An automatic monitoring system and method for groundwater level

Abstract

This invention discloses an automatic groundwater level monitoring system and method. Based on real-time monitoring values ​​of the groundwater level in the construction area and / or weather forecast information, it predicts groundwater level changes and controls drainage operations based on the measured and / or predicted results. Drainage control is implemented according to on-site drainage capacity and user plan requirements. The drainage control method includes any combination of measured value control, predicted total water level increase control, economic drainage control, demand-based drainage control, and maximum drainage control. This invention avoids the untimely drainage and construction seepage problems easily caused by traditional single manual drainage control, achieving groundwater monitoring and automatic drainage in the work area, meeting the safety requirements of construction sites, especially deep foundation pit construction, and improving the intelligent control level of smart construction site systems.

Description

Technical Field

[0001] This invention belongs to the field of smart water conservancy construction site technology, specifically relating to an automatic groundwater monitoring system and method. Background Technology

[0002] During the construction of water conservancy projects, groundwater level monitoring equipment and drainage equipment are generally installed in areas with abundant and complex groundwater. When the monitored water level exceeds or is about to exceed the preset threshold, the drainage facilities are manually or automatically activated to carry out drainage operations in order to avoid accidents such as leakage at the work site, especially in deep foundation pit operations.

[0003] Current groundwater level monitoring measures mostly require regular manual inspection and operation. Even with automatic drainage systems, drainage operations are only initiated or stopped by comparing the monitored water level with a preset threshold. During prolonged or sudden rainfall, untimely drainage can still lead to seepage in the foundation pit, affecting project quality and schedule. Especially when abnormal underground water sources are present, failure to promptly identify the source can significantly reduce drainage effectiveness, again impacting project quality and schedule. Summary of the Invention

[0004] To address the above problems, this invention designs an automatic groundwater monitoring system and method. By introducing weather forecasting, groundwater source direction judgment, water level prediction, and drainage control measures, it achieves intelligent monitoring of groundwater levels, which has significant practical significance for the construction of smart construction sites, especially water conservancy construction sites.

[0005] This invention discloses an automatic groundwater level monitoring method, characterized by real-time monitoring and prediction of groundwater level changes in the construction area, and control of drainage operations based on measured and / or predicted results. This includes setting water level thresholds, initiating drainage when the upper threshold is reached or exceeded, and stopping drainage when the lower threshold is reached or below. The groundwater level change prediction includes predicting groundwater level changes based on precipitation information and / or based on continuous measured groundwater level information, determining the direction of the groundwater source, and controlling drainage according to drainage capacity and user needs to ensure no leakage in the work area, such as deep foundation pits. The precipitation information can be obtained through weather forecasts. The prediction of groundwater level changes based on continuous measured groundwater level information includes filtering prediction of groundwater level changes using model filtering technology, which can employ mature technologies. The precipitation includes rainfall, snowfall, hail, etc., with rainfall generally being the primary focus of the design prediction.

[0006] Furthermore, the automatic groundwater level monitoring method also includes the following steps:

[0007] S1: Groundwater level monitoring point layout and real-time monitoring: At least one monitoring point shall be set up in the designated monitoring area to collect data and monitor in real time; the monitoring point shall be equipped with a groundwater level monitoring instrument, such as a pore water pressure gauge, mainly used for monitoring the rise and fall of groundwater level; if it is necessary to determine the direction of water source, at least two monitoring points shall be symmetrically arranged in the designated monitoring area, or at least four or more monitoring points shall be evenly and symmetrically arranged; the layout of monitoring points is generally a one-time operation, and the real-time monitoring state shall be entered after the equipment is debugged and normal.

[0008] S2: Predicting the normal growth rate of groundwater level using a water level prediction method: The water level prediction method includes using a prediction model and a filtering prediction algorithm to predict the recent water level growth rate and / or when the water level will reach a first water level threshold, i.e., a safe water level limit, by analyzing continuous groundwater level monitoring data; the water level monitoring data used for predicting the normal growth rate of groundwater level is collected during periods without drainage operations, i.e., groundwater level normal growth rate prediction and statistics are generally performed during periods without drainage operations; when water level changes are not significant, at least one day's worth of water level monitoring data is required, with a sampling period between 1 minute and 30 minutes; when water level changes are rapid, using... Data within several hours is also used for prediction calculations, with a sampling period of 1 second to 1 minute. The sampling period of the original water level monitoring data in the water level prediction method generally needs to ensure continuous sampling within the effective prediction period and that the number of samples is not less than 3. This is a basic requirement for model filtering. In fact, filtering usually starts to stabilize when the number of continuous sampling samples is not less than 10, and the filtering can maintain a stable state when the number of samples is more than 50. The prediction of the recent water level growth rate includes the prediction of the water level growth rate at the next sampling time after the current sampling time or at any sampling time. Generally, the prediction is mainly made at the next sampling time. In this step, the determination of the direction of the groundwater source can be initiated according to the user's needs.

[0009] The prediction model includes at least one of the following: constant velocity model, constant acceleration model, Singer model, "current" model, and polynomial model. The filtering prediction algorithm includes the Kalman filter prediction algorithm. Using the prediction model and the Kalman filter prediction algorithm to filter and predict continuously sampled data is a mature technology.

[0010] The constant velocity model essentially uses the average velocity calculated over the most recent period as the subsequent predicted velocity. Since groundwater levels generally change slowly and can be considered to increase at a uniform rate in the short term, the constant velocity model can meet the needs of predicting the rate of increase of groundwater levels.

[0011] S3: Prediction of the impact of precipitation on the rate of increase in groundwater level. This is achieved by obtaining local future precipitation information through the Internet or the Internet of Things. The precipitation information includes at least one combination of rainfall amount and duration, or rainfall amount and duration per unit time, or total rainfall within a certain period. This information can be obtained through weather forecasts. The rainfall amount includes light rain, light to moderate rain, moderate rain, moderate to heavy rain, heavy rain, heavy to torrential rain, torrential rain, torrential rain to extremely heavy torrential rain, and extremely heavy torrential rain. Specific rainfall values ​​can be found in the definition of rainfall by the National Meteorological Administration. This is combined with empirical statistical values ​​of the ground infiltration coefficient after rainfall, or the measured statistical relationship between rainfall (water) per unit time and the rate of increase in groundwater level. For details, please refer to the "Geotechnical Engineering Test and Monitoring Manual" to predict the impact of rainfall on the rate of increase in groundwater level.

[0012] S4: Short-term groundwater level anomaly prediction. Groundwater levels generally change slowly and almost stop changing after reaching a certain level. However, when abnormal leakage or seepage occurs, the rate of water level change can significantly exceed expectations, easily causing foundation pit leakage. The term "short-term" generally refers to minutes to several hours, requiring a smaller water level sampling interval, such as 1 minute, or a manually set shorter interval, such as 1 second to 60 seconds. Short-term groundwater level anomalies include groundwater level anomalies without drainage and groundwater level anomalies with drainage. The water level anomaly generally refers to an abnormal increase in water level; this invention also includes an abnormal decrease in water level, both of which can be monitored, predicted, and alarmed in real time. The short-term groundwater level anomaly prediction includes water level prediction, difference judgment, and confirmation prediction using a water level prediction method. The difference judgment includes determining whether the difference exceeds a preset first water level increase threshold. If it does, the sampling period is reduced, the sampling frequency is increased, and resampling and confirmation prediction are performed. The confirmation prediction includes using new encrypted sampling data for re-prediction, outputting or alarming the prediction result. The sampling data generally consists of no less than 10 sampling periods.

[0013] Regarding the difference judgment, when there is no rainfall and no drainage, the difference = |real-time predicted value - normal growth rate|; when there is rainfall and no drainage, the difference = |real-time predicted value - normal growth rate - predicted value affected by rainfall|; when there is no rainfall and drainage, the difference = |real-time predicted value - normal growth rate + drainage volume per unit time|; when there is rainfall and drainage, the difference = |real-time predicted value - normal growth rate - predicted value affected by rainfall + drainage volume per unit time|.

[0014] The normal growth rate includes any one of the following: the predicted normal growth rate of water level, the measured normal growth rate, the statistical mean of the predicted normal growth rate of water level, and the statistical mean of the measured normal growth rate of water level. There is no essential difference between the four, but the consistency of the standard should be noted in engineering applications.

[0015] The water level prediction method includes filtering prediction based on at least three consecutive frames of sampled data. It predicts the recent water level growth rate by analyzing continuous groundwater level monitoring data, using the same method as the prediction of normal groundwater level growth rate. It also initiates a water source direction judgment program based on user needs. If there is no water level anomaly or no need for short-term groundwater level anomaly prediction, this step can be omitted, and the process can proceed directly to the next step of drainage control.

[0016] S5: Drainage control, starting the drainage pump according to real-time control instructions to perform drainage operation, the drainage control method includes any combination of measured value control, predicted total water level growth rate control, economic drainage control, demand-driven drainage control, and maximum drainage control;

[0017] The measured value control includes: when the measured water level reaches the first water level threshold, drainage is automatically started, and the number of drainage pumps started and the drainage volume per unit time of each pump should ensure that the total drainage volume per unit time is greater than the predicted total water level growth rate; when the measured water level reaches the second water level threshold, drainage is stopped; the total water level growth rate is the sum of the normal growth rate, the groundwater level growth rate affected by rainfall, and the groundwater level growth rate affected by abnormal water level, or it may only consider the abnormal water level growth rate and ignore other minor factors;

[0018] The overall design concept of drainage control is as follows: When the predicted total rate of increase in groundwater level is less than or equal to the economic drainage rate, control is implemented according to the measured water level. That is, when the measured water level reaches the first water level threshold, drainage is automatically started; when the measured water level reaches the second water level threshold, drainage stops. When the predicted total rate of increase in groundwater level is greater than the economic drainage rate, it should be ensured that during the period of sustained high water level increase, such as the duration of rainfall, the increase in groundwater level does not exceed the water level control threshold, which includes the first water level threshold H1 or the third water level threshold H3. The calculation is limited to the predicted duration of abnormal water level. If the water level no longer increases abnormally beyond this time, there is no longer any drainage pressure, and drainage can proceed according to normal principles. If the predicted duration is too short, manual intervention is required. If the economic drainage rate cannot achieve this goal, the drainage rate needs to be increased, which is called demand-driven drainage control. Demand-driven drainage control should be implemented immediately after confirmation. If the predicted total rate of increase in water level or the drainage rate V3 calculated by demand-driven drainage control exceeds the total drainage capacity of the drainage pump, drainage should be started immediately and an alarm should be triggered. Generally, the maximum drainage capacity should be activated. If there is a clear forecast regarding the duration (T) of a sustained high groundwater level, the forecast can be used as a reference. If there is no clear forecast, such as when temporary leakage is unclear, a longer period, such as 1 to 3 days, can generally be used for calculation.

[0019] S6: Real-time water level monitoring and prediction during drainage, including temporarily adjusting the water level monitoring sampling frequency before drainage starts, reducing the sampling time interval (e.g., 1 minute), or manually setting a shorter sampling interval (e.g., 1s-60s); after drainage starts, continuously collecting no less than 10 frames of water level sampling data, initiating water level prediction, and displaying the prediction results, including the water level increase rate and when the second water level threshold (i.e., the lower limit of the water level, at which point drainage can be stopped) is reached. The water level increase rate during drainage should ideally be negative, which can also be referred to as water level deceleration; if the predicted water level increase rate is greater than 0, an alarm is triggered, and the drainage volume should generally be increased until the water level increase rate... If the predicted time to reach the second water level threshold is too long, the drainage volume can be increased. Increasing the drainage volume includes increasing the pump pressure (or increasing the speed) and / or increasing the number of drainage pumps started. The displayed prediction results also include displaying the measured growth rate and the predicted growth rate after one hour, as well as the predicted time for the water level to reach the third water level threshold. If the measured water level reaches the second water level threshold and the total water level growth rate is not significantly abnormal, the drainage control ends and returns to step S1 to enter normal real-time monitoring. Otherwise, continue to complete step S6. The significant abnormality includes the total water level growth rate continuously exceeding the normal growth rate and surpassing the first water level growth rate threshold.

[0020] Furthermore, the monitoring method also includes determining the direction of the groundwater source, the determination method comprising:

[0021] A method for determining the source of groundwater seepage involves, for interconnected groundwater levels, deploying at least one groundwater level monitoring sensor in each of at least two opposite directions, or one in each of four or more opposite directions within the interconnected area. Short-cycle high-frequency sampling technology is used to acquire water level monitoring data from each sensor. The sampling period is 1 to 60 seconds. Longer sampling periods can cause the water level changes to be averaged out, making it difficult to discern the sequence of water level changes at different monitoring points. Therefore, short-cycle high-frequency sampling technology should be used, and the direction of the water source should be determined based on the phase of the water level change curves from each sensor. Generally, the direction of the sensor where the water level rises first or at the fastest rate is the main source of groundwater seepage. The direction of the water source is typically determined after step S2 or S4 based on user requirements, or a sensor monitoring the flow direction can be used to determine the direction of the groundwater source.

[0022] Furthermore, the method for predicting the impact of precipitation on the rate of increase in groundwater level includes:

[0023] Let V be the predicted rainfall per unit time. y The duration of rainfall is T yLet σ be the surface permeability coefficient, ε be the influence coefficient of surface runoff and soil water retention, and 0 ≤ ε ≤ 1. Generally, when the surface hardening rate is high and the drainage network is well-developed, the influence coefficient ε can be taken as a smaller value, such as 0.1. For soft soil, such as cultivated land or wasteland, the influence coefficient ε can be taken as a larger value, such as 0.9 or 1.0. Then the permeability η is...

[0024] η=εσ / V y

[0025] When the calculation result η>1, then take η=1;

[0026] The predicted impact V4 of rainfall on the rate of increase in groundwater level is:

[0027] V4 = V y η;

[0028] The total water level height increment H4 has the following effect:

[0029] H4 = V4T y

[0030] When only forecast information such as light rain, moderate rain, heavy rain, rainstorm, and torrential rain is available, the average rainfall, median, or maximum value of the corresponding level can be used for calculation. For example, the National Meteorological Administration defines light rain as 0–4.9 mm in 12 hours, and the maximum value is used to convert the rainfall per unit time to 0.4 mm / h. Based on the ground characteristics around the monitoring point, such as clay, silty clay, silt, loess, and silty sand, the corresponding permeability coefficient is obtained based on experience and common sense such as the "Geotechnical Engineering Test and Monitoring Manual". The upper limit of the permeability coefficient is used as the prediction calculation parameter for the impact of rainfall on the rate of increase of groundwater level.

[0031] The measured statistical relationship between rainfall per unit time and groundwater level growth rate includes: obtaining the measured value of the impact of a single rainfall event on the groundwater level growth rate based on historical monitoring experience or historical measured data, such as subtracting the normal growth rate of the groundwater level from the actual growth rate during rainfall; statistically averaging the measured values ​​of the impact of multiple rainfall events of the same amount on the groundwater level growth rate, and using this average value as a prediction of the impact of this type of rainfall on the groundwater level growth rate. Different types of rainfall should be statistically analyzed separately. Since there is a certain delay between the start of rainfall and the groundwater level response, a certain delay effect can be considered during actual measurement, such as a delay of 0.5 hours to 2 hours.

[0032] The impact of other precipitation events such as snowfall and hail on groundwater is predicted in the same way as rainfall, only the delay and timeliness of the impact differ. If there is no precipitation forecast information or no need to predict the impact of precipitation on groundwater level growth, this step can be omitted, and you can directly proceed to the next step of short-term groundwater level anomaly prediction.

[0033] Furthermore, the predicted total water level growth rate control includes: if the predicted total water level growth rate exceeds the total drainage capacity of the drainage pump, i.e. the maximum drainage volume per unit time, drainage will be started immediately and an alarm will be triggered.

[0034] Furthermore, the economic drainage control includes: calculating the drainage start time of the drainage pump at the economic speed based on the predicted total water level growth rate V1, economic drainage rate V2, water level control threshold H0, the predicted start time t1 of the sustained high water level growth rate, and the duration of the sustained high rate T. The economic drainage rate includes the drainage rate with the highest energy efficiency corresponding to the economic speed of the drainage pump, which is essentially the lowest energy consumption and generally does not calculate the backup drainage capacity. The water level control threshold includes a first water level threshold H1 or a third water level threshold H3. The third threshold is the highest water level threshold, while the first water level threshold is the normal safe water level threshold, and the gap between the third water level threshold and the first water level threshold is a safety margin.

[0035] Let the current time t and the current water level h be the starting time for drainage. That is, the economic drainage speed control drainage must be started before the current time t.

[0036] When the forecasted water level growth rate has started to be consistently high, that is, the start time t1 is before or at the current time t, the total water level growth rate V1 with the consistently high water level growth rate already includes the normal growth rate V0.

[0037] △t=(H0-h-(V1-V2)(T-t+t1)) / V2 (Formula 1)

[0038] When the water level growth rate is consistently high but has not yet started, that is, when the starting time t1 is after the current time t, the water level growth rate before time t1 is the normal growth rate V0.

[0039] △t=(H0-h-V1T-V0(t1-t)+V2(T+t1-t)) / V2 (Formula 2)

[0040] When the third water level threshold H3 is used as the water level control threshold, H0 in Formula 1 to Formula 2 needs to be replaced with H3; or when the first water level threshold H1 is used as the water level control threshold, H0 in Formula 1 to Formula 2 needs to be replaced with H1.

[0041] Furthermore, the demand drainage control includes: when economic drainage control cannot effectively control the water level within the water level control threshold in a timely manner, that is, when Δt calculated according to the corresponding case of Formula 1 or Formula 2 is less than 0, economic drainage should theoretically start before the current moment, so the drainage speed needs to be increased to V3.

[0042] When the predicted rate of increase in water level has already begun to be consistently high, that is, when the starting time t1 is before or at the current time t.

[0043] V3=(h+V1(T+t1-t)-H0) / (T+t1-t) (Formula 3)

[0044] When the water level growth rate is consistently high but has not yet started, that is, when the starting time t1 is after the current time t, the water level growth rate before time t1 is the normal growth rate V0.

[0045] V3=(h+V1T+V0(t1-t)-H0) / (T+t1-t) (Formula 4)

[0046] When the third water level threshold H3 is used as the water level control threshold, H0 in Formulas 3 to 4 needs to be replaced with H3; or when the first water level threshold H1 is used as the water level control threshold, H0 in Formulas 3 to 4 needs to be replaced with H1.

[0047] Furthermore, the maximum drainage control includes: if the predicted total increase in water level or the drainage rate V3 calculated by the demand drainage control exceeds the total drainage capacity of the drainage pump, the maximum drainage capacity should be activated immediately and an alarm should be triggered.

[0048] Furthermore, the drainage control also includes time-based drainage control, which involves calculating the drainage volume per unit time based on the measured water level h, the total water level increase rate V1, the second water level threshold H2, and the expected single-start drainage duration T1. Based on the required drainage volume per unit time, the corresponding number of drainage pumps are started and the corresponding pump speeds are set. In other words, given a clear single-start drainage time, the required drainage capacity is determined.

[0049] An automatic groundwater monitoring system includes a water level monitoring sensor, a drainage device, and a drainage controller, which implements any of the above-described automatic groundwater level monitoring methods. The water level monitoring sensor transmits water level information to the drainage controller, which executes the automatic groundwater monitoring method to monitor the groundwater level and sends drainage control information to the drainage device, which then performs drainage operations.

[0050] An automatic groundwater monitoring program that implements any of the above-mentioned automatic groundwater level monitoring methods.

[0051] The advantages and beneficial effects of the present invention are as follows: The groundwater automatic monitoring system and method designed in this invention fully consider the requirements of intelligent construction of smart construction sites, give full play to the advantages of intelligent systems, process and analyze monitoring data in real time, automatically control the start and stop of drainage devices according to the analysis results, and can formulate drainage plans in advance according to the groundwater level prediction, thereby effectively avoiding abnormal seepage accidents in deep foundation pit operations of construction sites, especially water conservancy construction sites. Attached Figure Description

[0052] Figure 1This is a flowchart of an automatic groundwater level monitoring method. Detailed Implementation

[0053] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings and examples. The following examples are only used to more clearly illustrate the technical solutions of the present invention and should not be construed as limiting the scope of protection of the present invention.

[0054] This invention discloses an automatic groundwater level monitoring method, comprising real-time monitoring and prediction of groundwater level changes in a construction area based on real-time groundwater level monitoring values ​​and / or weather forecast information, and controlling drainage operations based on measured and / or predicted results. This embodiment sets upper and lower water level thresholds; drainage operations are initiated when the upper threshold is reached or exceeded, and stopped when the lower threshold is reached or below. The groundwater level change prediction includes predicting groundwater level changes based on precipitation information and / or predicting groundwater level changes based on continuous measured groundwater level information, determining the direction of the groundwater source, and controlling drainage according to the actual drainage capacity of the on-site drainage device and the user's drainage plan requirements to ensure no leakage in the work area, such as deep foundation pits. The precipitation information can be obtained through weather forecasts. Predicting groundwater level changes based on continuous measured groundwater level information includes filtering and predicting groundwater level changes using model filtering technology, which can employ mature technologies. Precipitation includes rain, snow, hail, etc. In this embodiment, the construction area generally does not experience snow or hail, so the design prediction is primarily based on rainfall.

[0055] Preferred, such as Figure 1 As shown, the automatic groundwater level monitoring method further includes the following steps:

[0056] S1: Groundwater level monitoring point layout and real-time monitoring: At least one monitoring point shall be set up in the designated monitoring area to collect data and monitor in real time; the monitoring point shall be equipped with a groundwater level monitoring instrument, such as a pore water pressure gauge, mainly used for monitoring the rise and fall of groundwater level; in this embodiment, one water level monitoring point shall be set up in each of the four directions of the foundation pit (east, south, west, and north) to more accurately determine the source of groundwater and to more accurately measure the rise and fall of groundwater level.

[0057] S2: Predicting the normal growth rate of groundwater level using a water level prediction method: The water level prediction method includes using a prediction model and a filtering prediction algorithm to predict the recent water level growth rate and / or when the first water level threshold will be reached under the said water level growth rate by analyzing continuous groundwater level monitoring data. Normal growth rate prediction and statistics of groundwater level are performed during periods without drainage operations. Generally, when water level changes are not significant, at least one day's worth of water level monitoring data is required, with a sampling period between 1 minute and 30 minutes. This embodiment sets five typical normal monitoring sampling periods: 1 minute, 10 minutes, 20 minutes, 30 minutes, and 60 minutes. The system defaults to a normal sampling interval of 10 minutes. When water level changes are very small, continuous monitoring data for more than one week is generally available. In this embodiment, when using a default sampling interval of 10 minutes, only 1-2 days of measured data are typically added for filtering analysis. When continuously collecting sampling data for one week, the sampling period is generally set to 3 minutes. Two sampling periods are available: 0 minutes or 60 minutes. When water levels change rapidly, data from several minutes to several hours can also be used for prediction calculations. The corresponding sampling period is generally 1 second to 1 minute. In general, continuous sampling data should be more than 50 frames, and the reliability of the stage analysis prediction results is relatively high. When purchasing water level monitoring sensors, it is necessary to pay attention to the ability to support manual setting of sampling periods over a wide range. Most commercially available sensors can support the above requirements. In the water level prediction method, the setting of the original water level monitoring data sampling period should generally ensure continuous sampling within the effective prediction period, and the number of samples should not be less than 3. In this embodiment, cyclic filtering begins after 3 consecutive sample data samples. When the number of samples is more than 50, the filtered prediction results can be accepted. The prediction of the recent water level growth rate includes the prediction of the water level growth rate at the next sampling time after the current sampling time or at any sampling time. Generally, the prediction of the next sampling time is the main focus. In this step, the determination of the direction of the groundwater source can also be initiated according to user needs.

[0058] The prediction model includes at least one of the following: constant velocity model, constant acceleration model, Singer model, "current" model, and polynomial model. In this embodiment, the above models are optional and can be selected by the user during software implementation. The filtering prediction algorithm includes the Kalman filter prediction algorithm. Using the prediction model and the Kalman filter prediction algorithm to filter and predict continuously sampled data is a mature technology.

[0059] While a constant velocity model can meet the needs of predicting groundwater level growth, in reality, the rate of increase slows down when the groundwater level reaches a certain height, and groundwater level changes significantly when surface precipitation is high or seepage occurs. Considering that the "current" model combines the characteristics of both constant velocity and constant acceleration models, and has strong real-time performance, adaptability, and robustness, this embodiment selects the "current" model as the system's default prediction model; related models and Kalman filtering prediction techniques are mature technologies.

[0060] S3: Prediction of the impact of precipitation on groundwater level growth rate. In this embodiment, precipitation is also referred to as rainfall. Local future rainfall or precipitation information is obtained through the Internet or the Internet of Things. The rainfall information includes at least one combination of rainfall amount and duration, or rainfall amount and duration per unit time, or total rainfall within a certain period. In this embodiment, forecast information such as light rain, light to moderate rain, moderate rain, moderate to heavy rain, heavy rain, heavy to torrential rain, torrential rain, torrential rain to extremely heavy rain, and extremely heavy rain is obtained through weather forecast information. The corresponding rainfall amount value is obtained according to the definition of rainfall amount by the National Meteorological Administration. Combined with the empirical statistical value of the ground leakage coefficient after rainfall given in the "Geotechnical Engineering Test and Monitoring Manual", or the measured statistical relationship between rainfall (water) amount per unit time and groundwater level growth rate, the impact of rainfall on groundwater level growth rate is predicted. In this embodiment, the default sampling interval is automatically set to 1 minute when there is rainfall, and the system restores the default sampling interval to 10 minutes after the rainfall ends. If there is no precipitation forecast information, this step can be omitted and proceed directly to the next step S4.

[0061] S4: Short-term groundwater level anomaly prediction. This embodiment provides a manually set sampling interval of 1s-60s for short-term groundwater level anomaly prediction. Upon detecting an abnormally rapid increase in water level, the system automatically sets the default sampling interval to 10s. If there is no drainage operation and the water level increase is normal, the system automatically sets the default sampling interval to 10 minutes. The short-term groundwater level anomaly includes groundwater level anomalies without drainage and groundwater level anomalies with drainage. The water level anomaly generally refers to an abnormal increase in water level, but this invention also includes an abnormal decrease in water level, both of which can be monitored, predicted, and alarmed in real time. The short-term groundwater level anomaly prediction includes water level prediction using water level prediction methods, differential... The difference judgment includes determining whether the difference exceeds a preset first water level growth rate threshold. If it does, the sampling period is reduced, the sampling frequency is increased, and resampling and confirmation prediction are performed. In this embodiment, the first water level growth rate threshold is set to 0.5 mm / h, meaning the difference between the measured water level growth rate and the normal water level growth rate should be less than or equal to this threshold. Otherwise, an anomaly is considered, and the system automatically initiates short-term groundwater level anomaly prediction. The confirmation prediction includes re-predicting using new encrypted sampling data and outputting or alarming the prediction result. In this embodiment, the difference calculation and judgment are performed on the cyclic filtering prediction result after the sampling data generally reaches 50 samples.

[0062] Regarding the calculation and judgment of the difference, when there is no rainfall and no drainage, the difference = |real-time predicted value - normal growth rate|; when there is rainfall and no drainage, the difference = |real-time predicted value - normal growth rate - predicted value affected by rainfall|; when there is no rainfall and drainage, the difference = |real-time predicted value - normal growth rate + drainage volume per unit time|; when there is rainfall and drainage, the difference = |real-time predicted value - normal growth rate - predicted value affected by rainfall + drainage volume per unit time|.

[0063] The normal growth rate includes any one of the following: predicted normal growth rate, measured normal growth rate, statistical mean of predicted normal growth rate, and statistical mean of measured normal growth rate. There is no essential difference between the four, but consistency of the standard should be considered in engineering applications. This embodiment uses the statistical mean of measured normal growth rates from the most recent 10 or more times as the normal growth rate.

[0064] The prediction model and filtering prediction algorithm in the water level prediction method are the same as those in step S2, which predicts the normal rate of increase of groundwater level. If an anomaly is detected, the water source direction judgment program can be started according to user needs to find the source of the anomaly. If there is no water level anomaly or no need for short-term groundwater level anomaly prediction, step S4 can be omitted and the process can proceed directly to the next step, S5, drainage control.

[0065] S5: Drainage control, starting the drainage pump according to real-time control instructions to perform drainage operation, the drainage control method includes any combination of measured value control, predicted total water level growth rate control, economic drainage control, demand-driven drainage control, and maximum drainage control;

[0066] The overall design scheme of drainage control in this embodiment is as follows: When the predicted total growth rate of groundwater level is less than or equal to the economic drainage rate, control is performed according to the measured water level. That is, when the measured water level reaches the first water level threshold, drainage is automatically started; when the measured water level reaches the second water level threshold, drainage is stopped. When the predicted total growth rate of groundwater level is greater than the economic drainage rate, it should be ensured that during the period of sustained high water level growth, such as the duration of rainfall, the increase in groundwater level does not exceed the water level control threshold, which includes the first water level threshold H1 or the third water level threshold H3. The calculation is limited to the predicted abnormal water level duration. If the water level no longer increases abnormally beyond this time, there is no longer any drainage pressure, and drainage can proceed according to normal principles. If the predicted duration is too short, manual intervention is required. If the economic drainage rate cannot achieve this goal, the drainage rate needs to be increased, which is called demand drainage control. Demand drainage control should be executed immediately after confirmation. If the predicted total growth rate of water level or the drainage rate V3 calculated by demand drainage control exceeds the total drainage capacity of the drainage pump, drainage should be started immediately and an alarm should be triggered. Generally, the maximum drainage capacity should be activated. If there is a clear forecast of the duration T of the persistently high groundwater level, the forecast can be used as a reference. If there is no clear forecast, such as when the temporary leakage phenomenon is unclear, a longer period, such as 1 to 3 days, can generally be used for calculation. Considering the treatment cycle, this embodiment defaults to an abnormal time of 2 days. During the drainage process, the actual value of the water level change is monitored. If it continues to increase significantly, on the one hand, the drainage volume should be increased, and on the other hand, the system should issue an alarm to prompt manual handling.

[0067] The measured value control includes: when the measured water level reaches the first water level threshold, drainage is automatically started, and the number of drainage pumps started and the drainage volume per unit time of each pump should ensure that the total drainage volume per unit time is greater than the predicted total water level growth rate; when the measured water level reaches the second water level threshold, drainage is stopped; the total water level growth rate is the sum of the normal growth rate, the groundwater level growth rate affected by rainfall, and the groundwater level growth rate affected by abnormal water level, or it may only consider the abnormal water level growth rate and ignore other minor factors;

[0068] S6: Real-time water level monitoring and prediction during drainage, including temporary adjustment of the water level monitoring sampling frequency and reduction of the sampling time interval before drainage starts. This embodiment provides a manual sampling interval selection of 1s-60s for the drainage process. The default sampling interval is 10s after drainage starts. After no drainage operation and the water level increase is normal, the system will set the default sampling interval to 10 minutes. After drainage starts, at least 10 frames of water level sampling data are continuously collected, water level prediction is started, and the prediction results are displayed, including the water level increase and when the second water level threshold (i.e., the lower limit of the water level, at which point drainage can be stopped) is reached. The water level increase during drainage should be negative, which can also be called water level deceleration. If the predicted water level increase is large... If the value is 0, an alarm is triggered, and the drainage volume should generally be increased until the water level growth rate meets the requirements. If the predicted time to reach the second water level threshold is too long, the drainage volume can also be increased. Increasing the drainage volume includes increasing the pump pressure (or increasing the speed) and / or increasing the number of drainage pumps started. The display of the prediction results also includes displaying the measured growth rate and the predicted growth rate after one hour, as well as the predicted time for the water level to reach the third water level threshold. If the measured water level reaches the second water level threshold and the total water level growth rate is not significantly abnormal, the drainage control ends and returns to step S1 to enter normal real-time monitoring. Otherwise, continue to complete step S6. The significant abnormality includes the total water level growth rate continuously exceeding the normal growth rate and surpassing the first water level growth rate threshold.

[0069] Preferably, the monitoring method further includes determining the direction of the groundwater source, the determination method comprising:

[0070] A method for determining the source of groundwater leakage involves, for interconnected groundwater levels, at least two groundwater level monitoring sensors are arranged in opposite directions within the interconnected area, or one groundwater level monitoring sensor is arranged in four or more opposite directions. Short-cycle high-frequency sampling technology is used to acquire water level monitoring data from each sensor. The sampling period is 1 to 60 seconds; in this embodiment, the water level monitoring sampling period is set to 1 second. The direction of the sensor with the first phase and the fastest water level rise is identified as the main source direction of groundwater leakage. Generally, the water source direction is determined after step S2 or S4 according to user requirements; alternatively, sensors monitoring the flow direction can be used to determine the groundwater source direction.

[0071] Preferably, the method for predicting the impact of rainfall on the rate of increase in groundwater level includes:

[0072] Let V be the predicted rainfall per unit time. y In this embodiment, the unit is defined as mm / h; the duration of rainfall is T. y In this embodiment, the unit is defined as h; the surface permeability coefficient σ is defined as mm / h; the influence coefficient of surface runoff and soil self-storage ε is 0≤ε≤1; generally, when the surface hardening rate is high and the drainage network is well-developed, the influence coefficient ε can be taken as a smaller value, such as 0.1. For soft soil farmland or wasteland, the influence coefficient ε can be taken as a larger value, such as 0.9 or 1.0. In this embodiment, the influence of surface runoff at the construction site is very small, so ε = 0.9 is taken. Therefore, the permeability η is...

[0073] η=εσ / V y

[0074] When the calculation result η>1, then take η=1;

[0075] The predicted impact V4 of rainfall on the rate of increase in groundwater level is:

[0076] V4 = V y η;

[0077] The total water level height increment H4 has the following effect:

[0078] H4 = V4T y

[0079] When only forecast information such as light rain, moderate rain, heavy rain, rainstorm, and torrential rain is available, the average rainfall, median, or maximum value of the corresponding level can be used for calculation. For example, the rainfall for light rain as defined by the China Meteorological Administration is 0–4.9 mm in 12 hours, and the maximum value is taken to convert the rainfall per unit time to 0.4 mm / h. Based on the ground characteristics around the monitoring point, such as clay, silty clay, silt, loess, silt, etc., the corresponding permeability coefficient is obtained according to experience and common sense such as the "Geotechnical Engineering Test and Monitoring Manual". The upper limit of the permeability coefficient is taken as the prediction calculation parameter for the impact of rainfall on the rate of increase of groundwater level.

[0080] Preferably, the predicted total water level growth rate control includes: when the predicted total water level growth rate exceeds the total drainage capacity of the drainage pump, i.e. the maximum drainage volume per unit time, drainage is immediately started and an alarm is triggered.

[0081] Preferably, the economic drainage control includes: calculating the drainage start time of the drainage pump at the economic speed based on the predicted total water level growth rate V1, the economic drainage rate V2, the water level control threshold H0, the predicted start time t1 of the sustained high water level growth rate, and the duration of the sustained high rate T. The economic drainage rate includes the drainage rate with the highest energy efficiency (essentially the lowest energy consumption) corresponding to the economic speed of the drainage pump, and generally does not calculate the standby drainage capacity; the water level control threshold includes a first water level threshold H1 or a third water level threshold H3; the third threshold is the highest water level threshold, while the first water level threshold is the normal safe water level threshold, and the gap between the third water level threshold and the first water level threshold is a safety margin.

[0082] Let the current time t and the current water level h be the starting time for drainage. That is, the economic drainage speed control drainage must be started before the current time t.

[0083] When the forecasted water level growth rate has started to be consistently high, that is, the start time t1 is before or at the current time t, the total water level growth rate V1 with the consistently high water level growth rate already includes the normal growth rate V0.

[0084] △t=(H0-h-(V1-V2)(T-t+t1)) / V2 (Formula 1)

[0085] When the water level growth rate is consistently high but has not yet started, that is, when the starting time t1 is after the current time t, the water level growth rate before time t1 is the normal growth rate V0.

[0086] △t=(H0-h-V1T-V0(t1-t)+V2(T+t1-t)) / V2 (Formula 2)

[0087] Preferably, the demand drainage control includes: when economic drainage control cannot effectively control the water level within the water level control threshold in a timely manner, that is, when Δt calculated according to the corresponding case of Formula 1 or Formula 2 is less than 0, economic drainage should start before the current moment in principle, so the drainage speed needs to be increased to V3.

[0088] When the predicted rate of increase in water level has already begun to be consistently high, that is, when the starting time t1 is before or at the current time t.

[0089] V3=(h+V1(T+t1-t)-H0) / (T+t1-t) (Formula 3)

[0090] When the water level growth rate is consistently high but has not yet started, that is, when the starting time t1 is after the current time t, the water level growth rate before time t1 is the normal growth rate V0.

[0091] V3=(h+V1T+V0(t1-t)-H0) / (T+t1-t) (Formula 4)

[0092] In this embodiment, when the third water level threshold H3 is used as the water level control threshold, H0 in Formulas 1 to 4 needs to be replaced with H3.

[0093] Preferably, the maximum drainage control includes: if the predicted total increase in water level or the drainage rate V3 calculated by the demand drainage control exceeds the total drainage capacity of the drainage pump, the maximum drainage capacity should be activated immediately and an alarm should be triggered.

[0094] Preferably, the drainage control further includes time-based drainage control, which involves calculating the drainage volume per unit time based on the measured water level h, the total water level increase rate V1, the second water level threshold H2, and the expected single-start drainage duration T1, and starting the corresponding number of drainage pumps and setting the corresponding speed of the drainage pumps according to the drainage volume per unit time requirement. That is, under the premise that the single-start drainage time is clear, the drainage capacity that needs to be activated is determined.

[0095] In this embodiment, the unit time drainage volume V3 of time-based drainage control is calculated using the second water level threshold H2 as the lower limit of the water level.

[0096] V3=(h+V1T-H2) / T1 (Formula 5)

[0097] If the calculated drainage volume V3 exceeds the total drainage capacity of the drainage system, the planned drainage time should generally be modified and a new plan should be submitted.

[0098] Another embodiment 2 differs from embodiment 1 in that monitoring points are arranged in eight directions: east, south, west, north, southeast, northwest, northeast, and southwest of the foundation pit, for a total of eight water level monitoring points.

[0099] Another embodiment 3 differs from embodiment 1 in that the first water level threshold H1 is used as the water level control threshold for economic drainage control and demand drainage control. In this case, only H3 in formulas 1 to 4 needs to be replaced with H1.

[0100] Another embodiment, 4, differs from embodiment 1 in that the drainage capacity is designed to be greater than the combined maximum known abnormal seepage and rainfall capacity, and the drainage control only retains the measured value control function.

[0101] Another embodiment 5, an automatic groundwater monitoring system, includes a water level monitoring sensor, a drainage device, and a drainage controller, to implement an automatic groundwater level monitoring method as described in any of the above embodiments. The water level monitoring sensor transmits water level information to the drainage controller, the drainage controller executes the automatic groundwater monitoring method, monitors the groundwater level, and sends drainage control information to the drainage device, which then performs drainage operations.

[0102] Another embodiment, 6, is an automatic groundwater monitoring program that implements the automatic groundwater level monitoring method described in any of the above embodiments.

[0103] The basic principle of this invention is as follows: based on precipitation information from weather forecasts and ground permeability, the impact of precipitation on groundwater levels can be predicted; based on measured groundwater level change trends, models can be used to predict groundwater level change trends over a period of time; based on the measured and predicted groundwater level changes, timely drainage plans can be formulated and drainage operations can be executed; based on the phase changes of water level from sensors installed at different locations, the main direction of groundwater source in the monitored area can be determined. The implementation of this invention can avoid the untimely drainage and construction seepage problems easily caused by traditional single-manual drainage control, meeting the safety requirements of construction sites, especially deep foundation pit construction, and improving the intelligent control level of smart construction site systems.

[0104] The above description is only a partial embodiment of an automatic groundwater monitoring system and method of the present invention. In fact, the control method, the prediction method, or multiple control methods can be simplified or recombined according to the actual needs of the construction site. However, when the design concept of linking groundwater level change prediction with drainage control is consistent with the present invention, these simplifications or combinations should also be considered within the scope of protection of the present invention, and will not be listed one by one here.

Claims

1. A method for automatic monitoring of groundwater level, characterized in that, The method comprises the following steps: S1: arranging and monitoring the underground water level in real time, at least one monitoring point is arranged in the monitoring area, and data is collected for real-time monitoring; S2: predicting the normal increasing rate of the underground water level by a water level prediction method, the water level prediction method comprises using a prediction model and a filtering prediction algorithm, and predicting the recent increasing rate of the water level and / or when the first water level threshold is reached under the increasing rate of the water level by analyzing the continuous monitoring data of the underground water level; the sampling period of the original water level monitoring data in the water level prediction method needs to be set to ensure that the sampling is continuous in the effective prediction period, and the number of samples is not less than 3; The prediction model comprises at least one of a constant speed model, a constant acceleration model, a Singer model, a "current" model and a polynomial model, and the filtering prediction algorithm comprises a Kalman filtering prediction algorithm; S3: predicting the influence of rainfall on the increasing rate of the underground water level, obtaining the future rainfall information of the local area through the Internet or the Internet of Things; the rainfall information comprises at least one of the rainfall amount and the rainfall duration, the rainfall amount per unit time and the rainfall duration, or the total rainfall amount in a certain period of time and the like, and the influence of the rainfall on the increasing rate of the underground water level is predicted by combining the statistical value of the ground infiltration coefficient after the rainfall; S4: short-term underground water level anomaly prediction, the short-term underground water level anomaly comprises underground water level anomaly without drainage and underground water level anomaly with drainage; the short-term underground water level anomaly prediction comprises water level prediction, difference judgment and confirmation prediction by the water level prediction method, the difference judgment comprises judging whether the difference exceeds the preset first water level increasing rate threshold, if yes, the sampling period is reduced, the sampling frequency is increased, the sampling is re-performed and the confirmation prediction is performed; the confirmation prediction comprises re-predicting by using the new encrypted sampling data, and outputting or alarming the prediction result; Without rainfall and without drainage, the difference = |real-time prediction value-normal increasing rate|; with rainfall and without drainage, the difference = |real-time prediction value-normal increasing rate-rainfall influence prediction value|; Without rainfall and with drainage, the difference = |real-time prediction value-normal increasing rate+unit time drainage amount|; with rainfall and with drainage, the difference = |real-time prediction value-normal increasing rate-rainfall influence prediction value+unit time drainage amount|; the normal increasing rate comprises any one of the normal increasing rate prediction value, the measured value, the statistical mean of the normal increasing rate prediction value, and the statistical mean of the normal increasing rate measured value; S5: drainage control, starting the drainage pump according to the real-time control instruction to perform the drainage operation, and the drainage control method comprises any combination of the measured value control, the predicted water level total increasing rate control, the economic drainage control, the demand drainage control and the maximum drainage control. ​ The measured value control includes: when the measured water level reaches a first water level threshold, automatically starting drainage, and the number of started drainage pumps and the drainage capacity of each pump per unit time should ensure that the total drainage capacity per unit time is greater than the predicted total water level increase rate; when the measured water level reaches a second water level threshold, stopping drainage; the total water level increase rate is the sum of the normal increase rate, the rainfall-influenced underground water level increase rate, and the water level anomaly-influenced underground water level increase rate; S6: Real-time water level monitoring and prediction during the drainage process, including temporarily adjusting the water level monitoring sampling frequency before starting the drainage, reducing the sampling time interval, continuously collecting no less than 10 frames of water level sampling data after starting the drainage, and starting water level prediction; checking whether the measured water level reaches the second water level threshold and the total water level increase rate has no obvious abnormality, then ending the drainage control and returning to step S1 to enter normal real-time monitoring; otherwise, continuing to complete the present step S6; the obvious abnormality includes that the total water level increase rate continuously exceeds the normal increase rate and exceeds the first water level increase rate threshold.

2. The method of claim 1, wherein, The monitoring method further includes underground water source direction judgment, and the judgment method includes: For regions with interconnected underground water levels, at least two underground water level monitoring sensors are arranged in opposite directions in the connected region, or four or more underground water level monitoring sensors are arranged in relative directions; short-period high-frequency sampling technology is used to obtain water level monitoring data of each sensor, and the sampling period is 1s-60s; the direction of the sensor with the highest water level or the fastest water level increase rate is the main source direction of the underground water leakage.

3. The method of claim 1, wherein, The rainfall influence prediction method on the water level increase rate includes: The unit time rainfall of the record is V y , the rainfall duration is T y , the ground infiltration coefficient σ, the ground runoff and soil self-storage influence coefficient ε, 0≤ε≤1, then the infiltration rate η is η = εσ / V y When the calculation result η>1, η=1 is taken; The rainfall influence prediction value V4 on the water level increase rate is V4 = V y η; The total water level height increase H4 is H4= V4T y .

4. The method of claim 1, wherein, The predicted total water level increase rate control includes: when the predicted total water level increase rate exceeds the total drainage capacity of the drainage pumps, immediately starting drainage and alarming.

5. The method of claim 1, wherein, The economic drainage control includes: according to the predicted total water level increase rate V1, the economic drainage speed V2, the water level control threshold H0, the start time t1 of the continuously high predicted water level increase rate, and the continuously high duration T, calculating the drainage starting time under the economic rotation speed of the drainage pumps; the economic drainage speed includes the drainage speed with the highest energy consumption effective utilization rate corresponding to the economic rotation speed of the drainage pumps; the water level control threshold includes the first water level threshold H1 or the third water level threshold H3; Let t be the current time, h be the current water level height, and △t be the time when the drainage is started after the current time t; When the continuously high predicted water level increase rate has started, the start time t1 is before the current time t or is the current time t; △t=(H0-h-(V1-V2)(T-t+t1)) / V2 (Formula 1) When the continuously high predicted water level increase rate has not started, the start time t1 is after the current time t, and the water level increase rate before the time t1 is the normal increase rate V0; △t=(H0-h-V1T-V0(t1-t)+V2(T+t1-t)) / V2 (Formula 2).

6. The method of claim 1, wherein, The demand drainage control includes: when the economic drainage control cannot achieve timely and effective control of the water level within the water level control threshold, increasing the drainage speed to V3. When the forecast water level increase rate continues to be higher than normal, and the start time t1 is before or equal to the current time t, V3=(h+V1(T+t1-t)-H0) / (T+t1-t) (Formula 3) When the forecast water level increase rate continues to be higher than normal, and the start time t1 is before or equal to the current time t, V3=(h+V1(T+t1-t)-H0) / (T+t1-t) (Formula 3) 7. The method of claim 1, wherein, The maximum drainage control includes: if the forecast water level increase rate or the drainage speed V3 calculated by the demand drainage control exceeds the total drainage capacity of the drainage pumps, the maximum drainage capacity should be started immediately and an alarm should be given.

8. The method of claim 1, wherein, The drainage control also includes time-based drainage control, which includes calculating the unit time drainage amount according to the measured water level h, the water level total increase rate V1, the second water level threshold H2, and the predicted single start drainage duration T1, and starting the corresponding number of drainage pumps and setting the corresponding rotation speed of the drainage pumps according to the unit time drainage amount demand; The second water level threshold H2 is used as the lower limit of the water level to calculate the unit time drainage amount V3 of the time-based drainage control V3=(h+V1T-H2) / T1 (Formula 5).

9. An automatic underground water monitoring system, comprising a water level monitoring sensor, a drainage device, and a drainage controller, which implements the automatic underground water monitoring method of any one of claims 1-8, the water level monitoring sensor transmits water level information to the drainage controller, the drainage controller executes the automatic underground water monitoring method, monitors the underground water level, and sends drainage control information to the drainage device, and the drainage device executes the drainage operation.

Images