A stability calculation method and system for a diaphragm wall of a subway station

Abstract

The present application provides a method and system for calculating the stability of a diaphragm wall of a subway station. The stability calculation method includes: S101, obtaining the construction plan of the diaphragm wall; S102, using a construction acquisition device to measure and obtain in real time the pouring depth data of the current construction trench segment and the liquid level data of the dewatering well corresponding to the current construction trench segment; S104, after the pouring of the current construction trench segment is completed, using an ultrasonic detection device to obtain the verticality data of the side wall of the construction trench segment. When the verticality data is within its corresponding preset range, input the verticality data, the change curve of the pouring depth data, and the change curve of the liquid level data into the stability scoring model; take the next construction trench segment as the current construction trench segment and execute S102. Through the comprehensive consideration of the verticality data, the change of the pouring depth data, and the liquid level data, the stability calculation during the construction process of the diaphragm wall is realized.

Description

Technical Field

[0001] This application relates to the technical field of calculation of construction quality stability, and particularly relates to a method and system for calculating the stability of diaphragm walls of subway stations. Background Art

[0002] With the rapid development of the national rail transit industry, diaphragm walls, as excellent structural engineering, have been widely used in the retaining structures of subway stations. However, when the diaphragm wall structure is constructed in some complex geological conditions, many problems will be faced. Taking the case of a deep water-rich soft silty fine sand layer as the geological condition, the soil layer under this geological condition has high liquefaction property, poor self-stability, is prone to soil disturbance, and the diaphragm wall trench wall is prone to phenomena such as necking and exposed reinforcement, collapse, and joint mud inclusion. It will also cause sand and water leakage at the excavation of the diaphragm wall joint, the steel cage cannot be lowered in place, and large-area bulging of the wall surface, seriously affecting the safe excavation of the foundation pit. In related technologies, most are to improve by controlling the mud quality. Although the possibility of these problems is reduced to a certain extent, there is still uncertainty. The calculation of the stability of the diaphragm wall is particularly important under such geological conditions. Related technologies have also proposed technical means for monitoring the stress of the diaphragm wall.

[0003] Exemplarily, the application No. CN202110965412.9 discloses a dynamic back-analysis calculation and fusion algorithm and an intelligent monitoring system. The collected deep horizontal displacement monitoring data is fed back to the intelligent monitoring platform to form a real-time displacement change curve. Through a dynamic back-analysis calculation and fusion algorithm, the monitoring of the deformation of the diaphragm wall, the internal force of the diaphragm wall, the horizontal displacement and vertical displacement of the top of the diaphragm wall is realized. That is to say, it mainly focuses on the analysis and calculation of the monitored displacement and deformation data after the overall construction is completed. What it discloses is not a calculation method based on the data of the diaphragm wall during the construction process of the diaphragm wall, and it is not suitable for timely adjusting the construction during the construction process.

[0004] Exemplarily, the application No. CN201910441572.6 discloses a device for automatic multi-parameter detection of diaphragm walls underwater, which can replace the original laborious and time-consuming manual random point selection inspection method. However, since it is applicable to the diaphragm wall used as the water cutoff, anti-seepage, load-bearing, and water retaining structure of the wharf, the provided device for automatic multi-parameter detection mainly transmits the detection images and data of the underwater execution equipment underwater in a timely manner for display, and is not applicable to non-wharf application scenarios and is not applicable to the construction of diaphragm walls of subway stations.

[0005] Based on this, this application aims to provide a method and system for calculating the stability of diaphragm walls of subway stations to solve the above problems. Summary of the Invention

[0006] The purpose of this application is to provide a method and system for calculating the stability of diaphragm walls of subway stations, which solves the problem of the lack of means for calculating the stability of diaphragm walls of subway stations.

[0007] The purpose of this application is achieved by the following technical solutions:

[0008] This application provides a method for calculating the stability of diaphragm walls of subway stations, and the stability calculation method includes:

[0009] S101, obtaining the construction plan of the diaphragm wall;

[0010] S102, pouring concrete for the current construction trench section according to the construction plan, and using a construction acquisition device to measure and obtain the pouring depth data of the current construction trench section and the liquid level data of the dewatering well corresponding to the current construction trench section in real time; the pouring depth data and the liquid level data are respectively used to indicate the changes in the pouring depth and the dewatering well liquid level.

[0011] S104, when the pouring of the current construction trench section is completed, using an ultrasonic detection device to obtain the verticality data of the side wall of the construction trench section. When the verticality data is within its corresponding preset range, input the verticality data, the change curve of the pouring depth data, and the change curve of the liquid level data into the stability scoring model to obtain the stability score of the current construction trench section and send it to the user device, take the next construction trench section as the current construction trench section and execute S102;

[0012] Wherein, the construction acquisition device includes a pressure sensor arranged on the conduit of the construction trench section of the diaphragm wall and a liquid level sensor arranged on the dewatering well of the diaphragm wall.

[0013] Furthermore, the stability scoring model is obtained by training a deep learning model, and the training process of the stability scoring model includes the following steps:

[0014] Obtaining a training set, the training set includes a plurality of training data, and each training data includes verticality data, a pouring depth change curve, and a liquid level change curve, as well as labeled data of the stability score corresponding to the verticality data, the pouring depth change curve, and the liquid level change curve;

[0015] For each training data in the training set, perform the following processing:

[0016] Input the verticality data, the pouring depth change curve, and the liquid level change curve in the training data into a preset deep learning model to obtain predicted data of the stability score corresponding to the verticality data, the pouring depth change curve, and the liquid level change curve;

[0017] Update the model parameters of the deep learning model based on the prediction data and the labeled data;

[0018] Detect whether a preset training end condition is satisfied; if so, use the trained deep learning model as the stability scoring model; if not, continue to train the deep learning model with the next piece of training data.

[0019] Further, the stability calculation method further includes:

[0020] When the number of stability scores of the construction trench sections for which the stability scores have been completed meets a preset number, obtain the stability calculation result of the diaphragm wall in the previous construction stage according to the stability scores, trench key coefficients, and stability scoring criteria of each construction trench section for which the stability scores have been completed, where the stability calculation result includes normal and abnormal;

[0021] When the stability calculation result is abnormal, generate an abnormal prompt message and send it to the user device.

[0022] Further, the stability calculation method further includes:

[0023] S103, obtain a water pump control signal according to the pouring depth data and the liquid level data, where the water pump control signal is used to control the water pump group of the dewatering well to pump water so that the water level of the dewatering well is lower than the current pouring height of the current construction trench section.

[0024] Further, the method for obtaining the water pump control signal includes:

[0025] Obtain the target liquid level data corresponding to the pouring depth interval where the pouring depth data is located;

[0026] According to the target liquid level data and the liquid level data, generate a water pump control signal in a PID manner to control the operation of the water pump group so that the actual liquid level meets the requirements of the target liquid level.

[0027] Further, the generating a water pump control signal in a PID manner according to the target liquid level data and the liquid level data includes:

[0028] Obtain the target liquid level data and the predicted liquid level data corresponding to the pouring depth interval where the pouring depth data is located, where the predicted liquid level data is obtained according to a preset target liquid level curve;

[0029] When the difference between the target liquid level data and the predicted liquid level data is within its corresponding preset range, update the target liquid level data according to the predicted liquid level data;

[0030] When the difference between the target liquid level data and the predicted liquid level data is not within its corresponding preset range, update the target liquid level curve according to the target liquid level data;

[0031] Calculate the error according to the target liquid level data and the liquid level data, and input the error into a pre-set PID control algorithm to obtain a water pump control signal.

[0032] Further, arrange a plurality of dewatering wells in the foundation pit of the diaphragm wall, and use each dewatering well as a precipitation well to drain water before and during concrete pouring to lower the groundwater level, so that the water in the soil above the water level gradient line seeps down, reducing the fluidity of the soil while causing consolidation of the sandy soil layer and reducing the lateral pressure of groundwater on the groove wall.

[0033] Further, the construction plan of the diaphragm wall further includes: wrapping color-striped cloth around the positions of the soft soil layer on the steel reinforcement cage of the diaphragm wall and within the range of 3m above and below, for isolating the contact between the soil on both sides of the construction groove section and the steel reinforcement.

[0034] Further, the conduit refers to the perfusion conduit of the current construction groove section.

[0035] In a second aspect, the present application further provides a stability calculation system for the diaphragm wall of a subway station. The stability calculation system includes a controller, a construction acquisition device and an ultrasonic detection device connected to the controller. The construction acquisition device includes a pressure sensor disposed on the conduit of the construction groove section of the diaphragm wall and a liquid level sensor disposed on the precipitation well of the diaphragm wall; the controller is configured to:

[0036] S101, obtain the construction plan of the diaphragm wall;

[0037] S102, perform concrete pouring for the current construction groove section according to the construction plan, and use the construction acquisition device to measure and obtain the pouring depth data of the current construction groove section and the liquid level data of the precipitation well corresponding to the current construction groove section in real time; the pouring depth data and the liquid level data are respectively used to indicate the changes in the pouring depth and the liquid level of the precipitation well.

[0038] S104, when the current construction groove section is poured, use the ultrasonic detection device to obtain the perpendicularity data of the side wall of the construction groove section. When the perpendicularity data is within its corresponding preset range, input the perpendicularity data, the change curve of the pouring depth data and the change curve of the liquid level data into the stability scoring model to obtain the stability score of the current construction groove section and send it to the user device, and use the next construction groove section as the current construction groove section and execute S102.

[0039] Further, the stability scoring model is obtained by training a deep learning model, and the training process of the stability scoring model includes the following steps:

[0040] Obtain a training set, where the training set includes a plurality of training data, and each training data includes verticality data, a casting depth change curve, and a liquid level change curve, as well as annotation data of the stability score corresponding to the verticality data, the casting depth change curve, and the liquid level change curve;

[0041] For each training data in the training set, perform the following processing:

[0042] Input the verticality data, the casting depth change curve, and the liquid level change curve in the training data into a preset deep learning model to obtain prediction data of the stability score corresponding to the verticality data, the casting depth change curve, and the liquid level change curve;

[0043] Update the model parameters of the deep learning model based on the prediction data and the annotation data;

[0044] Detect whether a preset training end condition is satisfied; if so, use the trained deep learning model as the stability scoring model; if not, continue to train the deep learning model with the next training data.

[0045] Further, the controller is further configured to:

[0046] When the number of stability scores of the construction trench sections for which the stability scores have been completed meets a preset number, obtain the stability calculation result of the diaphragm wall in the previous construction stage according to the stability scores of each construction trench section for which the stability scores have been completed, the trench key coefficient, and the stability scoring standard, and the stability calculation result includes normal and abnormal;

[0047] When the stability calculation result is abnormal, generate an abnormal prompt message and send it to the user device.

[0048] Further, the controller is further configured to:

[0049] S103, obtain a water pump control signal according to the casting depth data and the liquid level data, and the water pump control signal is used to control the water pump group of the dewatering well to pump water so that the water level of the dewatering well is lower than the current casting height of the current construction trench section.

[0050] Further, the controller obtains the water pump control signal in the following manner:

[0051] Obtain the target liquid level data corresponding to the casting depth interval where the casting depth data is located;

[0052] Generate a water pump control signal in a PID manner based on the target liquid level data and the liquid level data, and control the operation of the water pump group to make the actual liquid level meet the requirements of the target liquid level.

[0053] Further, the controller generates a water pump control signal in a PID manner according to the target liquid level data and the liquid level data in the following way:

[0054] Obtain the target liquid level data and the predicted liquid level data corresponding to the pouring depth interval where the pouring depth data is located. The predicted liquid level data is obtained according to a preset target liquid level curve.

[0055] When the difference between the target liquid level data and the predicted liquid level data is within its corresponding preset range, update the target liquid level data according to the predicted liquid level data.

[0056] When the difference between the target liquid level data and the predicted liquid level data is not within its corresponding preset range, update the target liquid level curve according to the target liquid level data.

[0057] Calculate the error based on the target liquid level data and the liquid level data, and input the error into a preset PID control algorithm to obtain the water pump control signal.

[0058] Further, arrange a plurality of dewatering wells in the foundation pit of the diaphragm wall, and use each dewatering well as a precipitation well to drain water before and during concrete pouring to lower the groundwater level, so that the water in the soil above the water level gradient seeps down, reducing the fluidity of the soil while causing consolidation of the sandy soil layer and reducing the lateral pressure of the groundwater on the groove wall.

[0059] Further, the construction plan of the diaphragm wall also includes: wrapping colored strip cloth on the weak soil layer position of the steel cage of the diaphragm wall and within the range of 3m above and below, for separating the contact between the soil on both sides of the construction groove section and the steel bars.

[0060] Further, the conduit refers to the perfusion conduit of the current construction groove section.

[0061] Advantages of the present application: By the collaborative calculation of various data acquisitions, the construction safety and quality are ensured. Specifically, the concrete pouring of the current construction trench section is carried out according to the construction plan. The pouring depth data of the current construction trench section is obtained in real time through the pressure sensor arranged on the conduit, and the corresponding liquid level data is obtained in real time through the liquid level sensor arranged on the dewatering well. These data are used to indicate the changes in the pouring depth and the liquid level of the dewatering well. After the pouring of the current construction trench section is completed, the verticality data of the side wall of the construction trench section is obtained by using an ultrasonic detection device. When the verticality data is within the preset range, the change curves of this data, the pouring depth data, and the liquid level data are input into the stability scoring model to generate the stability score of the current construction trench section and send it to the user device, realizing the stability calculation of the diaphragm wall of the subway station.

[0062] The technical solution provided in this embodiment obtains the verticality data by using an ultrasonic detection device after each construction trench section is completed, and takes into account the influence of the changes in the pouring depth and the liquid level of the dewatering well during the perfusion process on the stability of the current construction trench section of the diaphragm wall. Combining the change curves of the pouring depth and the liquid level data, a comprehensive evaluation is carried out through the stability scoring model. Through the above calculation method, a stability score reflecting the stability status of the wall is provided, realizing the stability calculation of the diaphragm wall during the construction process. Once the monitored stability is not ideal, the subsequent construction plan of the diaphragm wall can be adjusted in time. By using a pressure sensor, a liquid level sensor, and an ultrasonic detection device, in cooperation with the stability scoring model, the calculation results make the monitoring and control process more intelligent and precise. The provided construction data acquisition device is more suitable for the stability calculation of the diaphragm wall of the subway station than the device for underwater multi-parameter automatic detection of the diaphragm wall mentioned in the background technology.

[0063] In summary, the technical solution comprehensively considers the changes in the pouring depth and the liquid level of the dewatering well reflected by the real-time obtained pouring depth data and liquid level data, and the verticality data, participates in the stability calculation during the construction process of the diaphragm wall, effectively ensuring the stability and safety of the construction process. Through the stability calculation results, comprehensive stability monitoring and management means are provided for the construction of the diaphragm wall, creating conditions for timely adjustment of the construction quality and construction plan, and ensuring the smooth progress of the project. Description of the Drawings

[0064] The present application will be further described below with reference to the drawings and embodiments.

[0065] Figure 1 It is a schematic flow chart of a stability calculation method provided by an embodiment of the present application.

[0066] Figure 2 It is a schematic flow chart of a method for obtaining a water pump control signal provided by an embodiment of the present application.

[0067] Figure 3 It is a schematic flow diagram for generating a water pump control signal provided by an embodiment of the present application.

[0068] Figure 4 It is a partial schematic flow diagram of a stability calculation method provided by an embodiment of the present application.

[0069] Figure 5 It is a schematic flow diagram of another stability calculation method provided by an embodiment of the present application. Detailed implementation manners

[0070] Next, in combination with the accompanying drawings and specific implementation manners, the present application will be further described. It should be noted that, on the premise of no conflict, the following-described embodiments or technical features can be arbitrarily combined to form new embodiments. The following will illustrate the implementation procedures of the present application with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present application from the content disclosed in this specification. The present application can also be implemented or applied through other different specific implementation procedures. The details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present application. It should be understood that the preferred embodiments are only for explaining the present application, rather than for limiting the protection scope of the present application.

[0071] Hereinafter, some terms in the present application will be explained to facilitate the understanding of those skilled in the art.

[0072] The diaphragm wall is a peripheral retaining structure for forming an underground structure. The diaphragm wall mentioned in the present application refers to the diaphragm wall for forming the peripheral retaining structure of a subway station, which is used to bear the earth pressure and water pressure and transfer them to the deep stable soil layer to ensure the stability of the subway station structure.

[0073] Machine Learning (ML) is an interdisciplinary subject involving multiple fields such as probability theory, statistics, approximation theory, convex analysis, and algorithm complexity theory. By using statistical and algorithmic techniques, a computer can learn from data and make decisions or predictions. The core concept is to use data to train a model and then use the trained model to score, predict, or classify new data.

[0074] Deep learning is a special type of machine learning that represents and realizes great functions and flexibility by learning to use nested concept hierarchies, where each concept is defined in relation to simple concepts, and more abstract representations are calculated in a less abstract way. Machine learning and deep learning generally include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and learning from demonstration.

[0075] The main purpose of the stability calculation of the diaphragm wall is to ensure safety and quality during construction, prevent accidents such as wall instability and collapse caused by changes in groundwater, geological conditions or improper construction, so as to ensure the smooth progress of the project and the reliability of the final structure. In the related technologies, only the further calculation after the completion of the construction of the diaphragm wall is considered. However, the present application provides a stability calculation method and system for the diaphragm wall of a subway station. By reusing the changes in the pouring depth and the water level of the dewatering well reflected by the real-time obtained pouring depth data and water level data, the stability calculation during the construction process of the diaphragm wall is realized, effectively ensuring the stability and safety during the construction process. The stability calculation results provide a comprehensive means of stability monitoring and management for the construction of the diaphragm wall, improving the construction quality and the conditions for timely adjustment of the construction plan, and ensuring the smooth progress of the project. In specific applications, the stability calculation method provided by the present application can be used for preliminary and rapid monitoring during the construction process of the diaphragm wall, and then the means provided by the related technologies can be used for continuous monitoring after the construction is completed. The method will be described first below, and then the system and others will be described.

[0076] Method embodiment.

[0077] See Figure 1 , Figure 1 is a schematic flow chart of a stability calculation method provided by an embodiment of the present application.

[0078] The present embodiment provides a stability calculation method for the diaphragm wall of a subway station, including:

[0079] S101, obtaining the construction plan of the diaphragm wall;

[0080] S102, according to the construction plan, perform concrete pouring for the current construction trench, and use the construction acquisition device to measure and obtain the pouring depth data of the current construction trench and the water level data of the dewatering well corresponding to the current construction trench in real time; the pouring depth data and the water level data are respectively used to indicate the changes in the pouring depth and the water level of the dewatering well.

[0081] S104, when the pouring of the current construction trench is completed, use an ultrasonic detection device to obtain the perpendicularity data of the side wall of the construction trench. When the perpendicularity data is within its corresponding preset range, input the perpendicularity data, the change curve of the pouring depth data and the change curve of the water level data into the stability scoring model to obtain the stability score of the current construction trench and send it to the user device; take the next construction trench as the current construction trench and execute S102.

[0082] Among them, the construction acquisition device includes a pressure sensor disposed on a conduit in a construction slot section of the diaphragm wall and a liquid level sensor disposed in a dewatering well of the diaphragm wall.

[0083] The stability calculation method provided in this embodiment ensures construction safety and quality through collaborative calculation of various acquired data. Specifically, concrete pouring of the current construction slot section is carried out according to the construction plan. The pouring depth data of the current construction slot section is obtained in real time through the pressure sensor disposed on the conduit, and the corresponding liquid level data is obtained in real time through the liquid level sensor disposed in the dewatering well. The obtained data is used to indicate the changes in the pouring depth and the liquid level of the dewatering well. After the pouring of the current construction slot section is completed, the ultrasonic detection device is used to obtain the perpendicularity data of the side wall of the construction slot section. When the perpendicularity data is within the preset range, the change curves of this data, the pouring depth data, and the liquid level data are input into the stability scoring model to generate the stability score of the current construction slot section and send it to the user device, realizing the stability calculation of the diaphragm wall of the subway station.

[0084] Among them, the pressure sensor can be installed inside or on the outer wall of the concrete pouring conduit in the construction slot section of the diaphragm wall. By monitoring the pressure change inside or outside the conduit, the filling depth of the concrete can be inferred. The conduit can be a perfusion conduit. As the concrete is poured, the pressure inside the conduit will increase accordingly. The sensor converts the pressure into an electrical signal to reflect the concrete pouring situation in real time. Combining with the depth of the perfusion conduit in the current construction slot section, the pouring depth of the concrete is monitored in real time. The conduit can also be a conduit separately disposed in the construction slot section for the pressure sensor to monitor the filling depth of the concrete, so as to reflect the concrete perfusion situation in real time. The liquid level sensor is installed in the corresponding dewatering well of the diaphragm wall and is used to monitor the change of the underground water level in the well in real time. The liquid level sensor mentioned in this application can be a resistive liquid level sensor, which uses the resistance change caused by the liquid level change to detect the liquid level height. The ultrasonic detection device can measure the deviation between the concrete side wall and the vertical direction by emitting and receiving ultrasonic waves, and evaluate the perpendicularity of the construction slot section after completion through the obtained perpendicularity data. The perpendicularity data can be expressed as an angle, that is, the angle at which the side wall of the construction slot section tilts to one side. In specific applications, it can be that the perpendicularity data is obtained by measuring any one side wall; it can also be that for the perpendicularity data of two adjacent construction slot sections, the side walls on different sides are measured. For example, for one construction slot section, the inner side wall is measured, and for the other construction slot section, the outer side wall is measured. The advantage of doing this is that it can avoid the need for the same calculation for the other side, thereby saving calculation resources; the smaller amount of data can speed up the data analysis, enabling the user to make decisions faster. In this case, the perpendicularity data also includes identification information of the measurement type to indicate the measurement method.

[0085] Compared with the related art that only considers further calculations after the construction of the diaphragm wall is completed, the technical solution provided in this embodiment obtains verticality data using an ultrasonic detection device after each construction trench is completed, and takes into account the changes in the pouring depth and the water level of the dewatering well during the pouring process, as well as the influence on the stability of the current construction trench of the diaphragm wall. By combining the change curves of the pouring depth and the water level data, a comprehensive evaluation is carried out through a stability scoring model, and a stability score reflecting the stability status of the wall is provided through the above calculation method, realizing the stability calculation of the diaphragm wall during the construction process. Once the monitored stability is not ideal, the subsequent construction plan of the diaphragm wall can be adjusted in a timely manner. By using a pressure sensor, a water level sensor, and an ultrasonic detection device, and cooperating with the stability scoring model, the calculation results make the monitoring and control process more intelligent and precise.

[0086] In summary, by using the changes in the pouring depth and the water level of the dewatering well reflected by the real-time obtained pouring depth data and water level data to participate in the stability calculation during the construction process of the diaphragm wall, the stability and safety of the construction process are effectively guaranteed. The stability calculation results provide a comprehensive means of stability monitoring and management for the construction of the diaphragm wall, creating conditions for timely adjustment of the construction quality and construction plan, and ensuring the smooth progress of the project.

[0087] At the same time, the construction acquisition device provided in this technical solution is obviously more suitable for the stability calculation of the diaphragm wall of the subway station mentioned in this application compared to the device for underwater multi-parameter automatic detection of the diaphragm wall mentioned in the background art.

[0088] In an exemplary embodiment, the stability calculation method further includes:

[0089] S103. Obtain a water pump control signal according to the pouring depth data and the water level data, where the water pump control signal is used to control the water pump group of the dewatering well to pump water, so that the water level of the dewatering well is lower than the current pouring height of the current construction trench.

[0090] It can be understood that during the concrete pouring process, the real-time obtained pouring depth data and water level data are reused to generate a water pump control signal for controlling the water pump group of the dewatering well to pump water, ensuring that the water level of the dewatering well is always lower than the current pouring height of the current construction trench, so as to reduce the lateral pressure of groundwater on the groove wall and prevent the wall from collapsing. Among them, S103 can be executed after S102.

[0091] The advantages of doing so are as follows: through real-time monitoring based on calculation results and controlling the groundwater level through the water pump control signal, the pressure of groundwater on the groove wall can be reduced, effectively preventing accidents such as groove wall collapse, and improving construction safety. By reusing the pouring depth data and liquid level data obtained in real time, the water pump control signal can be generated in a timely manner, dynamically adjusting the pumping volume of the dewatering wells, maintaining an appropriate groundwater level, and ensuring the stability of the groove wall. Reusing the pouring depth data and liquid level data for calculation can avoid repeated data collection and improve data utilization efficiency. By comprehensively utilizing the existing data, precise control of the water pump can be achieved, ensuring the construction stability of the diaphragm wall.

[0092] See Figure 2 , Figure 2 which is a schematic flow chart of obtaining a water pump control signal provided by an embodiment of the present application.

[0093] In an exemplary embodiment, the method for obtaining the water pump control signal includes:

[0094] S201, obtaining target liquid level data corresponding to the pouring depth interval where the pouring depth data is located;

[0095] S202, generating a water pump control signal in a PID manner according to the target liquid level data and the liquid level data, and controlling the operation of the water pump group so that the actual liquid level meets the requirements of the target liquid level.

[0096] First, determine the pouring depth interval where the current pouring depth data is located. Each depth interval corresponds to a preset target liquid level data. It can be understood that according to the engineering design and construction plan, multiple consecutive pouring depth intervals are preset. For example, the multiple consecutive pouring depth intervals are greater than 33m and not less than 36m, greater than 36m and not less than 39m, etc. Then each pouring depth interval corresponds to a target liquid level data (such as 30m, 33m, 45m). Whichever pouring depth interval the pouring depth indicated by the pouring depth data is in, the target liquid level data corresponding to that interval is taken as the target liquid level data of the pouring depth data. It can be considered that the correspondence between the pouring depth interval and the target liquid level data is preset according to the engineering design and construction plan, reflecting the ideal state of the expected groundwater level or dewatering well water level under a specific depth interval.

[0097] The PID (Proportional-Integral-Derivative) control algorithm is used to generate a water pump control signal to adjust and control the operation of the water pump group. During specific operation, first, the target liquid level data and the (actual) liquid level data are compared, and the error value is calculated based on the difference between the actual liquid level and the target liquid level as the deviation amount. The deviation amount indicates the distance between the current liquid level and the target liquid level, reflecting the degree of water pump adjustment required. Among them, the PID control algorithm calculates the control outputs of the proportional term (P term), integral term (I term), and derivative term (D term) respectively according to the deviation amount. Specifically:

[0098] The proportional term (P term) refers to generating a control quantity proportional to the error according to the current error amount for immediate adjustment.

[0099] The integral term (I term) refers to considering the cumulative history of the error to eliminate long-term deviations.

[0100] The derivative term (D term) refers to predicting the trend of error change to suppress future overshoot phenomena.

[0101] The control quantity calculated by the PID algorithm is converted into the operation parameters of the water pump group, including instructions such as start, stop, and speed regulation, to affect the pumping efficiency and water level adjustment ability of the water pump group.

[0102] The advantage of doing this is that by precisely matching the pouring depth interval with the target liquid level, precise control of the groundwater level or the water level in the precipitation well can be achieved, meeting the construction requirements at different depths. Obtaining the target liquid level data in intervals can maintain relatively stable control parameters within different pouring depth intervals, without having to respond to each tiny depth change, thus avoiding the situation of frequently adjusting the target liquid level data due to changes in the actual pouring depth and simplifying the complexity of real-time adjustment.

[0103] See Figure 3 , Figure 3 which is a schematic flowchart of a process for generating a water pump control signal provided by an embodiment of the present application.

[0104] In an exemplary embodiment, generating the water pump control signal in a PID manner according to the target liquid level data and the liquid level data includes:

[0105] S301, obtaining the target liquid level data and the predicted liquid level data corresponding to the pouring depth interval where the pouring depth data is located, and the predicted liquid level data is obtained according to a preset target liquid level curve;

[0106] S302, when the difference between the target liquid level data and the predicted liquid level data is within its corresponding preset range, updating the target liquid level data according to the predicted liquid level data;

[0107] S303. When the difference between the target liquid level data and the predicted liquid level data is not within its corresponding preset range, update the target liquid level curve according to the target liquid level data;

[0108] S304. Calculate the error based on the target liquid level data and the liquid level data, and input the error into a preset PID control algorithm to obtain a water pump control signal.

[0109] Among them, the target liquid level curve can be pre-set data in the construction plan, which consists of a set of data points. These data points describe the desired groundwater level height at different pouring depths. The data points in the target liquid level curve can be obtained through engineering design and simulation. In practical applications, the target liquid level curve can be discrete data points, or can be obtained in the form of a continuous curve through interpolation or fitting to facilitate the acquisition of predicted liquid level data. In this case, the predicted liquid level data is a predicted value calculated based on the preset target liquid level curve and the current depth.

[0110] When the difference between the target liquid level data and the predicted liquid level data is within the preset range, directly use the predicted liquid level data as the target liquid level data. This is because the target liquid level data in specific applications is obtained based on the interval of pouring depth data, which is used to save the calculation process during the construction of the diaphragm wall. The pouring depth data is a value obtained after processing the sensing signals acquired by multiple sensors and is used to indicate the pouring depth situation, while the predicted liquid level data is calculated based on system modeling during the formulation of the construction plan. When the difference between the target liquid level data and the predicted liquid level data is within the preset range, it indicates that the prediction model has a high effectiveness. Directly using the predicted liquid level data can improve the prediction accuracy of the water pump group control process and strengthen the data-based decision-making process, making the construction process more scientific and accurate. When the difference exceeds the preset range, the target liquid level curve will be adjusted and updated according to the target liquid level data obtained from the pouring depth interval to ensure the accurate use of the target liquid level curve for subsequent construction segments.

[0111] The advantage of doing this is that the predicted liquid level data is calculated based on system modeling and historical data, which reflects the trend and law of groundwater level changes. On the premise that the target liquid level curve is accurately judged, directly using the predicted liquid level data can improve the prediction accuracy of groundwater level changes during the water pump group control process.

[0112] In an exemplary embodiment, the step of updating the target liquid level curve according to the target liquid level data when the difference between the target liquid level data and the predicted liquid level data is not within its corresponding preset range includes:

[0113] When the difference between the target liquid level data and the predicted liquid level data is not within the preset range, a new target liquid level curve is generated using interpolation or fitting techniques based on the target liquid level data. As an example, spline interpolation is performed using the SciPy library in Python, an interpolation function is created using the interp1d function in the SciPy library, and the updated target liquid level curve is generated using the interpolation function.

[0114] In an exemplary embodiment, the stability scoring model is obtained by training a deep learning model, and the training process of the stability scoring model includes the following steps:

[0115] Obtain a training set, where the training set includes a plurality of training data, and each training data includes verticality data, a pouring depth change curve, and a liquid level change curve, as well as annotation data of the stability score corresponding to the verticality data, the pouring depth change curve, and the liquid level change curve;

[0116] For each training data in the training set, perform the following processing:

[0117] Input the verticality data, the pouring depth change curve, and the liquid level change curve in the training data into a preset deep learning model to obtain predicted data of the stability score corresponding to the verticality data, the pouring depth change curve, and the liquid level change curve;

[0118] Based on the predicted data and the annotation data, update the model parameters of the deep learning model;

[0119] Detect whether a preset training end condition is satisfied; if so, use the trained deep learning model as the stability scoring model; if not, continue to train the deep learning model using the next training data.

[0120] The stability scoring model is used to predict the stability score of the construction trench section based on the change trends of the pouring depth and the precipitation well liquid level obtained by the construction acquisition device and the verticality data after the current construction trench section is poured. For example, during the training process, a large number of sample data are used, including verticality data, a pouring depth change curve, and a liquid level change curve, as well as the corresponding stability scores. Through these data, a preset deep learning model is trained. Based on the predicted data output by the deep learning model and the actually annotated stability score data, the backpropagation algorithm is used to adjust and update the parameters of the deep learning model so that it can more accurately predict the influence of each variable in the diaphragm wall construction on the stability. Detect whether a preset training end condition is satisfied, such as reaching the maximum number of iterations or the model performance stabilizing within a certain threshold range. If the end condition is satisfied, the stability scoring model is obtained; if not, continue to train the model using the next training data.

[0121] In a specific application, the pouring depth change curve can be represented as the first time series data of the pouring depth data, the liquid level change curve can be represented as the second time series data of the liquid level data, and the verticality data is represented as a single data point. The selected deep learning model is a fusion model that combines a recurrent neural network (RNN), specifically using a long short-term memory network (LSTM) to process the two time series data sets, and a convolutional neural network (CNN) to extract local features. The model includes a fully connected layer to process the single data point (verticality data). In terms of the model architecture, the input layer receives the two time series data sets (pouring depth change curve and liquid level change curve) and a verticality data point. On the feature extraction layer, for the time series data, the LSTM layer and the CNN layer are used to extract features, while for the verticality data point, it is directly fed into the fully connected layer. The fusion layer integrates different features to facilitate capturing the mutual relationships between them. The output layer finally outputs the stability score based on the fused features. Thus, complex non-linear relationships can be captured, improving the accuracy and reliability of the score.

[0122] The advantage of this is that using a deep learning model for the training and prediction of the stability score realizes a certain degree of automated processing. The deep learning model can learn and optimize based on a large amount of data, reducing the need for manual intervention and improving efficiency and consistency. By design, by establishing an appropriate number of neuron computing nodes and a multi-layer operation hierarchy, and selecting the appropriate input layer and output layer, a preset deep learning model can be obtained. Through the learning and optimization of this deep learning model, a functional relationship from input to output is established. Although the functional relationship between input and output cannot be found 100%, it can approximate the real correlation relationship as much as possible. The stability score model obtained by such training can obtain the corresponding stability score based on the verticality data, pouring depth change curve, and liquid level change curve of the same construction slot section, with good calculation accuracy and high reliability.

[0123] In the embodiments of the present application, a stability score model can be trained. In some other alternative embodiments, the present application can adopt a pre-trained stability score model. The preset deep learning model can be a convolutional neural network model or a recurrent neural network model. The present application does not limit the preset training end condition, which can be, for example, that the number of training times reaches a preset number (the preset number can be, for example, 1 time, 3 times, 10 times, 100 times, 1000 times, 10000 times, etc.), or it can be that all the training data in the training set have completed one or more trainings, or it can be that the total loss value obtained in this training is not greater than the preset loss value.

[0124] The training data in the training set can be historical construction data and monitoring records, including verticality data, pouring depth change curves, and liquid level change curves obtained from multiple sets of on-site data collection, as well as the scores given by technicians; it can also be simulated data generated by modeling and simulation based on known physical models and assumed conditions to simulate the stability under different combinations of verticality data, pouring depth change curves, and liquid level change curves.

[0125] In a specific application, when the verticality data for two adjacent construction slots are measured on the side walls of different sides, the verticality data also includes identification information of the measurement type to indicate the measurement method. In this case, the training set of the stability scoring model clearly identifies whether each (verticality data) data point is the measurement result of the inner side wall or the outer side wall. The stability scoring model can learn the characteristics of the inner and outer side wall data and take this into account during prediction, thereby improving the generalization ability of the model, avoiding the establishment and maintenance of separate models for each side wall type, and simplifying model management.

[0126] See Figure 4 , Figure 4 is a partial flowchart of a stability calculation method provided by an embodiment of the present application.

[0127] In an exemplary embodiment, the stability calculation method further includes:

[0128] S105, when the number of stability scores of the construction slots for which stability scoring has been completed meets a preset number, obtain the stability calculation result of the diaphragm wall in the previous construction stage according to the stability score, slot key coefficient, and stability scoring standard of each construction slot for which stability scoring has been completed. The stability calculation result includes normal and abnormal.

[0129] S106, when the stability calculation result is abnormal, generate an abnormal prompt message and send it to the user device.

[0130] In a specific application, the slot key coefficient can indicate the positional importance of the construction slot in the overall diaphragm wall. For example, whether the slot is located at the corner, edge, or key support part of the structure. If so, the coefficient is greater than 1 (for example, preset to 1.2), and if not, the coefficient is 1; when the slot key coefficient is greater than 1, when obtaining the stability calculation result of the diaphragm wall in the previous construction stage according to the stability score, slot key coefficient, and stability scoring standard of each construction slot for which stability scoring has been completed, copy the stability score of the corresponding construction slot as the stability scores of two construction slots for stability calculation result judgment. When the slot key coefficient is 1, calculate normally.

[0131] Among them, the preset quantity is, for example, 5, 8, 12, etc., the stability score obtained by the stability scoring model is, for example, 90, 75, A, C, etc., and the stability scoring standard is, for example, when the average of the stability scores of multiple construction trench segments is 85, 75, or the mode is B, C or above, the monitoring result is normal. The higher the number, the better the stability, and the earlier the letter is sorted in the positive order, the better the stability.

[0132] As an example, the stability scores are A, A, B, C, C, C respectively, the key coefficients of the trench segments are all 1, and the stability scoring standard is that the mode of the stability scores is B or above. Then it is considered that the mode of the stability scores is C, and the monitoring result is abnormal, indicating that even if the verticality data is within its corresponding preset range, there are potential hazards in the stability of the current diaphragm wall and further inspection is required.

[0133] As another example, the stability scores are A, A, B, C, C, C respectively, and the corresponding key coefficients of the trench segments are 1.2, 1.2, 1, 1, 1 respectively. The stability scores corresponding to the key coefficients of the trench segments greater than 1 are copied respectively, and all the obtained stability scores are: A, A, A, A, B, C, C, C. The stability scoring standard is that the mode of the stability scores is B or above. Then it is considered that the mode of the stability scores is A, and the monitoring result is normal, indicating that the verticality data is within its corresponding preset range and there are no potential hazards in the stability of the current diaphragm wall.

[0134] This application does not limit the user equipment, such as tablet computers, industrial computers, laptops, etc. The manifestation form of the abnormal prompt information can be voice notification, text message notification or APP pop-up window, etc.

[0135] After the scoring is completed, according to the stability score, trench segment key coefficient and stability scoring standard of each construction trench segment, the stability calculation result of the trench segment in the previous construction stage is obtained. The stability calculation result includes normal and abnormal states. The normal state indicates that the stability of the construction trench segment is good and there are no obvious problems; the abnormal state generates an abnormal prompt information and sends it to the user equipment of the project supervisor or the on-site person in charge to take necessary measures in time.

[0136] The advantage of doing this is that by real-time monitoring and evaluating the stability of multiple completed construction trench segments, potential risks can be immediately discovered, which is convenient for providing an early warning and feedback mechanism and reducing potential structural problems.

[0137] In an exemplary embodiment, a plurality of dewatering wells are arranged in the foundation pit of the diaphragm wall, and each dewatering well is used as a precipitation well to drain water before and during concrete pouring to lower the groundwater level, so that the water in the soil body above the water level gradient line seeps down, reducing the fluidity of the soil body while causing consolidation of the sandy soil layer and reducing the lateral pressure of the groundwater on the trench wall.

[0138] A plurality of dewatering wells are pre-arranged in the diaphragm wall foundation pit, and these dewatering wells are used as precipitation wells to perform drainage operations before and during concrete pouring. Through the pumping function of the precipitation wells, the groundwater level in the foundation pit is lowered, and the lateral pressure of the groundwater on the groove wall is reduced. The precipitation activity causes the moisture in the soil above the water level gradient line to seep downward, thereby reducing the moisture content in the soil. As the moisture decreases, the sandy soil layer begins to consolidate, enhancing the overall stability of the soil mass. The consolidation of the soil mass and the reduction of moisture reduce the lateral pressure of the groundwater on the groove wall and lower the risk of groove wall collapse. Concrete pouring is carried out under the condition that the groundwater level is effectively controlled to ensure the pouring quality and the stability of the groove section.

[0139] In an exemplary embodiment, the construction plan of the diaphragm wall further includes: wrapping color-striped cloth around the position of the soft soil layer on the steel reinforcement cage of the diaphragm wall and within a range of 3 m above and below it, for isolating the contact between the soil masses on both sides of the construction groove section and the steel bars, and avoiding the intrusion of the groove wall soil mass into the steel reinforcement cage or leaning against the steel reinforcement cage due to the shrinkage of the construction groove section, resulting in mud inclusion and leakage of steel bars in the diaphragm wall.

[0140] In the construction of the diaphragm wall, the steel reinforcement cage is the main support structure of the wall. Color-striped cloth is wrapped around the position where the steel reinforcement cage contacts the soft soil layer and within a range of 3 m above and below it. The function of the color-striped cloth is to isolate the direct contact between the soil masses on both sides of the construction groove section and the steel reinforcement cage, forming a physical barrier. The isolation measures provided in this embodiment help to avoid the mud inclusion phenomenon caused by soil intrusion and the problem of leakage of steel bars in the diaphragm wall due to poor contact between the soil and the steel bars.

[0141] See Figure 5 , Figure 5 which is a schematic flow chart of another stability calculation method provided by the embodiment of the present application.

[0142] In a specific application, a method for calculating the stability of the diaphragm wall of a subway station is provided, including

[0143] Starting, obtain the construction plan. The construction plan includes the construction arrangement of each construction groove section and the identification information of the construction groove section, such as the 1# construction groove section, the 2# construction groove section; it also includes the target liquid level curve, and the target liquid level curve includes a plurality of data points for indicating the predicted liquid level data corresponding to different pouring depth intervals. Perform the concrete pouring and data collection of the 1# construction groove section. While pouring the concrete, use instruments (construction acquisition devices) to measure the depth of the concrete and the corresponding water level of the precipitation well, and at the same time perform liquid level control. The process of liquid level control includes:

[0144] Obtain the target liquid level data based on the pouring depth data, obtain the target liquid level curve, (from the target liquid level curve) obtain the predicted liquid level data, determine whether the difference between the two (the target liquid level data and the predicted liquid level data) is within the corresponding preset range. When it is within the preset range, (according to the predicted liquid level data) update the target liquid level data. When it is not within the preset range, (according to the target liquid level data) update the target liquid level curve; Calculate the error based on the target liquid level data and the liquid level data and input it into the pre-set PID control algorithm to obtain the water pump control signal, and control the water pump group in the PID mode to ensure that the water level is maintained at an appropriate position. After the pouring of the 1# construction slot section is completed, use ultrasonic waves to detect its verticality, and judge whether there is a problem with the verticality. If not, perform a stability score for the 1# construction slot section according to the changes in verticality, concrete depth, and water level. If the verticality detection shows a problem, that is, the verticality data of the wall is not within the preset range, it means that there are deviations or errors during the construction process. Send a prompt message to the user equipment, and the project supervision personnel or the on-site person in charge will handle it. The project supervision personnel or the on-site person in charge analyze the reasons for the verticality deviation from aspects such as unstable formwork support, uneven foundation settlement, and improper construction, and take corresponding corrective measures according to the reasons for the problem. After processing, re-judge the verticality.

[0145] Judge whether the number of stability scores is greater than 5. If so, obtain the stability calculation result of the diaphragm wall in the previous construction stage according to each stability score, the key coefficient of the stability slot section, and the stability score standard, and judge whether the result is normal. When the state is abnormal, generate an abnormal prompt message and send it to the project supervision personnel or the on-site person in charge.

[0146] In another specific application, a construction plan for a diaphragm wall is provided. The construction plan includes the following construction steps: 1. Construction preparation; 2. Guide wall construction; 3. Pipe well advance dewatering and slurry preparation; 4. Grooving; 5. Cleaning sediment and replacing slurry; 6. Brushing the I-beam joint; 7. Manufacturing the steel cage and setting anti-bypass iron sheets; 8. Lifting and placing the steel cage; 9. Setting the anti-sand layer; 10. Lifting and placing the lock pipe; 11. Pouring concrete; 12. Pulling out the lock pipe; 13. Grooving the next construction slot section and obtaining the stability calculation result in the previous construction stage. The information of the construction plan also includes the pre-obtained construction arrangements for each construction slot section, the identification information of the construction slot section, and the target liquid level curve.

[0147] During the construction preparation stage, the diaphragm wall is divided into multiple construction slot sections, and the corresponding identification information (numbers) are 1# construction slot section, 2# construction slot section... N# construction slot section, and the construction is carried out in the order of the number arrangement.

[0148] The shape of the cast-in-place guide wall is selected as "┐┌". The construction of the guide wall satisfies that the bottom penetrates more than 150 mm below the bottom elevation of the capping beam. Connect and penetrate the road surface steel bars and the guide wall steel bars. During construction, key inspections are carried out on the deviation of the center line, verticality and the inner net width of the guide wall.

[0149] The well point pre-drainage is temporary precipitation. The precipitation wells only operate during the construction of the diaphragm wall, and the operation time is short. Arranging precipitation wells separately will result in too high cost, which is not conducive to project management from the perspective of reducing project cost. Therefore, dewatering wells in the pit are constructed in advance. After completion, the dewatering wells are first used as precipitation wells. If the dewatering wells cannot meet the precipitation requirements, an appropriate number of precipitation wells are set outside the foundation pit. According to the construction position of the diaphragm wall, a certain number of precipitation wells at nearby corresponding positions are selected for pumping. To prevent impacts on the surrounding area, the precipitation wells in other areas should stop pumping construction, and the hydraulic gradient outside the foundation pit is reduced by means of sectional and zonal precipitation.

[0150] During the construction of the diaphragm wall trench, ensure that the groundwater level is lower than the mud liquid level. Use an automated remote control device to supplement the mud in a timely manner. The mud supplement should be carried out when the grab bucket is lifted 2 m above the ground; the speed should be kept balanced when the groover lowers and lifts the bucket. For the construction of different deep trench sections, the shallower trench sections should be constructed last, and the deeper trench sections should be excavated first. The excavation depth at the bottom of the same trench section should be kept consistent. After the construction of the same trench section is completed, ultrasonic detection is carried out. During construction, control the concrete at the joint to be dense without mud inclusion. Use an extra-long vibrating rod for concrete vibration, and increase the vibration ratio at the interface of the I-beam of the diaphragm wall. The steel reinforcement cage of the diaphragm wall at the position of the silty soil and fine sand layer and 1 m above and below it is wrapped with colored strip cloth, and it is tied and fixed with buckles or thin steel bars every 2 to 3 m. The colored strip cloth is used to separate the soil on both sides of the trench section from the steel bars in the trench, avoiding the phenomenon of diaphragm wall mud inclusion and steel bar exposure caused by trench section shrinkage. The time interval between the concrete pouring time and the steel reinforcement cage installation time should not be too long, with 4 h as the standard. The above construction plan for the diaphragm wall and its corresponding stability calculation method are applicable to the construction of diaphragm walls in deep, water-rich, soft silty fine sand strata, where the groundwater is rich, the water head is high, the soil is soft, the self-stability is poor, and the silty fine sand layer is located in the shallow layer and is deep.

[0151] In another specific application, a method for calculating the stability of a diaphragm wall of a subway station is provided, including:

[0152] S101, obtaining the construction plan of the diaphragm wall;

[0153] S102, pouring concrete for the current construction trench section according to the construction plan, and using a construction acquisition device to measure and obtain the pouring depth data of the current construction trench section and the liquid level data of the precipitation well corresponding to the current construction trench section in real time; the pouring depth data and the liquid level data are respectively used to indicate the changes in the pouring depth and the liquid level of the precipitation well.

[0154] S103. Obtain the target liquid level data and the predicted liquid level data corresponding to the pouring depth interval where the pouring depth data is located. The predicted liquid level data is obtained according to a preset target liquid level curve. When the difference between the target liquid level data and the predicted liquid level data is within its corresponding preset range, update the target liquid level data according to the predicted liquid level data. When the difference between the target liquid level data and the predicted liquid level data is not within its corresponding preset range, update the target liquid level curve according to the target liquid level data. Calculate the error based on the target liquid level data and the liquid level data, input the error into a preset PID control algorithm to obtain a water pump control signal, and use the water pump control signal to control the operation of the water pump group so that the actual liquid level meets the requirements of the target liquid level (make the water level in the dewatering well lower than the current pouring height of the current construction trench section).

[0155] S104. When the pouring of the current construction trench section is completed, use an ultrasonic detection device to obtain the perpendicularity data of the side wall of the construction trench section. When the perpendicularity data is within its corresponding preset range, input the perpendicularity data, the change curve of the pouring depth data, and the change curve of the liquid level data into a stability scoring model to obtain the stability score of the current construction trench section and send it to the user device. Take the next construction trench section as the current construction trench section and execute S102.

[0156] S105. When the number of stability scores of the construction trench sections for which stability scoring has been completed meets a preset number, obtain the stability calculation result of the diaphragm wall in the previous construction stage according to the stability score, the key coefficient of the trench section, and the stability scoring standard of each construction trench section for which stability scoring has been completed. The stability calculation result includes normal and abnormal.

[0157] S106. When the stability calculation result is abnormal, generate an abnormal prompt message and send it to the user device.

[0158] Among them, the training process of the stability scoring model includes the following steps:

[0159] Obtain a training set. The training set includes multiple training data. Each training data includes perpendicularity data, a pouring depth change curve, and a liquid level change curve, as well as the labeled data of the stability score corresponding to the perpendicularity data, the pouring depth change curve, and the liquid level change curve.

[0160] For each training data in the training set, perform the following processing:

[0161] Input the perpendicularity data, the pouring depth change curve, and the liquid level change curve in the training data into a preset deep learning model to obtain the predicted data of the stability score corresponding to the perpendicularity data, the pouring depth change curve, and the liquid level change curve.

[0162] Update the model parameters of the deep learning model based on the predicted data and the labeled data;

[0163] Detect whether a preset training end condition is satisfied; if so, use the trained deep learning model as the stability scoring model; if not, continue to train the deep learning model using the next piece of training data.

[0164] System embodiment.

[0165] This embodiment provides a stability calculation system for the diaphragm wall of a subway station. For the specific working process and beneficial effects of the stability calculation system, reference can be made to the corresponding processes and beneficial effects in the foregoing method embodiments, which will not be elaborated herein.

[0166] The stability calculation system includes a controller, a construction data acquisition device and an ultrasonic detection device connected to the controller. The construction data acquisition device includes a pressure sensor disposed on a conduit of a construction section of the diaphragm wall and a liquid level sensor disposed in a dewatering well of the diaphragm wall; the controller is configured to:

[0167] S101, obtain the construction plan of the diaphragm wall;

[0168] S102, perform concrete pouring for the current construction section according to the construction plan, and use the construction data acquisition device to measure and obtain the pouring depth data of the current construction section and the liquid level data of the dewatering well corresponding to the current construction section in real time; the pouring depth data and the liquid level data are respectively used to indicate the changes in the pouring depth and the liquid level of the dewatering well.

[0169] S104, when the pouring of the current construction section is completed, use the ultrasonic detection device to obtain the perpendicularity data of the side wall of the construction section. When the perpendicularity data is within its corresponding preset range, input the perpendicularity data, the change curve of the pouring depth data, and the change curve of the liquid level data into the stability scoring model to obtain the stability score of the current construction section and send it to the user device. Take the next construction section as the current construction section and execute S102.

[0170] In an exemplary embodiment, the controller is further configured to:

[0171] S103, obtain a water pump control signal according to the pouring depth data and the liquid level data, where the water pump control signal is used to control the water pump group in the dewatering well to pump water, so that the water level in the dewatering well is lower than the current pouring height of the current construction section.

[0172] In an exemplary embodiment, the controller obtains the water pump control signal in the following manner:

[0173] Obtain the target liquid level data corresponding to the pouring depth interval where the pouring depth data is located;

[0174] According to the target liquid level data and the liquid level data, generate a water pump control signal in a PID manner to control the operation of the water pump group so that the actual liquid level meets the requirements of the target liquid level.

[0175] In an exemplary embodiment, the controller generates a water pump control signal in a PID manner according to the target liquid level data and the liquid level data in the following way:

[0176] Obtain the target liquid level data and the predicted liquid level data corresponding to the pouring depth interval where the pouring depth data is located, and the predicted liquid level data is obtained according to a preset target liquid level curve;

[0177] When the difference between the target liquid level data and the predicted liquid level data is within its corresponding preset range, update the target liquid level data according to the predicted liquid level data;

[0178] When the difference between the target liquid level data and the predicted liquid level data is not within its corresponding preset range, update the target liquid level curve according to the target liquid level data;

[0179] Calculate the error according to the target liquid level data and the liquid level data, and input the error into a preset PID control algorithm to obtain a water pump control signal.

[0180] In an exemplary embodiment, the stability scoring model is obtained by training a deep learning model, and the training process of the stability scoring model includes the following steps:

[0181] Obtain a training set, where the training set includes a plurality of training data, and each training data includes verticality data, a pouring depth change curve, and a liquid level change curve, as well as labeled data of the stability score corresponding to the verticality data, the pouring depth change curve, and the liquid level change curve;

[0182] For each training data in the training set, perform the following processing:

[0183] Input the verticality data, the pouring depth change curve, and the liquid level change curve in the training data into a preset deep learning model to obtain predicted data of the stability score corresponding to the verticality data, the pouring depth change curve, and the liquid level change curve;

[0184] Update the model parameters of the deep learning model based on the predicted data and the labeled data;

[0185] Check whether the preset training end condition is met; if so, use the trained deep learning model as the stability scoring model; if not, continue to train the deep learning model using the next training data.

[0186] In an exemplary embodiment, the controller is further configured to:

[0187] When the number of stability scores of the construction trench sections that have completed stability scoring meets the preset number, the stability calculation results of the underground continuous wall in the previous construction stage are obtained according to the stability score of each construction trench section that has completed stability scoring, the trench section key coefficient and the stability scoring standard, and the stability calculation results include normal and abnormal results;

[0188] When the stability calculation result is abnormal, abnormal prompt information is generated and sent to the user device.

[0189] In an exemplary embodiment, a plurality of drainage wells are arranged in the foundation pit of the underground continuous wall, and each drainage well is used as a precipitation well for drainage before and during concrete pouring to lower the groundwater level, so that the moisture in the soil above the water level slope line seeps down, the fluidity of the soil is reduced, and the sandy soil layer is consolidated, thereby reducing the lateral pressure of the groundwater on the trench wall.

[0190] In an exemplary embodiment, the construction plan of the underground continuous wall also includes: wrapping the soft soil layer position on the steel cage of the underground continuous wall and within a range of 3m above and below, so as to isolate the contact between the soil and the steel bars on both sides of the construction trench section, and to prevent the trench wall soil caused by the shrinkage of the construction trench section from invading the steel cage or leaning on the steel cage, causing mud inclusion and leakage of steel bars in the underground continuous wall.

[0191] In an exemplary embodiment, the conduit refers to the perfusion conduit of the current construction slot section.

[0192] In the various implementation modes of the specification of the present application, the size of the serial number of each process does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the present application.

[0193] In the description, claims and drawings of the present application, the terms "first", "second", "third", etc. (if any) are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments of the present application described herein can be implemented in an order different from those illustrated or described herein. In addition, the terms "comprising" and "corresponding to" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that comprises a series of steps or units need not be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0194] From the perspectives of the purpose of use, efficacy, progress and novelty, the present application is described and has met the requirements of enhanced function and use emphasized by the patent law. The above description and the accompanying drawings of the present application are only preferred embodiments of the present application and do not limit the present application thereto. Therefore, all those that are similar or identical to the structure, device, features, etc. of the present application, that is, all equivalent substitutions or modifications made according to the scope of the patent application of the present application, shall fall within the scope of protection of the patent application of the present application.

Claims

1. A calculation method for the stability of a diaphragm wall of a subway station, characterized in that, The described stability calculation method includes: S101, obtaining the construction plan of the diaphragm wall; S102, performing concrete pouring for the current construction slot according to the construction plan, and using a construction acquisition device to measure and obtain the pouring depth data of the current construction slot and the liquid level data of the dewatering well corresponding to the current construction slot in real time; the pouring depth data and the liquid level data are respectively used to indicate the changes in the pouring depth and the liquid level of the dewatering well; S104, when the pouring of the current construction slot is completed, using an ultrasonic detection device to obtain the perpendicularity data of the side wall of the construction slot. When the perpendicularity data is within its corresponding preset range, input the perpendicularity data, the change curve of the pouring depth data, and the change curve of the liquid level data into the stability scoring model to obtain the stability score of the current construction slot and send it to the user device, take the next construction slot as the current construction slot and execute S102; Among them, the construction acquisition device includes a pressure sensor arranged on the conduit of the construction slot of the diaphragm wall and a liquid level sensor arranged on the dewatering well of the diaphragm wall.

2. The stability calculation method of the diaphragm wall of a subway station according to claim 1, characterized in that The stability scoring model is obtained by training a deep learning model. The training process of the stability scoring model includes the following steps: Obtaining a training set, the training set includes a plurality of training data, and each training data includes perpendicularity data, a pouring depth change curve, and a liquid level change curve, as well as labeled data of the stability score corresponding to the perpendicularity data, the pouring depth change curve, and the liquid level change curve; For each training data in the training set, perform the following processing: Input the perpendicularity data, the pouring depth change curve, and the liquid level change curve in the training data into a preset deep learning model to obtain predicted data of the stability score corresponding to the perpendicularity data, the pouring depth change curve, and the liquid level change curve; Based on the predicted data and the labeled data, update the model parameters of the deep learning model; Detect whether the preset training end condition is satisfied; if so, use the trained deep learning model as the stability scoring model; if not, continue to train the deep learning model with the next training data.

3. The stability calculation method of the diaphragm wall of a subway station according to claim 2, characterized in that, The stability calculation method further includes: When the number of stability scores of the construction slots for which stability scores have been completed meets a preset number, obtain the stability calculation result of the diaphragm wall in the previous construction stage according to the stability scores, slot key coefficients, and stability scoring criteria of each construction slot for which stability scores have been completed. The stability calculation result includes normal and abnormal; When the stability calculation result is abnormal, generate an abnormal prompt message and send it to the user device.

4. The stability calculation method of the diaphragm wall of a subway station according to claim 1, characterized in that, The stability calculation method further includes: S103, obtaining a water pump control signal according to the pouring depth data and the liquid level data, and the water pump control signal is used to control the water pump group of the dewatering well to pump water so that the water level of the dewatering well is lower than the current pouring height of the current construction slot.

5. The stability calculation method of the diaphragm wall of a subway station according to claim 4, characterized in that, The method for obtaining the water pump control signal includes: Obtaining the target liquid level data corresponding to the pouring depth interval where the pouring depth data is located; According to the target liquid level data and the liquid level data, a water pump control signal is generated in a PID manner to control the operation of the water pump group so that the actual liquid level meets the target liquid level requirement.

6. The stability calculation method of the diaphragm wall of a subway station according to claim 5, characterized in that, The step of generating a water pump control signal in a PID manner according to the target liquid level data and the liquid level data comprises: Obtaining target liquid level data and predicted liquid level data corresponding to the pouring depth interval where the pouring depth data is located, wherein the predicted liquid level data is obtained according to a preset target liquid level curve; When the difference between the target liquid level data and the predicted liquid level data is within the corresponding preset range, updating the target liquid level data according to the predicted liquid level data; When the difference between the target liquid level data and the predicted liquid level data is not within the corresponding preset range, updating the target liquid level curve according to the target liquid level data; The error is calculated based on the target liquid level data and the liquid level data, and the error is input into a preset PID control algorithm to obtain a water pump control signal.

7. The stability calculation method of the diaphragm wall of a subway station according to claim 1, characterized in that, A plurality of drainage wells are arranged in the foundation pit of the underground continuous wall, and each drainage well is used as a precipitation well for drainage before and during concrete pouring to lower the groundwater level, so that the moisture in the soil above the water level slope line seeps down, the fluidity of the soil is reduced, and the sandy soil layer is consolidated, thereby reducing the lateral pressure of the groundwater on the trench wall.

8. The stability calculation method of the diaphragm wall of a subway station according to claim 1, characterized in that, The construction plan of the underground continuous wall also includes: wrapping the soft soil layer on the steel cage of the underground continuous wall and within a range of 3m above and below, so as to isolate the contact between the soil and the steel bars on both sides of the construction trench section.

9. The stability calculation method of the diaphragm wall of a subway station according to claim 1, characterized in that, The conduit refers to the perfusion conduit of the current construction slot section.

10. A stability calculation system for a diaphragm wall of a subway station, characterized in that, The stability calculation system includes a controller, a construction collection device and an ultrasonic detection device connected to the controller, wherein the construction collection device includes a pressure sensor of a conduit provided in a construction trench section of the underground continuous wall and a liquid level sensor provided in a dewatering well of the underground continuous wall; the controller is configured as follows: S101, obtaining a construction plan of the underground continuous wall; S102, pouring concrete in the current construction slot section according to the construction plan, using a construction acquisition device to measure and obtain in real time the pouring depth data of the current construction slot section and the liquid level data of the precipitation well corresponding to the current construction slot section; the pouring depth data and the liquid level data are used to indicate changes in the pouring depth and the liquid level of the precipitation well, respectively; S104, when the current construction slot section is cast, the verticality data of the side wall of the construction slot section is obtained by using an ultrasonic detection device. When the verticality data is within the corresponding preset range, the verticality data, the change curve of the casting depth data and the change curve of the liquid level data are input into a stability scoring model to obtain the stability score of the current construction slot section and send it to the user device to set the next construction slot section as the current construction slot section and execute S102.

Images