates, it can drive the fixing claw 37 to slide in the slide groove 34 and abut against the inner wall of the monitoring well 2.

[0099] It should be noted that, based on the above structure, when the motor rotates, the limiting plate 35 and the limiting groove 36 can force the fixed claw 37 to reciprocate in the slide groove 34. The fixed claw 37 has a rod-shaped structure matching the slide groove 34, and an arc block is arranged at the end of the rod-shaped structure. The monitoring module 3 is fixed in the water of the monitoring well 2 through the squeezing effect of all the arc blocks, and the monitoring operation of water layers at different depths is realized through the action of the submersible 33. More specifically, the monitoring module 3 includes a monitoring base 32, in which a submersible 33 is installed, which is used to change the height of the monitoring module 3 in the groundwater of the monitoring well 2. The sensor array 31 is installed below the monitoring base 32 and directly contacts the groundwater to monitor the water level change in real time. In order to achieve accurate positioning of the monitoring module 3 in the well, a plurality of slide grooves 34 are designed inside the monitoring base 32, and fixed claws 37 are installed in these slide grooves 34. A limit plate 35 is also provided in the monitoring base 32, and the limit plate 35 is connected to a motor connected to the control module signal through gears. A plurality of limit grooves 36 are provided on the limit plate 35, and the fixed claws 37 cooperate with the limit grooves 36 through pins. When the limit plate 35 rotates, it can drive the fixed claws 37 to slide in the slide grooves 34, and then the fixed claws 37 abut against the inner wall of the monitoring well 2, thereby fixing the monitoring module 3.

[0100] The closing member 44 includes a closing base 441 fixedly connected to the closing cylinder 431, a closing disk 442 is rotatably arranged in the closing base 441, a plurality of closing grooves 443 are provided on the closing disk 442, a motor connected to the control module signal is also arranged in the closing base 441, the motor is connected to the closing disk 442 through gears, a plurality of closing leaves 444 are also arranged in the closing base 441, limiting rods 445 are provided on the upper and lower end surfaces of the closing leaves 444, and the limiting rods 445 cooperate with the openings in the closing grooves 443 and the closing base 441. When the motor is started, the closing leaves 444 are opened and closed by the rotation of the closing disk 442.

[0101] It should be noted that the closure member 44 is composed of a closure base 441 and a closure disk 442, and the closure base 441 is fixedly connected to the closure cylinder 431; the closure disk 442 is rotatably arranged in the closure base 441, and a plurality of closure grooves 443 are provided thereon, and these grooves are used to guide the opening and closing action of the closure leaves 444; a motor is installed in the closure base 441, and the motor is connected to the control module through a signal line, and the motor is connected to the closure disk 442 through a gear, thereby driving the rotation of the closure disk 442; a plurality of closure leaves 444 are also installed in the closure base 441, and limit rods 445 are provided on the upper and lower end surfaces of each closure leaf 444; these limit rods 445 cooperate with the closure grooves 443 and the openings in the closure base 441, so that when the motor is started, the rotation of the closure disk 442 can drive the opening and closing of the closure leaves 444, thereby realizing the opening and closing functions of the closure member 44.

[0102] The working principle of the closure member 44 is based on the combination of mechanical transmission and control signals. The motor receives a signal from the control module and starts or stops according to the command of the signal. The rotation of the motor is transmitted to the closure disk 442 through the gear, so that the closure disk 442 rotates. The closure groove 443 on the closure disk 442 cooperates with the limiting rod 445 on the closure leaf 444. When the closure disk 442 rotates, the limiting rod 445 moves in the closure groove 443, driving the closure leaf 444 to open or close. This design enables the closure leaf 444 to accurately control the opening of the closure cylinder 431 to prevent foreign matter from entering or allow water to flow through when necessary. The opening and closing action of the closure leaf 444 is accurate and reliable, ensuring the safety and functionality of the monitoring module 3 under different working conditions.

[0103] Embodiment 5:

[0104] This embodiment only describes the parts that are different from the embodiment 1, specifically, as shown in the attached Fig.11 As shown, a method for automatically monitoring abnormal groundwater level fluctuation data includes the following steps:

[0105] Step 1, data collection, lowering the monitoring module 3 to the water surface of the monitoring well 2 through the lifting unit 42, and controlling the monitoring module 3 to the set position of the collection mode;

[0106] Step 2, data processing, after receiving the data collected by the sensor array 31 in step 1, the fusion module determines and removes the abnormal fluctuation data of the groundwater level, and obtains fused data by fusing the data collected by at least three other groups of sensor arrays 31, wherein the fused data corresponds to the time series of the removed abnormal fluctuation data;

[0107] Step 3: data interaction: the fused data in step 2 and the collected normal data are sent to the cloud platform through the communication module.

[0108] Specifically, the specific steps of the method are as follows:

[0109] Step 1, data collection: This step involves using the lifting unit 42 to accurately lower the monitoring module 3 to the water surface of the monitoring well 2, and controlling the monitoring module 3 to reach the preset collection mode position. This process requires precise control of the movement of the lifting unit 42 to ensure that the monitoring module 3 can reach the correct monitoring depth and start data collection at the appropriate position.

[0110] Step 2, data processing: After data collection is completed, the fusion module starts to work and receives the data collected by the sensor array 31. The main task of the fusion module is to determine and eliminate abnormal fluctuation data, which may be caused by sensor failure, environmental interference or other abnormal factors. After eliminating the abnormal data, the fusion module will use at least three groups of data collected by the normal sensor array 31 for fusion processing to obtain fused data. This step ensures that even if there are problems with some sensor data, it can be supplemented and corrected by the data of other normal sensors, so as to obtain accurate groundwater level data.

[0111] Step 3, data interaction: The processed fusion data and normal data will be sent to the cloud platform through the communication module. This step involves the remote transmission of data, and the cloud platform can further analyze and store the data, while providing data sharing and remote access functions.

[0112] Based on the above steps, the automatic control technology ensures that the monitoring module 3 can accurately reach the monitoring location and start data collection, while the data fusion technology is used to improve the accuracy and reliability of the data. By eliminating abnormal data and using multi-sensor data fusion, this method can provide more stable and accurate groundwater level monitoring results. The communication module is responsible for transmitting these results to the cloud platform to achieve remote management and analysis of data.

[0113] The monitoring module 3 includes a surface mode and an underwater mode, and the underwater mode includes a suspended mode and a fixed mode; the surface mode is that the monitoring module 3 floats on the water surface of the monitoring well 2; the suspended mode is that the monitoring module 3 is suspended in the water of the monitoring well 2 through a submersible 33; the fixed mode is that the monitoring module 3 is fixed on the inner wall of the monitoring well 2 through a fixing claw 37.

[0114] For the water surface mode, in the water surface mode, the monitoring module 3 is designed to float on the water surface of the monitoring well 2. In this mode, the monitoring module 3 is easy to deploy and recover, and is suitable for quickly monitoring water level changes on the water surface. Since the monitoring module 3 is in direct contact with the water surface, it can monitor the rise and fall of the water level in real time, and it also has a good monitoring effect on environmental factors such as floating objects and temperature changes on the water surface. In addition, the monitoring module 3 in the water surface mode does not require a complicated lowering and lifting mechanism, which reduces equipment costs and maintenance difficulties.

[0115] For the suspension mode, in the suspension mode, the monitoring module 3 is suspended at a specific depth in the water of the monitoring well 2 through the submersible 33. This mode enables the monitoring module 3 to monitor at any depth of the water body without being restricted by the water surface and the bottom of the well. The suspension mode is suitable for situations where monitoring is required in a specific water layer, such as monitoring the concentration of pollutants or temperature changes at a specific depth in the water. In addition, the monitoring module 3 in the suspension mode can reduce the error caused by contact with the well wall and improve the accuracy of the monitoring data.

[0116] For the fixed mode, in the fixed mode, the monitoring module 3 is fixed to the inner wall of the monitoring well 2 by the fixing claws 37. This mode allows the monitoring module 3 to perform long-term and stable monitoring at a specific position of the well wall. The fixed mode is suitable for situations where it is necessary to continuously monitor the groundwater level changes, permeability and other parameters near the well wall. Since the monitoring module 3 is fixed to the well wall, it can reduce the error caused by water flow disturbance and improve the stability and reliability of the data.

[0117] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An automatic monitoring system for abnormal groundwater level fluctuation data, comprising a cloud platform and a plurality of monitoring stations (1) with monitoring wells (2), characterized in that: The monitoring well (2) is provided with a monitoring module (3), wherein a sensor array (31) for monitoring groundwater data is integrated in the monitoring module (3), and the monitoring well (2) is also provided with a processing module (4) for cleaning the sensor array (31). The monitoring station (1) comprises: a control module connected to the processing module (4) by signals, the control module being used to control the position of the monitoring module (3) in the monitoring well (2) and also being used to control the processing process of the sensor array (31); A fusion module, connected to the sensor array (31) signal, used to determine and remove abnormal fluctuation data of the groundwater level, and to obtain fused data by fusing the data collected by the sensor array (31), wherein the fused data corresponds to the time series of the removed abnormal fluctuation data; The communication module is connected with the fusion module and the cloud platform signal to realize information interaction.

2. The automatic monitoring system for abnormal groundwater level fluctuation data according to claim 1 is characterized by: The fusion module includes a judgment unit, and the judgment unit includes: A decomposition subunit is connected to the sensor array (31) signal and uses collective empirical mode decomposition to reduce noise on the data set collected by the sensor array (31); The calculation subunit is connected to the decomposition subunit signal, and uses the isolation forest algorithm to construct multiple isolated trees, and calculates the path length of each isolated tree, and normalizes the average path length of the calculation result to obtain the abnormal score value; the LOF algorithm is also used to calculate the local abnormal factor of each sample, and the calculated local abnormal factor is normalized; The fusion subunit is connected to the calculation subunit signal, and performs weighted fusion of the abnormal score value calculated in the calculation subunit and the normalized local outlier factor value to obtain a fusion score; The judging subunit is connected to the fusion subunit signal and uses the Raida criterion to judge the fusion score. The judging process is as follows: the mean and standard deviation of the fusion score are calculated, and the fusion score that deviates from the mean by more than 3 times the standard deviation is recorded as an outlier.

3. The automatic monitoring system for abnormal groundwater level fluctuation data according to claim 2 is characterized by: The normalized calculation in the calculation subunit satisfies: in, LOF nor is the normalized local outlier factor value of the i-th sample; LOF(i) is the local outlier factor value of the i-th sample; The fusion process in the fusion subunit satisfies: F(S,LOF nor )=ωα×S+ωβ×LOF nor , Where S is the anomaly score of the isolation forest; α and β are fusion weights; ω is the neighborhood size; The determination process of the determination subunit satisfies: Here, σ is the standard deviation.

4. The automatic monitoring system for abnormal groundwater level fluctuation data according to claim 2 is characterized by: The fusion module includes a fusion unit connected to the judgment subunit signal, the sensor array includes a number of water level sensors evenly spaced and used to collect groundwater levels in monitoring wells, when an abnormal value appears in the water level sensor, the fusion unit calls at least three groups of normal data in the same time series for fusion, and the fusion process satisfies: f i (x)=[a min +(a max -a min )a i (x)]A i (x)+[b min +(b max -b min )b i (x)]B i (x)+[c min +(c max -c min )c i (x)]C i (x); Among them, A i (x), B i (x), C i (x) are information functions of three sets of normal data, including mean function or variance function; a max 、a min They are the maximum and minimum AD values ​​of a water level sensor in adjacent time series respectively; b max , b min are the maximum and minimum AD values ​​of another water level sensor in adjacent time series respectively; b max , b min They are the maximum and minimum AD values ​​of another water level sensor in adjacent time series; a i (x), b i (x), c i (x) are the AD value curves of the three water level sensors; f i (x) is the normal data after fusion.

5. The automatic monitoring system for abnormal groundwater level fluctuation data according to claim 1 is characterized by: The processing module (4) comprises: A processing base (41) is arranged at the wellhead of the monitoring well (2); A lifting unit (42) is arranged in the processing base (41) and is used to change the position of the monitoring module (3) on the water level; A cleaning unit (43) is disposed in the processing base (41) and is used to clean the sensor array (31); The cleaning unit (43) comprises: A closed cylinder (431), the diameter of which decreases from top to bottom, and a closed member (44) is provided at the upper end, and the lower end is fixed to the bottom of the processing base (41); A cleaning nozzle (432) is arranged on the inner wall of the processing base (41), and the cleaning nozzle (432) and the sealing member (44) are both connected to the control module signal; When the lifting unit (42) lifts the monitoring module (3) to the upper end of the closing member (44), the control module controls the closing member (44) to close the upper end opening of the closing cylinder (431), and controls the cleaning nozzle (432) to clean the monitoring module (3). A water outlet (45) is also provided at the bottom end of the interior of the processing base (41).

6. The automatic monitoring system for abnormal groundwater level fluctuation data according to claim 5 is characterized by: The monitoring module (3) comprises a monitoring base (32), wherein a submersible (33) for changing the height of the monitoring module (3) in the groundwater of the monitoring well (2) is arranged in the monitoring base (32), wherein the sensor array (31) is arranged below the monitoring base (32), wherein a plurality of slide grooves (34) are provided in the monitoring base (32), wherein a limit plate (35) is arranged in the monitoring base (32), wherein the limit plate (35) is connected to a motor connected to a control module signal via a gear, wherein a plurality of limit grooves (36) are arranged on the limit plate (35), wherein a fixing claw (37) is arranged in the slide groove (34) and the fixing claw (37) cooperates with the limiting groove (36) via a pin, wherein when the limiting disk (35) rotates, the fixing claw (37) can be driven to slide in the slide groove (34) and abut against the inner wall of the monitoring well (2).

7. The automatic monitoring system for abnormal groundwater level fluctuation data according to claim 5 is characterized by: The closure member (44) comprises a closure base (441) fixedly connected to the closure cylinder (431), a closure disk (442) rotatably arranged in the closure base (441), a plurality of closure grooves (443) being provided on the closure disk (442), a motor connected to a control module signal being also arranged in the closure base (441), the motor being connected to the closure disk (442) via a gear, a plurality of closure leaves (444) being also arranged in the closure base (441), limiting rods (445) being provided on both upper and lower end surfaces of the closure leaves (444), and the limiting rods (445) being matched with the closure grooves (443) and the openings in the closure base (441), and when the motor is started, the closure leaves (444) are opened and closed by the rotation of the closure disk (442).

8. An automatic monitoring method for abnormal groundwater level fluctuation data, characterized in that: An automatic monitoring system for abnormal groundwater level fluctuation data according to claim 6 or 7 comprises the following steps: Step 1, data collection, lowering the monitoring module (3) to the water surface of the monitoring well (2) through the lifting unit (42), and controlling the monitoring module (3) to a set position of the collection mode; Step 2, data processing, after receiving the data collected by the sensor array (31) in step 1, the fusion module determines and removes the abnormal fluctuation data of the groundwater level, and obtains fused data by fusing the data collected by at least three other groups of sensor arrays (31), wherein the fused data corresponds to the time series of the removed abnormal fluctuation data; Step 3: data interaction: the fused data in step 2 and the collected normal data are sent to the cloud platform through the communication module.

9. The automatic monitoring method of abnormal groundwater level fluctuation data according to claim 8 is characterized by: The monitoring module (3) includes a surface mode and an underwater mode, and the underwater mode includes a suspended mode and a fixed mode; The water surface mode is that the monitoring module (3) floats on the water surface of the monitoring well (2); The suspension mode is that the monitoring module (3) is suspended in the water of the monitoring well (2) through the submersible (33); The fixing mode is that the monitoring module (3) is fixed on the inner wall of the monitoring well (2) by means of fixing claws (37).

Images