Intelligent control method and system for foundation pit dewatering

Abstract

The invention relates to the technical field of foundation pit dewatering, in particular to an intelligent control method and system for foundation pit dewatering, and the method comprises the steps: carrying out the historical data collection of an area where foundation pit excavation and dewatering are planned, and finding an underground water level rule and an underground water flowing direction rule of the area under various air temperature and dewatering conditions through the historical data. After the rule is obtained, the flow direction and the water level of the underground water can be predicted by using the meteorological data at the future time point, and then the position of the precipitation well is arranged by using the flow mode of the underground water, so that the position of the precipitation well is more reasonable. Besides, the water level in the dewatering well is controlled through the water pump, during control, the underground water level rule is used for predicting the underground water change amount in the future time, the change amount is used for adjusting a water pump control signal in advance, and therefore the method has the good external interference resisting capacity.

Description

Technical Field

[0001] The present invention relates to the technical field of foundation pit dewatering, and specifically to an intelligent control method and system for foundation pit dewatering. Background Art

[0002] Foundation pit dewatering refers to a series of measures to lower the groundwater level when excavating a foundation pit to ensure that the construction can be carried out in a dry environment and avoid a series of problems caused by too high groundwater level, such as slope instability, foundation quicksand, bottom heave, bottom piping, and reduction of foundation bearing capacity.

[0003] The existing foundation pit dewatering means is usually well point dewatering, that is, dewatering wells are dug around the foundation pit, and the dewatering wells are used to lower the groundwater level in the area where the foundation pit is located and pump out the accumulated water in the foundation pit to keep the foundation pit dry. The existing control of foundation pit dewatering mainly relies on the monitoring of the water level in the dewatering well. When the water level rises to a certain level, the water pump is controlled to pump water. However, the activities of groundwater will change, and the method of controlling the water pump by water level monitoring has the disadvantages of lag effect and weak ability to resist external interference (such as rainfall). Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide an intelligent control method and system for foundation pit dewatering to solve the above technical problems.

[0005] In order to achieve the above purpose, the present invention adopts the following technical solutions:

[0006] An intelligent control method for foundation pit dewatering of the present invention includes the steps of:

[0007] Obtaining the groundwater level sampling values, the precipitation amounts at multiple historical time points, and the air temperatures at multiple historical time points of multiple sampling points around the foundation pit; and obtaining the precipitation prediction data and the air temperature prediction data during the precipitation construction period of the foundation pit;

[0008] Constructing a water level change characteristic model and a groundwater flow direction characteristic model for the area where the foundation pit is located based on the groundwater level sampling values, the precipitation amounts, and the air temperatures at multiple historical time points of the multiple sampling points, wherein the water level change characteristic model characterizes the groundwater level change characteristics under various air temperature and precipitation change conditions, and the groundwater flow direction characteristic model characterizes the groundwater flow characteristics under various air temperature and precipitation conditions;

[0009] Determining the predicted groundwater flow characteristics during the precipitation construction period of the foundation pit based on the precipitation prediction data, the air temperature prediction data, and the groundwater flow direction characteristic model, and determining the positions of multiple dewatering wells based on the predicted groundwater flow characteristics, wherein the multiple dewatering wells are arranged around the foundation pit;

[0010] When performing foundation pit dewatering, a water pump control signal is generated, and the water level change amount is determined based on the water level change characteristic model, precipitation prediction data, and temperature prediction data. A water pump adjustment signal is generated based on the water level change amount, and the water pump is adjusted based on the water pump control signal and the water pump adjustment signal.

[0011] The present application also provides an intelligent control system for foundation pit dewatering, including:

[0012] An acquisition module for acquiring the underground water level sampling values, precipitation amounts at multiple historical time points, and temperatures at multiple historical time points of multiple sampling points around the foundation pit; and acquiring precipitation prediction data and temperature prediction data for the dewatering construction period of the foundation pit;

[0013] A model construction module for constructing a water level change characteristic model and an underground water flow direction characteristic model of the area where the foundation pit is located based on the underground water level sampling values, precipitation amounts at multiple historical time points, and temperatures at multiple historical time points of the multiple sampling points. Among them, the water level change characteristic model characterizes the underground water level change characteristics under various temperature and precipitation change conditions, and the underground water flow direction characteristic model characterizes the underground water flow characteristics under various temperature and precipitation conditions;

[0014] A prediction module for determining the predicted underground water flow characteristics during the dewatering construction period of the foundation pit based on the precipitation prediction data, the temperature prediction data, and the underground flow direction characteristic model, and determining the positions of multiple dewatering wells based on the predicted underground water flow characteristics, where the multiple dewatering wells are arranged around the foundation pit;

[0015] A control module for generating a water pump control signal when performing foundation pit dewatering, determining the water level change amount based on the water level change characteristic model, precipitation prediction data, and temperature prediction data, generating a water pump adjustment signal based on the water level change amount, and adjusting the water pump based on the water pump control signal and the water pump adjustment signal.

[0016] The beneficial effects of the present invention are as follows: An intelligent control method and system for foundation pit dewatering according to the present invention collect historical data of the area where foundation pit excavation and dewatering are planned to be implemented, and use the historical data to find the laws of the groundwater level and the direction of groundwater flow under various temperature and precipitation conditions in this area. After obtaining the laws, the meteorological data at future time points can be used to predict the direction of groundwater flow and the water level, and then the positions of the dewatering wells can be arranged according to the groundwater flow pattern, making the positions of the dewatering wells more reasonable. In addition, this application uses a water pump to control the water level in the dewatering well. During the control, the law of the groundwater level is used to predict the change amount of groundwater in the future, and the change amount is used to adjust the water pump control signal in advance, so that this application has good anti-external interference ability. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The present invention will be further described below in conjunction with the drawings and embodiments:

[0018] Figure 1 is a schematic diagram of the construction site result of foundation pit dewatering shown in an embodiment of the present application;

[0019] Figure 2 is a flowchart of the intelligent control method for foundation pit dewatering shown in an embodiment of the present application;

[0020] Figure 3 is a schematic diagram of the distribution of the positions of the dewatering wells in an embodiment of the present application;

[0021] Figure 4 is a schematic diagram of the production process of the water pump control signal in an embodiment of the present application;

[0022] Figure 5 is a structural diagram of an intelligent control system for foundation pit dewatering shown in an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] The principle of foundation pit dewatering is to lower the groundwater level below the foundation pit excavation surface through artificial intervention means to keep the bottom and slope of the foundation pit dry, and prevent problems such as slope instability, quicksand, piping, and bottom heave caused by groundwater infiltration, so as to ensure construction safety and engineering stability. Figure 1 is a schematic diagram of the construction site result of foundation pit dewatering shown in an embodiment of the present application. As Figure 1 shown, a plurality of dewatering wells 120 are arranged around the foundation pit 110 in this application, and the bottom of the dewatering well 110 is at least 0.5 meters lower than the bottom of the foundation pit 120. The bottom of the dewatering well 120 extends into the aquifer, and water can be pumped through the water pump inside the dewatering well 120 to form a dry area around the foundation pit 110 to keep the foundation pit dry. In addition, a plurality of open ditches or hidden ditches can be built between the foundation pit 120 and the dewatering well for drainage.

[0024] Figure 2 It is a schematic flow chart of an intelligent control method for foundation pit dewatering shown in an embodiment of the present application. As Figure 2 shown, the present application includes the steps:

[0025] S210, obtaining the groundwater level sampling values at multiple sampling points around the foundation pit at multiple historical time points, the precipitation amounts at multiple historical time points, and the air temperatures at multiple historical time points; and obtaining the precipitation prediction data and the air temperature prediction data during the dewatering construction period of the foundation pit;

[0026] In the present application, a groundwater level monitor is used to detect the groundwater level sampling values of multiple sampling points.

[0027] The sampling points are circularly distributed with the position of the foundation pit as the base point, and the distance between each sampling point and the foundation pit remains the same to form an annular sampling zone. The precipitation amounts at multiple historical time points and the air temperatures at multiple historical time points are obtained by extracting historical meteorological data.

[0028] The precipitation prediction data and the air temperature prediction data during the dewatering construction period of the foundation pit are obtained from the meteorological prediction data.

[0029] S220, constructing a water level change characteristic model and a groundwater flow direction characteristic model for the area where the foundation pit is located based on the groundwater level sampling values at multiple sampling points at multiple historical time points, the precipitation amounts at multiple historical time points, and the air temperatures at multiple historical time points. Among them, the water level change characteristic model characterizes the groundwater level change characteristics under various air temperature and precipitation change conditions, and the groundwater flow direction characteristic model characterizes the groundwater flow characteristics under various air temperature and precipitation conditions;

[0030] Air temperature and precipitation are the main factors affecting groundwater changes. Precipitation (including rain, snow, etc.) is one of the main sources of groundwater recharge. During the period of abundant precipitation, surface water gradually replenishes the underground aquifer through soil infiltration, resulting in an increase in the groundwater level. On the contrary, in arid seasons or areas with little rain, the groundwater cannot be effectively replenished, and the water level may drop. Air temperature indirectly affects the groundwater level by influencing the evaporation rate. Higher temperatures will accelerate the evaporation of surface water bodies (such as lakes, rivers) and soil surface moisture, reducing the amount of water entering the groundwater system. In addition, in some areas, high temperatures will also increase the transpiration of vegetation, which also consumes some of the water that might otherwise infiltrate into the ground. In the long run, rising air temperatures may lead to regional aridification, further reducing the groundwater recharge volume and prompting the water level to drop.

[0031] Therefore, after collecting historical data, the present application constructs characteristic models reflecting the groundwater level and the groundwater flow direction based on the historical data.

[0032] (1) Water level change characteristic model

[0033] (1-1) Calculate the change characteristic value WP_C of the groundwater level sampling values at any two adjacent time points t , the change characteristic value R_C of the precipitation t and the change characteristic value W_C of the temperature t , where the change characteristic value is a difference or a change rate, and t represents the time point;

[0034] In step (1-1), calculating the change characteristic values of the groundwater level, precipitation, and temperature at adjacent time points can be a difference or a change rate. The difference is the difference between two adjacent points, and the change rate may be the difference divided by the time interval. Here, the time interval may need to be clarified, such as daily or hourly. For example, the groundwater level observation in a certain area is on a daily scale, and perhaps the time points here are also daily data. So the change rate may use the daily change rate, such as (current value - previous day value) / previous day value, or directly the difference.

[0035] The mathematical expression of the change characteristic value ΔA is:

[0036] ΔA = A t -A t-1

[0037] Or

[0038]

[0039] A t represents the groundwater level, precipitation, or temperature at time point t, and A t-1 represents the groundwater level, precipitation, or temperature at time point t-1.

[0040] (1-2) Based on the change characteristic values WP_C t of the same time point, the change characteristic value R_C of the precipitation t and the change characteristic value W_C of the temperature t construct the first sample data Data_sam at multiple historical time points t , Data_sam t = (WP_C t , R_C t , W_C t );

[0041] Each sample contains three characteristics: groundwater level change, precipitation change, and temperature change. In this step, it is necessary to ensure that the data is aligned and the time points are consistent.

[0042] (1-3) Remove the first sample data Data_sam where the change characteristic value WP_C of the groundwater level sampling value t is less than the preset thresholdt , obtain the target data set;

[0043] Filter out samples with the change in groundwater level less than the preset threshold. The threshold here may need to be set according to the actual situation, such as the extreme precipitation threshold, or determined according to the data distribution. For example, if the change is very small, it may be considered that these data have little impact on the analysis and can be excluded to reduce noise.

[0044] For example, if the standard deviation of the daily change in groundwater level in historical data is 5 mm, the threshold can be set to ±2 mm.

[0045] Filtering condition: If |Δ groundwater levelt| < threshold |Δ groundwater levelt| < threshold, then eliminate this sample.

[0046] (1-4) Cluster the sample data vectors in the target data set based on the change characteristic values of precipitation and the change characteristic values of temperature to obtain the first sample clusters under various temperature and precipitation conditions;

[0047] Cluster the target data set based on the change characteristic values of precipitation and temperature, so as to cluster various precipitation and temperature conditions. Here, a clustering algorithm may be needed, such as DBSCAN or hierarchical clustering. Here, a suitable distance metric needs to be selected, such as the Euclidean distance, and then group according to the change characteristic values of precipitation and temperature to obtain sample clusters under different climate conditions.

[0048] In addition, before clustering, the data needs to be normalized to facilitate setting a general variance threshold later.

[0049] (1-5) Calculate the data volume in multiple first sample clusters and the variance of the change characteristic values of the groundwater level sampling values, take the first sample clusters with the data volume greater than the preset quantity threshold as the intermediate clusters, and screen out outliers from the intermediate clusters with the variance greater than the preset variance threshold to obtain the target clusters with the variance less than or equal to the preset variance threshold;

[0050] The variance reflects the aggregation degree of the water level data in the cluster, and the quantity reflects whether the data in the cluster is formed accidentally. In this application, if the data volume in the cluster is too small, it may be formed accidentally and cannot reflect the law of the climate conditions in the cluster. Therefore, in this embodiment, only the first sample clusters with the data volume greater than the preset quantity threshold are retained.

[0051] In addition, under normal circumstances, if there is no water discharge caused by human activities, the groundwater levels should be similar under similar climate conditions. Therefore, it is necessary to eliminate abnormal data from the clusters to ensure the high aggregation of the groundwater level sampling values in the clusters and eliminate interfering data. The process of eliminating abnormal data includes:

[0052] (1-5-1) Calculate the variance of the water level in the cluster;

[0053] (1-5-2) When the variance of the water levels within the cluster is greater than a preset variance threshold, remove the data with the largest deviation rate from the water level average within the cluster, and return to (1-5-1) until the variance of the water levels within the cluster is less than or equal to the preset variance threshold.

[0054] (1-6) Calculate the mean and standard deviation of the change characteristic values of the groundwater level sampling values in the target cluster, and construct a reference range for the change characteristic values based on the mean and standard deviation, where the median of the reference range is the typical value.

[0055] After a series of processing procedures described above, the remaining target clusters can reflect the groundwater level pattern under a certain climate condition. Therefore, a reference range is constructed based on the standard deviation and mean within the cluster, and the expression of the reference range is:

[0056] [μ-kσ,μ+kσ][μ-kσ,μ+kσ]

[0057] where μ is the mean, σ is the standard deviation, and k is the confidence coefficient (for example, 1.96 corresponds to a 95% confidence interval).

[0058] (2) Groundwater flow direction characteristic model

[0059] (2-1) For each historical time point, determine the flow vectors flow of the groundwater at multiple target points based on multiple groundwater level sampling values t (γ,v), where γ represents the flow direction and v represents the flow velocity;

[0060] Among them, the flow vector consists of the flow direction and the flow velocity. In the prior art, the flow direction is generally represented by the hydraulic head. For the convenience of determining the position of the precipitation well in the subsequent process of this application, a flow vector is constructed.

[0061] The construction of the flow vector is based on existing groundwater flow analysis tools, such as MODFLOW. The construction process of the flow vector flow t (γ,v) includes:

[0062] (2-1-1) Calculate the hydraulic heads of multiple sampling points based on multiple groundwater level sampling values;

[0063] The hydraulic head is a core concept in fluid mechanics, referring to the mechanical energy possessed by a unit weight of liquid, usually expressed in height (meters or meters of water column, mH2O). It consists of the following three parts: position head, pressure head, and velocity head. The calculation formula for the hydraulic head is:

[0064]

[0065] In the formula, H总 where \(H\) represents the total head, \(z\) represents the elevation head, \(P\) represents the hydrostatic pressure, \(\rho\) represents the density of water, \(g\) represents the acceleration due to gravity, and \(v\) represents the flow velocity. represents the pressure head, represents the velocity head.

[0066] Under normal circumstances, the flow velocity of groundwater is relatively slow, so the velocity head can be ignored. In addition, it can be converted through the water level, so there is no need to measure the specific groundwater pressure.

[0067] (2-1-2) Input the heads of multiple sampling points into the groundwater simulated flow analysis tool to obtain a flow grid map, where the flow grid map includes the flow vectors of multiple grids;

[0068] In this application, the heads of multiple sampling points and the existing groundwater simulated flow analysis tool are used for analysis to obtain a flow grid map, which contains the groundwater flow direction and flow velocity of each grid.

[0069] (2-1-3) Extract the flow vectors of multiple target points from the flow grid map.

[0070] Finally, extract the groundwater flow direction and flow velocity of all target points from the flow grid map to form the flow vector.

[0071] (2-2) Based on the flow vectors \(flow(\gamma, v)\) at multiple historical time points, precipitation \(R\) t and air temperature \(W\) t construct the second sample data \(Data\_sam'\) at multiple historical time points, \(Data\_sam'=(flow(\gamma, v), R, W)\); t t ′, t t t t )

[0072] (2-3) Cluster the multiple second sample data based on precipitation and air temperature to obtain second sample clusters under various air temperature and precipitation conditions;

[0073] (2-4) Calculate the data volume and variance of the flow vectors within the multiple second sample clusters, use the second sample clusters with a data volume greater than the preset quantity threshold as intermediate clusters, and perform outlier screening on the intermediate clusters with a variance greater than the preset variance threshold to obtain target clusters with a variance less than or equal to the preset variance threshold;

[0074] (2-5) Calculate the mean and standard deviation of the flow vectors in the target clusters, and construct a reference range of the flow vectors based on the mean and standard deviation, where the median of the reference range is the typical value.

[0075] The principle of steps (2-2)-(2-5) is similar to that described above. Please refer to the above for understanding and details will not be repeated here.

[0076] S230. Based on the precipitation prediction data, the temperature prediction data, and the underground flow dynamic characteristic model, determine the predicted groundwater flow characteristics during the precipitation construction period of the foundation pit, and determine the positions of multiple precipitation wells based on the predicted groundwater flow characteristics, where the multiple precipitation wells are arranged around the foundation pit.

[0077] After obtaining the underground flow dynamic characteristic model through the model construction described above, in order to more reasonably arrange the positions of the precipitation wells, this application determines the positions of the precipitation wells by predicting the groundwater flow characteristics during the precipitation construction period, including the steps:

[0078] S231. Obtain a set of flow vectors at multiple future time points during the precipitation construction period, where the set of flow vectors includes the flow vectors of multiple sampling points.

[0079] The flow vector data needs to cover the entire precipitation construction period (such as 7 days) and be output according to time points (such as every hour or daily).

[0080] S232. For any sampling point S i , calculate the mean value of the flow vectors at multiple future time points to obtain the average flow vectors of multiple sampling points.

[0081] The average flow vector can represent the approximate flow velocity and direction of groundwater in the future time period.

[0082] S233. Construct a reference line S i between the multiple sampling points S i and the position O of the foundation pit, and determine the included angle γ i between the reference line S i O and the average flow vector of the sampling point S i , and based on the included angle γ i and the average flow vector of the sampling point S i , calculate the number N i of precipitation wells in the interception area corresponding to the sampling point S i , and the mathematical expression of the number N i of precipitation wells is:

[0083]

[0084] In the formula, round() represents rounding, and v i is the sampling point S iThe flow velocity of the average flow vector, v0 is the reference flow velocity, and n is the empirical value of the number of precipitation wells at the reference flow velocity.

[0085] Finally, use the reference line S i O and the sampling point S i The included angle γ of the average flow vector i represents the effective cut-off flow velocity of the precipitation well during interception, and more precipitation wells need to be set in the area with a larger flow velocity. Therefore, the empirical value n and the reference flow velocity v0 are used to determine the number of precipitation wells in each area.

[0086] Figure 3 is a schematic diagram of the precipitation well position distribution in an embodiment of the present application. The precipitation well positions determined through the above process are as Figure 3 shown.

[0087] S240. When performing foundation pit dewatering, generate a water pump control signal, determine the water level change amount based on the water level change characteristic model, precipitation prediction data, and temperature prediction data, generate a water pump adjustment signal based on the water level change amount, and adjust the water pump based on the water pump control signal and the water pump adjustment signal.

[0088] When determining the change amount, substitute the precipitation prediction data and the temperature prediction data into the water level change characteristic model to obtain the underground water level change characteristics at multiple future time points; construct the water level change amount based on the current underground water level and the underground water level change characteristics.

[0089] In the present application, the work of the water pump in the precipitation well is intelligently controlled by setting a warning line, and then the warning line is adjusted based on the predicted underground water change amount to avoid the situation where the water pump cannot drain the accumulated water in the well in time when encountering precipitation. Or, in order to drain the accumulated water in the well, it needs to operate at a high power, which increases energy consumption and reduces the service life of the water pump.

[0090] In the present application, the generation process of the basic water pump control signal includes:

[0091] S2401, obtain the real-time precipitation well water level H s , foundation pit water level H j , underground water level H u and the soil permeability K of the area where the foundation pit is located, and obtain the performance curve of the water pump. Among them, the water pump performance curve includes the relationship between the power of the water pump and the pumping speed;

[0092] The performance curve of a water pump characterizes the relationship between the power of the water pump and the pumping speed. The optimal working power of the water pump is at the power value corresponding to the peak point (BEP) of the efficiency curve (η-Q curve). At this time, the efficiency is the highest, the water volume processed per unit power is the largest; the economy is the best, and the system energy consumption is the lowest; the operation stability is the best, avoiding vibration at low flow rates or overload risks at high flow rates.

[0093] S2402, based on the precipitation well water level H s , the foundation pit water level H j , the groundwater level H u and the soil permeability K of the area where the foundation pit is located to calculate the infiltration flow Q in the precipitation well. The mathematical expression of the infiltration flow Q is:

[0094]

[0095] S1 = πR1(H j -H d1 )

[0096]

[0097] In the formula, L1 is the vertical distance between the central axis of the precipitation well and the central axis of the foundation pit, L2 is the shortest distance between the central axis of the precipitation well and the central axis of the sampling point, S1 is the contact area between the precipitation well and the water seeping out of the foundation pit, S2 is the contact area between the precipitation well and the groundwater, R1 is the diameter of the foundation pit, R2 is the diameter of the precipitation well, H d1 is the depth of the foundation pit, H d2 is the depth of the bottom of the aquifer, H d3 is the depth of the precipitation well, and L is the thickness of the aquifer;

[0098] The principle of the above formula includes:

[0099] The infiltration flow Q is determined by the seepage formula, and the seepage water in the precipitation well is determined by the groundwater and the seepage water from the foundation pit. During the initial stage of precipitation, the water level in the foundation pit gradually decreases, and the foundation pit water level H j gradually decreases to 0. The groundwater level is initially stable, but as the pumping time lengthens, the groundwater level will also gradually decrease. Therefore, the infiltration flow Q does not remain constant but gradually decreases.

[0100] Please refer to Figure 1 for understanding. S1 is the contact area between the precipitation well and the water seeping out of the foundation pit. When the water in the foundation pit seeps into the precipitation well, the contact area with the seepage water is the side of the precipitation well facing the well pit. Therefore, the general contact area is the side area of the corresponding height cylinder. The groundwater may seep in through the side or through the side and the ground, depending on the position relationship between the precipitation well and the aquifer. If the bottom of the precipitation well is higher than the bottom of the aquifer, that is, Hd2 ≥H d3 , the contact area is the contact side + the bottom surface; if the bottom surface of the precipitation well is lower than the bottom plate of the aquifer, i.e., H d2 <H d3 , the contact area is the contact side area.

[0101] S2403, obtain the optimal working power of the water pump and the target pumping speed V corresponding to the optimal working power from the performance curve of the water pump, and generate a water pump control signal based on the infiltration flow rate Q and the target pumping speed V, so that the water pump maintains the optimal working power as much as possible and keeps the water level of the precipitation well below the warning line.

[0102] Figure 4 is a schematic diagram of the production process of the water pump control signal in an embodiment of the present application. As Figure 5 shown, generating a water pump control signal based on the infiltration flow rate Q and the target pumping speed V includes:

[0103] S1, compare the infiltration flow rate Q and the target pumping speed V, and compare the water level of the precipitation well with the warning line;

[0104] S2, when the target pumping speed V is greater than the infiltration flow rate Q and the water level of the precipitation well is lower than the warning line, generate a stop signal for controlling the water pump to stop, and return to S1 after waiting for a target duration;

[0105] When the target pumping speed V is greater than the infiltration flow rate Q, it indicates that the pumping capacity exceeds the infiltration volume and the water level may drop.

[0106] When the target pumping speed is greater than the infiltration flow rate and the water level is lower than the warning line, generate a stop signal to stop the water pump from working for a period of time and then return to S1. This may mean that the pumping speed exceeds the actual need, resulting in too low a water level, and it is necessary to pause to avoid excessive pumping.

[0107] S3, when the target pumping speed V is greater than the infiltration flow rate Q and the water level of the precipitation well is higher than or equal to the warning line, generate a first control signal for controlling the water pump to work at the optimal working power, and return to S1 after waiting for a target duration;

[0108] When the target pumping speed is greater than the infiltration flow rate but the water level is higher than or equal to the warning line, let the water pump operate at the optimal power. This may be because although the pumping capacity exceeds the infiltration volume, the water level is still relatively high, and it is necessary to maintain efficient operation to quickly drain the water.

[0109] This working condition is generally caused by the water level rising due to draining the water in the foundation pit into the precipitation well through the ditch. In this case, maintaining the optimal working efficiency of the water pump can gradually lower the water level in the precipitation well.

[0110] S4. When the target pumping rate V is less than or equal to the infiltration flow rate Q and the water level in the precipitation well is lower than the warning line, generate a first control signal for controlling the water pump to operate at the optimal working power, wait for the target duration, and then return to S1;

[0111] When the target pumping rate V is less than or equal to the infiltration flow rate Q, it indicates that when the water pump operates at the optimal power, the pumping capacity is insufficient to correspond to the seepage depth. However, the water level has not reached the warning line. Therefore, the water level may gradually rise. But as the water level in the foundation pit drops during the pumping process, the seepage rate will also gradually decrease. Therefore, it is only necessary to continue to maintain the water pump to operate at the optimal working power.

[0112] This working condition is generally caused by the rise of the water level in the foundation pit and precipitation well due to rain, or the rise of the groundwater level.

[0113] S5. When the target pumping rate V is less than or equal to the infiltration flow rate Q and the water level in the precipitation well is higher than or equal to the warning line, generate a second control signal for updating the working power of the water pump, wait for the target duration, and then return to S1, where the updated power P n ′ satisfies: P n ′ = P n + P0, P n is the power before update, and P0 is the increase.

[0114] Finally, if the water level cannot be maintained under the S4 working condition, the power of the water pump is cyclically increased to gradually control the water level line. The updated power P n ′ does not exceed the maximum power of the water pump.

[0115] The above process is the generation process of the basic control signal for maintaining the water level in the precipitation well. However, during the above process, if a heavy rain causes the water level to rise rapidly, the water pump cannot respond in a short time. Therefore, based on the predicted water level change amount, this application updates the total seepage volume and correspondingly adjusts the warning line, so as to correspond to the increase in seepage volume in the future time period, specifically including:

[0116] S2411. Update the groundwater level based on the water level change amounts at multiple future time points, and calculate the average infiltration flow rate Q at multiple future time points based on the updated groundwater level A ;

[0117] In this embodiment, only the water volume change caused by the rising groundwater level is considered. Although there will also be water accumulation in the foundation pit, the water volume in the foundation pit will not be too large. After a period of dewatering, it will not have a great impact on the seepage water volume in the dewatering wells. However, the impact brought by the change in the groundwater level is large and continuous. Therefore, the groundwater level is mainly updated. In another embodiment, the change in the foundation pit water level can also be introduced, and the water level rise can be predicted by using the predicted rainfall and the cross-sectional area of the foundation pit, which will not be elaborated here.

[0118] When it is more detailed, calculations are performed using the change characteristics and the current water level to obtain the updated predicted water level. For example, using the change rate method, the current water level is H, and the change rate in the future time period is 10%. Then the predicted water level is 1.1 times H.

[0119] After updating the groundwater level, the average infiltration flow rate Q at multiple future time points can be obtained. A 。

[0120] S2421, when the average infiltration flow rate Q at multiple future time points A is greater than or equal to the target pumping speed V, adjust the warning line position. The adjusted warning line position is H AL ′. Among them, the mathematical expression of the adjusted warning line position H AL ′ is:

[0121]

[0122] In the formula, T is the duration of multiple future time points, and H AL ″ is the set minimum water level.

[0123] Finally, calculate the increase value of the dewatering well water level that will be caused in the future time period Then use the increase value of the dewatering well water level as the warning line downward adjustment value to cope with it in the future time period.

[0124] In addition, in order to avoid excessive dewatering, a minimum water level H AL ″ is also set. If the warning line drops below the minimum water level H AL ″ after the downward adjustment, then use the minimum water level H AL ″ as the updated warning line.

[0125] Introducing the adjusted warning line into the generation process of the basic control signal can enable the water pump to operate at the best power as much as possible, and at the same time be able to cope with the possible heavy rainfall phenomenon in the future.

[0126] An intelligent control method for foundation pit dewatering according to the present invention collects historical data on the area where foundation pit excavation and dewatering are planned to be implemented, and uses the historical data to find the laws of the groundwater level and the direction of groundwater flow under various temperature and precipitation conditions in this area. After obtaining the laws, the meteorological data at future time points can be used to predict the direction of groundwater flow and the water level, and then the positions of the dewatering wells can be arranged according to the groundwater flow pattern, making the positions of the dewatering wells more reasonable. In addition, this application uses a water pump to control the water level in the dewatering well. During the control, the law of groundwater level is used to predict the change amount of groundwater at future times, and the change amount is used to adjust the water pump control signal in advance, so that this application has good anti-external interference ability.

[0127] As Figure 5 shown, this application also provides an intelligent control system for foundation pit dewatering, including:

[0128] An acquisition module, configured to acquire the groundwater level sampling values at multiple sampling points around the foundation pit at multiple historical time points, the precipitation amounts at multiple historical time points, and the temperatures at multiple historical time points; and acquire the precipitation prediction data and temperature prediction data during the dewatering construction period of the foundation pit;

[0129] A model construction module, configured to construct a water level change characteristic model and a groundwater flow direction characteristic model for the area where the foundation pit is located based on the groundwater level sampling values at multiple sampling points around the foundation pit at multiple historical time points, the precipitation amounts at multiple historical time points, and the temperatures at multiple historical time points. Among them, the water level change characteristic model characterizes the water level change characteristics under various temperature and precipitation change conditions, and the groundwater flow direction characteristic model characterizes the groundwater flow characteristics under various temperature and precipitation conditions;

[0130] A prediction module, configured to determine the predicted groundwater flow characteristics during the dewatering construction period of the foundation pit based on the precipitation prediction data, the temperature prediction data, and the underground flow direction characteristic model, and determine the positions of multiple dewatering wells based on the predicted groundwater flow characteristics, where the multiple dewatering wells are arranged around the foundation pit;

[0131] A control module, configured to generate a water pump control signal when performing foundation pit dewatering, determine the water level change amount based on the water level change characteristic model, the precipitation prediction data, and the temperature prediction data, generate a water pump adjustment signal based on the water level change amount, and adjust the water pump based on the water pump control signal and the water pump adjustment signal.

[0132] An intelligent control system for foundation pit dewatering of the present invention collects historical data on the area where foundation pit excavation and dewatering are planned to be implemented, and uses the historical data to find the laws of the groundwater level and the direction of groundwater flow under various temperature and precipitation conditions in this area. After obtaining the laws, the meteorological data at future time points can be used to predict the direction of groundwater flow and the water level, and then the positions of the dewatering wells can be arranged according to the groundwater flow mode, making the positions of the dewatering wells more reasonable. In addition, this application uses a water pump to control the water level in the dewatering well. During the control, the law of the groundwater level is used to predict the change amount of groundwater in the future time, and the change amount is used to adjust the water pump control signal in advance, so that this application has better anti-external interference ability.

Claims

1. An intelligent control method for foundation pit dewatering, characterized in that, Including the steps: Obtain the groundwater level sampling values at multiple sampling points around the foundation pit at multiple historical time points, the precipitation amounts at multiple historical time points, and the air temperatures at multiple historical time points; and obtain the precipitation prediction data and the air temperature prediction data during the precipitation construction period of the foundation pit; Construct a water level change characteristic model and a groundwater flow direction characteristic model for the area where the foundation pit is located based on the groundwater level sampling values at multiple sampling points at multiple historical time points, the precipitation amounts at multiple historical time points, and the air temperatures at multiple historical time points. Among them, the water level change characteristic model characterizes the groundwater level change characteristics under various air temperature and precipitation change conditions, and the groundwater flow direction characteristic model characterizes the groundwater flow characteristics under various air temperature and precipitation conditions; Determine the predicted groundwater flow characteristics during the precipitation construction period of the foundation pit based on the precipitation prediction data, the air temperature prediction data, and the groundwater flow direction characteristic model, and determine the positions of multiple precipitation wells based on the predicted groundwater flow characteristics, where the multiple precipitation wells are arranged around the foundation pit; When performing foundation pit dewatering, generate a water pump control signal, determine the water level change amount based on the water level change characteristic model, the precipitation prediction data, and the air temperature prediction data, generate a water pump adjustment signal based on the water level change amount, and adjust the water pump based on the water pump control signal and the water pump adjustment signal.

2. The intelligent control method for foundation pit dewatering according to claim 1, characterized in that Constructing a water level change characteristic model for the area where the foundation pit is located based on the groundwater level sampling values at multiple sampling points at multiple historical time points, the precipitation amounts at multiple historical time points, and the air temperatures at multiple historical time points includes: Calculate the change characteristic value WP_C of the groundwater level sampling values at any two adjacent time points t , the change characteristic value R_C of the precipitation t and the change characteristic value W_C of the air temperature t , where the change characteristic value is a difference or a change rate, and t represents a time point; Change characteristic value WP_C based on the same time point t , change characteristic value R_C of precipitation t and change characteristic value W_C of temperature t to construct first sample data Data_sam at multiple historical time points t , Data_sam t = (WP_C t , R_C t , W_C t ); Remove the change characteristic value WP_C of the groundwater level sampling value t The first sample data Data_sam less than the preset threshold t , to obtain the target data set; Cluster the sample data vectors in the target data set based on the change characteristic values of precipitation and the change characteristic values of air temperature to obtain the first sample clusters under various air temperature and precipitation conditions; Calculate the data amounts in multiple first sample clusters and the variances of the change characteristic values of the groundwater level sampling values, use the first sample clusters with data amounts greater than the preset quantity threshold as intermediate clusters, and perform outlier screening on the intermediate clusters with variances greater than the preset variance threshold to obtain target clusters with variances less than or equal to the preset variance threshold; Calculate the average value and standard deviation of the change characteristic values of the groundwater level sampling values in the target clusters, and construct a reference range of the change characteristic values based on the average value and the standard deviation, where the median of the reference range is the typical value.

3. The intelligent control method for foundation pit dewatering according to claim 1, characterized in that, Constructing a groundwater flow direction characteristic model for the area where the foundation pit is located based on the groundwater level sampling values at multiple sampling points at multiple historical time points, the precipitation amounts at multiple historical time points, and the air temperatures at multiple historical time points includes: For each historical time point, determine the flow vectors of the groundwater at multiple target points based on multiple groundwater level sampling values t (γ, v), where γ represents the flow direction and v represents the flow velocity; Flow vectors flow based on multiple historical time points t (γ, v), precipitation R t and temperature W t Construct second sample data Data_sam for multiple historical time points t ′, Data_sam t ′ = (flow t (γ, v), R t , W t ); Cluster multiple second sample data based on precipitation and air temperature to obtain second sample clusters under various air temperature and precipitation conditions; Calculate the data amounts in multiple second sample clusters and the variances of the flow vectors, use the second sample clusters with data amounts greater than the preset quantity threshold as intermediate clusters, and perform outlier screening on the intermediate clusters with variances greater than the preset variance threshold to obtain target clusters with variances less than or equal to the preset variance threshold; Calculate the average value and standard deviation of the flow vectors in the target cluster, and construct a reference range of the flow vectors based on the average value and the standard deviation, wherein the median of the reference range is the typical value.

4. An intelligent control method for foundation pit dewatering according to claim 3, characterized in that, Determine the flow vector flow of groundwater at multiple target points based on multiple groundwater level sampling values t (γ, v), including: Calculate the hydraulic head of multiple sampling points based on multiple groundwater level sampling values; Input the hydraulic heads of multiple sampling points into a groundwater simulated flow analysis tool to obtain a flow grid diagram, wherein the flow grid diagram includes flow vectors of multiple grids; Extract the flow vectors of multiple target points from the flow grid diagram.

5. The intelligent control method for foundation pit dewatering according to claim 1, characterized in that Determine the positions of multiple precipitation wells based on the predicted groundwater flow characteristics, including: Obtain a set of flow vectors at multiple future time points during the precipitation construction period, wherein the set of flow vectors includes the flow vectors of multiple sampling points; For any sampling point S i , calculate the mean of the flow vectors at multiple future time points to obtain the average flow vectors of multiple sampling points; Construct multiple sampling points S i Reference line S between the sampling points S and the foundation pit position O i O, and determine the reference line S i The included angle γ between O and the average flow vector of the sampling point S i And based on the included angle γ i And the average flow vector of the sampling point S i Calculate the number N of precipitation wells in the cut-off area corresponding to the sampling point S i The mathematical expression of the number N of precipitation wells is: i i The number N of precipitation wells i is:​ where round() represents rounding, and v i is the flow velocity of the average flow vector of the sampling point S i , v0 is the reference flow velocity, and n is the empirical value of the number of precipitation wells at the reference flow velocity.

6. The intelligent control method for foundation pit dewatering according to claim 1, characterized in that, Generate a water pump control signal, including: Obtain the real-time water level H of the precipitation well s , the foundation pit water level H j , the groundwater level H u and the soil permeability K of the area where the foundation pit is located, and obtain the performance curve of the water pump. Among them, the water pump performance curve includes the relationship between the power of the water pump and the pumping speed; Based on the precipitation well water level H s , the foundation pit water level H j , the groundwater level H u and the soil permeability K of the area where the foundation pit is located, calculate the infiltration flow rate Q in the precipitation well. The mathematical expression of the infiltration flow rate Q is as follows: S1 = πR1(H j -H d1 ) Wherein, L1 is the perpendicular distance between the central axis of the precipitation well and the central axis of the foundation pit, L2 is the shortest distance between the central axis of the precipitation well and the central axis of the sampling point, S1 is the contact area between the precipitation well and the water seeping out from the foundation pit, S2 is the contact area between the precipitation well and the groundwater, R1 is the diameter of the foundation pit, R2 is the diameter of the precipitation well, H d1 is the depth of the foundation pit, H d2 is the depth of the bottom of the aquifer, H d3 is the depth of the precipitation well, and L is the thickness of the aquifer; Obtain the optimal working power of the water pump and the target pumping speed V corresponding to the optimal working power from the performance curve of the water pump, and generate a water pump control signal based on the infiltration flow rate Q and the target pumping speed V, so that the water pump maintains the optimal working power as much as possible and keeps the water level of the precipitation well below the warning line.

7. An intelligent control method for foundation pit dewatering according to claim 6, characterized in that, Generate a water pump control signal based on the infiltration flow rate Q and the target pumping speed V, including: S1. Compare the infiltration flow rate Q with the target pumping speed V, and compare the water level of the precipitation well with the warning line; S2. When the target pumping speed V is greater than the infiltration flow rate Q and the water level of the precipitation well is lower than the warning line, generate a stop signal for controlling the water pump to stop, and return to S1 after waiting for a target duration; S3. When the target pumping speed V is greater than the infiltration flow rate Q and the water level of the precipitation well is higher than or equal to the warning line, generate a first control signal for controlling the water pump to work at the optimal working power, and return to S1 after waiting for a target duration; S4. When the target pumping speed V is less than or equal to the infiltration flow rate Q and the water level of the precipitation well is lower than the warning line, generate a first control signal for controlling the water pump to work at the optimal working power, and return to S1 after waiting for a target duration; S5. When the target pumping speed V is less than or equal to the infiltration flow rate Q and the water level of the precipitation well is higher than or equal to the warning line, a second control signal for updating the working power of the water pump is generated, and after waiting for the target duration, it returns to S1, where the updated power P of the updated water pump n ' satisfies: P n ' = P n + P0, P n is the power before update, and P0 is the increment.

8. An intelligent control method for foundation pit dewatering according to claim 6, characterized in that Generate a water pump adjustment signal based on the water level change amount, including: Update the groundwater level based on the water level change amounts at multiple future time points, and calculate the average infiltration flow rate Q at multiple future time points based on the updated groundwater level A ; Average infiltration flow rate Q at multiple future time points A When it is greater than or equal to the target pumping speed V, adjust the position of the warning line, and the adjusted position of the warning line is H AL ′, where the adjusted position of the warning line is H AL ′, and the mathematical expression is: where T is the duration of multiple future time points, and H AL ″ is the set lowest water level.

9. The intelligent control method for foundation pit dewatering according to claim 1, wherein Determine the water level change amount based on the water level change characteristic model, precipitation prediction data, and temperature prediction data, including: Substitute the precipitation prediction data and the temperature prediction data into the water level change characteristic model to obtain the groundwater level change characteristics at multiple future time points; Construct the water level change amount based on the current groundwater level and the groundwater level change characteristics.

10. An intelligent control system for foundation pit dewatering, characterized in that, Including: An acquisition module, configured to acquire the groundwater level sampling values, precipitation amounts at multiple historical time points, and temperatures at multiple historical time points of multiple sampling points around the foundation pit; and acquire precipitation prediction data and temperature prediction data during the precipitation construction period of the foundation pit; A model construction module for constructing a water level change characteristic model and an underground water flow direction characteristic model of the area where the foundation pit is located based on the underground water level sampling values at multiple historical time points, the precipitation amounts at multiple historical time points, and the air temperatures at multiple historical time points among the multiple sampling points. Among them, the water level change characteristic model characterizes the underground water level change characteristics under various air temperature and precipitation change conditions, and the underground water flow direction characteristic model characterizes the underground water flow characteristics under various air temperature and precipitation conditions; A prediction module for determining the predicted underground water flow characteristics during the precipitation construction period of the foundation pit based on the precipitation prediction data, the air temperature prediction data, and the underground flow direction characteristic model, and determining the positions of multiple precipitation wells based on the predicted underground water flow characteristics, where the multiple precipitation wells are arranged around the foundation pit; A control module for generating a water pump control signal during the foundation pit precipitation execution, determining the water level change amount based on the water level change characteristic model, the precipitation prediction data, and the air temperature prediction data, generating a water pump adjustment signal based on the water level change amount, and adjusting the water pump based on the water pump control signal and the water pump adjustment signal.

Images