A method for early warning and monitoring of foundation pit deformation based on multi-parameter variables

Abstract

This invention belongs to the field of foundation pit deformation monitoring technology and discloses a method for early warning and monitoring of foundation pit deformation based on multiple parameters. It addresses the problem that existing methods using sensors to monitor foundation pit deformation cannot provide early warnings or feedback to guide the design of foundation pit support structures. This invention utilizes past foundation pit data and current foundation pit deformation monitoring data, and by adjusting the weights of different variables, constructs a corresponding machine model to predict the deformation of the foundation pit. Compared to the passive sensor-based deformation monitoring methods of existing technologies, this invention provides proactive deformation prediction and early warning. It not only guides the design of foundation pit support structures and facilitates their optimization, but also provides strong guidance on foundation pit excavation and risk points during the excavation process, improving the safety of foundation pit support and excavation.

Description

Technical Field

[0001] This invention belongs to the field of foundation pit deformation monitoring technology, specifically relating to a foundation pit deformation early warning monitoring method based on multiple parameter variables. Background Technology

[0002] The support structure of a building foundation pit and the surrounding ground surface will deform due to the excavation of the pit, posing a serious threat to the safety of surrounding buildings, underground pipelines, structures, and roads. In particular, some deep foundation pit projects are located in urban areas with dense municipal facilities and high population density, so deformation control of deep foundation pits should be strictly controlled.

[0003] Currently, the main method for monitoring the deformation of foundation pits is to use sensors to collect data and transmit the data detected by the sensors to the controller. The controller then decides whether to issue an alarm or warning based on the set values ​​(i.e. whether the data exceeds the set range).

[0004] There are numerous existing technical documents regarding the use of sensors to monitor the deformation of foundation pits. For example, application number 2019106463601 discloses a conventional deformation monitoring system and method for urban building foundation pits, including a foundation pit body and a support structure. The monitoring system includes a monitoring host and several sensors, which are installed in the foundation pit body or the support structure. Each sensor includes a housing, and the housing contains a movable part and a sensing element for sensing the position of the movable part. The sensing element is electrically connected to the monitoring host. The conventional deformation monitoring system and method for urban building foundation pits of the present invention can directly output the angle and direction of the deformation, achieving low complexity and high accuracy.

[0005] For example, application number 2019104831897 discloses a method and system for monitoring foundation pit deformation. It sets up a reference sensor network and multiple monitoring sensor networks. Each sensor node in the reference sensor network transmits two radio waves of different frequencies to the first monitoring sensor network. Other monitoring sensor networks sequentially send their own measurement data and the received measurement data to the subsequent monitoring sensor networks, until the last monitoring sensor network sends all measurement data to the reference sensor network. The server receives all measurement data sent by the reference sensor network and obtains the center point position of each monitoring sensor network based on all the measurement data. The server calculates the deformation of the foundation pit based on the center point positions of each monitoring sensor network obtained multiple times. This technology monitors foundation pit deformation through a sensor network, eliminating the need for manual measurement and having low requirements for installation conditions, thus improving the convenience of measurement.

[0006] For example, application number 2021104700154 discloses a high-precision urban building foundation pit conventional deformation monitoring system and its monitoring method. The foundation pit includes the foundation pit body and the support structure. The monitoring system includes a monitoring host and several sensors. The sensors are installed in the foundation pit body or the support structure. Each sensor includes a housing, and the housing contains a movable part and a sensing element for sensing the position of the movable part. The sensing element is electrically connected to the monitoring host. The housing is tubular and vertically arranged, with both the upper and lower ends of the housing closed. A connecting pipe passes through the upper end of the housing, through which the cable for connecting the sensing element and the monitoring host passes. A lifting plate is slidably arranged inside the housing, and the lifting plate is tightly fitted to the inner wall of the housing. The lifting plate is connected to the lower end of the housing by a spring, and a protective fluid is filled between the lifting plate and the upper end of the housing. This technology can directly output the angle and direction of the deformation, achieving low complexity and high precision.

[0007] For example, application number 2021113986771 discloses a deep soil deformation monitoring device and calculation method for foundation pits. The monitoring device includes a measuring tube and stress sensors. The measuring tube has a certain thickness, and multiple sets of stress sensors are evenly spaced along the length of the measuring tube on its inner wall. Each set of stress sensors includes two stress sensors installed at the same height. The calculation method involves first determining the angle and radius corresponding to the center of curvature of each segment before installation; then determining the direction of soil deformation by analyzing the compressive or tensile stress generated by the stress sensors; and finally, calculating the magnitude of soil deformation using the arc length formula. This technology can visually display the development trend of deep soil deformation and can measure the amount of deep soil deformation in real time.

[0008] Based on the examples of existing technologies mentioned above, current foundation pit deformation monitoring is all based on data collected by sensors to obtain the amount of foundation pit deformation.

[0009] However, the data collected by sensors is merely a result and has little guiding significance for the factors that cause foundation pit deformation. Furthermore, the monitoring data collected by sensors cannot provide early warning and cannot provide good feedback for the design of foundation pit support structures. Summary of the Invention

[0010] To address the problem that existing methods using sensors to monitor foundation pit deformation cannot provide early warnings or feedback to guide the design of foundation pit support structures, this invention provides a foundation pit deformation early warning and monitoring method based on multiple parameters. This method has the function of actively predicting deformation, thus providing good guidance and feedback for the design of foundation pit support structures and the excavation of foundation pits.

[0011] To solve the technical problem, the technical solution adopted by this invention is as follows:

[0012] A method for early warning and monitoring of foundation pit deformation based on multi-parameter variables, characterized by the following steps:

[0013] (1) Collect data information of previous foundation pits and record monitoring data information of the deformation of the foundation pit. The data information of the foundation pit includes, but is not limited to: the shape and size of the foundation pit, the depth of the foundation pit, the geological conditions, the type of support structure of the foundation pit, the support capacity of the support structure, the surrounding environmental conditions, and the safety level of the foundation pit; the monitoring data information includes, but is not limited to: the total deformation and the deformation collected by the sensors at various time points, as well as the location information of the sensor placement.

[0014] (2) Classify the data and monitoring data of the foundation pit according to the geological conditions of the foundation pit and the type of support structure of the foundation pit; and classify them into at least soil nailing wall database, cement-soil gravity retaining wall database, steel-cement-soil mixing wall database, retaining structure database, and underground continuous wall database.

[0015] (3) Construct machine learning models using data from the soil nailing wall database, cement-soil gravity retaining wall database, steel-cement-soil mixing wall database, retaining structure database, and diaphragm wall database respectively; that is, construct a soil nailing wall machine learning model using data from the soil nailing wall database; construct a cement-soil gravity retaining wall machine learning model using data from the cement-soil gravity retaining wall database; construct a steel-cement-soil mixing wall machine learning model using data from the steel-cement-soil mixing wall database; construct a retaining structure machine learning model using data from the retaining structure database; and construct a diaphragm wall machine learning model using data from the diaphragm wall database.

[0016] (4) Collect the training database again, which includes at least the soil nailing wall training database, the cement-soil gravity retaining wall training database, the steel-cement-soil mixing wall training database, the retaining structure training database, and the diaphragm wall training database.

[0017] (5) The data from the soil nailing wall training database, the cement-soil gravity retaining wall training database, the steel-cement-soil mixing wall training database, the retaining structure training database, and the diaphragm wall training database are respectively put into the soil nailing wall-machine learning model, the cement-soil gravity retaining wall-machine learning model, the cement-soil mixing wall-machine learning model, the retaining structure-machine learning model, and the diaphragm wall-machine learning model for training, so as to obtain the soil nailing wall-machine model, the cement-soil gravity retaining wall-machine model, the cement-soil mixing wall-machine model, the retaining structure-machine model, and the diaphragm wall-machine model.

[0018] (6) Input the data information of the newly built foundation pit into the corresponding machine model and output the deformation amount of the foundation pit.

[0019] In some embodiments, when constructing various machine learning models, the size, depth, shape, surrounding environmental conditions, and safety level of the foundation pit are used as variables, and the weights of the variables are continuously adjusted to obtain the optimal machine learning model.

[0020] In some embodiments, groundwater level and temperature information from geological conditions are also included as variables, and their weights are continuously adjusted when constructing the machine learning model. That is to say, the geological conditions in this invention include not only stratigraphic information but also groundwater level and temperature information.

[0021] In some embodiments, soil moisture content in geological conditions is also used as a variable, and the weight of soil moisture content is continuously adjusted when constructing the machine learning model.

[0022] Similarly, when setting up the training database, soil moisture content is also used as a variable to train the machine learning model.

[0023] In some embodiments, the method further includes the following steps: separately constructing machine learning models for the support capacity (e.g., support capacity includes bearing capacity, bending moment resistance, anti-slip capacity, anti-overturning capacity, etc.), soil load (soil lateral pressure and soil pressure), and deformation of the foundation pit's support structure, and training them to obtain a support capacity-load-machine model, wherein the support capacity and soil load are used as variables; inputting the support capacity and soil load corresponding to the new foundation pit's support structure into the support capacity-load-machine model to obtain the corresponding deformation of the foundation pit. The deformation obtained in this step is used to play an auxiliary role. However, if any deformation in this step or in step (6) exceeds a set value, an alarm will be triggered.

[0024] In some embodiments, to ensure the stability of the foundation pit support structure, its support capacity is amplified during the design process (e.g., by setting a safety factor K). Preferably, when constructing the support capacity-load-machine model, the support capacity / safety factor of the support structure is used as a variable to construct the support capacity-load-machine model. That is, the support capacity-load-machine model is constructed using the data after the actual support capacity / safety factor of the support structure and the magnitude of the soil load.

[0025] Compared with the prior art, the present invention has the following beneficial effects:

[0026] The present invention provides a multi-parameter variable-based early warning and monitoring method for foundation pit deformation. This method utilizes existing foundation pit data and current foundation pit deformation monitoring data. By adjusting the weights of different variables, a corresponding machine model is constructed. This machine model is then used to predict the deformation of the foundation pit. Compared to existing technologies that passively monitor deformation using sensors, the deformation prediction and early warning method of this invention is proactive. It not only guides the design of foundation pit support structures and facilitates their optimization, but also provides strong guidance on foundation pit excavation and risk points during the excavation process, thereby improving the safety of foundation pit support and excavation. Attached Figure Description

[0027] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0028] The present invention will be further described below with reference to embodiments. These embodiments are merely some, not all, of the embodiments of the present invention. Other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are all within the protection scope of the present invention.

[0029] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. The terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, unless otherwise explicitly specified and limited, the terms "installed," "connected," and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a machine connection; they can refer to a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0030] Referring to the accompanying drawings, the early warning and monitoring method for foundation pit deformation based on multiple parameters of the present invention includes the following steps:

[0031] (1) Collect data information of previous foundation pits and record monitoring data information of the deformation of the foundation pit. The data information of the foundation pit includes, but is not limited to: the shape and size of the foundation pit, the depth of the foundation pit, the geological conditions, the type of support structure of the foundation pit, the support capacity of the support structure, the surrounding environmental conditions, and the safety level of the foundation pit; the monitoring data information includes, but is not limited to: the total deformation and the deformation collected by the sensors at various time points, as well as the location information of the sensor placement.

[0032] The data information for the foundation pit includes the supporting capacity of its support structure (e.g., bearing capacity, bending moment resistance, anti-slip capacity, anti-overturning capacity, etc.). The monitoring data also includes the lateral soil pressure and soil pressure of the foundation pit. These lateral soil pressure and soil pressure include both the lateral soil pressure and soil pressure when the foundation pit is not deformed, and the lateral soil pressure and soil pressure (vertical pressure) corresponding to various deformation amounts. For example, when the foundation pit deforms by x mm, the lateral soil pressure is F0. x The soil pressure is F x When the excavation pit deforms to y mm, the lateral pressure on the soil is F0. y The soil pressure is F y This is to facilitate the assessment of soil pressure and lateral soil pressure under different soil deformations.

[0033] (2) The data and monitoring data of the foundation pit are classified according to the geological conditions and the type of support structure of the foundation pit; and are at least divided into soil nailing wall database, cement-soil gravity retaining wall database, steel-cement-soil mixing wall database, retaining structure database, and diaphragm wall database. The soil nailing wall database contains data on foundation pits with soil nailing walls as the support structure, as well as monitoring data recording the deformation of the foundation pit. The cement-soil gravity retaining wall database contains data on foundation pits with cement-soil gravity retaining walls as the support structure, as well as monitoring data recording the deformation of the foundation pit. The steel-cement-soil mixing wall database contains data on foundation pits with steel-cement-soil mixing walls as the support structure, as well as monitoring data recording the deformation of the foundation pit. The retaining structure database contains data on foundation pits with retaining structures as the support structure, as well as monitoring data recording the deformation of the foundation pit. The diaphragm wall database contains data on foundation pits with diaphragm walls as the support structure, as well as monitoring data recording the deformation of the foundation pit.

[0034] In other words, each database contains data information of each foundation pit under that foundation pit support structure type and corresponding monitoring data information of each foundation pit.

[0035] Since the type of support structure for a foundation pit is generally closely related to geological conditions, the geological conditions and the type of support structure for the foundation pit can be integrated into one.

[0036] Common types of support structures for foundation pits include: slope protection, soil nailing walls, cement-soil gravity retaining walls, steel-cement-soil mixing walls, retaining structures, and diaphragm walls.

[0037] Slope excavation is suitable for cohesive soils, silty clay, silty soils, silt, and silty sand, but not for silt, fill soil, or recently loose soil. Slope excavation is suitable for shallow foundation pits with an excavation depth of no more than 7m. In areas with better soil conditions, the excavation depth for slope excavation can be appropriately increased.

[0038] Soil nailing walls are suitable for foundation pits in cohesive soils and weakly cemented sandy soils above the groundwater level. If encountering water-rich sandy soils, gravelly sands, or pebble layers, a water-stop curtain is required. They are not suitable for silt, backfill, or recently filled loose soil. Soil nailing walls are suitable for shallow foundation pits in soft soil areas with an excavation depth not exceeding 5m. In areas with better soil quality, the excavation depth can be appropriately increased, but the maximum excavation depth should not exceed 12m.

[0039] Cement-soil gravity retaining walls are suitable for almost all soil conditions, but when encountering silt, backfill, or thick loose fill, the cement content of the mixing piles should be appropriately increased. Under normal circumstances, the excavation depth should not exceed 7m when there are no environmental protection requirements, and should not exceed 5m when there are environmental protection requirements.

[0040] Steel-cement-soil mixing walls are suitable for almost all soil conditions, but when encountering silt, backfill, or thick loose fill, the cement content of the mixing piles should be appropriately increased. They are generally suitable for foundation pits with an excavation depth of no more than 15 cm in soft soil areas; in non-soft soil areas, the excavation depth can be appropriately increased.

[0041] Retaining structures are suitable for all soil conditions. In soft soil areas, the excavation depth should not exceed 20m for deep foundation pits, while in non-soft soil areas, the excavation depth can be appropriately increased. Diaphragm walls are suitable for all soil conditions.

[0042] Therefore, it is preferable to integrate the geological conditions of the foundation pit with the type of support structure. That is to say, under normal circumstances, the type of support structure of the foundation pit (such as slope protection, soil nailing wall, cement-soil gravity retaining wall, steel-cement-soil mixing wall, pile wall, diaphragm wall, etc.) is determined when the geological conditions are determined.

[0043] In the specific real-time process, the slope protection method for foundation pit support is relatively simple compared to the design of other foundation pit support structures. Therefore, if the foundation pit adopts the slope protection method, the data information of the foundation pit and the monitoring data information recording the deformation of the foundation pit can be deleted.

[0044] (3) Construct machine learning models using data from the soil nailing wall database, cement-soil gravity retaining wall database, steel-cement-soil mixing wall database, retaining structure database, and diaphragm wall database respectively; that is, construct a soil nailing wall machine learning model using data from the soil nailing wall database; construct a cement-soil gravity retaining wall machine learning model using data from the cement-soil gravity retaining wall database; construct a steel-cement-soil mixing wall machine learning model using data from the steel-cement-soil mixing wall database; construct a retaining structure machine learning model using data from the retaining structure database; and construct a diaphragm wall machine learning model using data from the diaphragm wall database.

[0045] (4) Collect the training database again, which includes at least the soil nailing wall training database, the cement-soil gravity retaining wall training database, the steel-cement-soil mixing wall training database, the retaining structure training database, and the diaphragm wall training database.

[0046] Specifically, regarding the training database, during the acquisition process, relevant units can be collaborated to obtain data information on newly constructed foundation pits and monitoring data on the deformation of the foundation pits. This facilitates the acquisition of a large training database to train the constructed machine learning model, thereby improving the training effect of the machine learning model.

[0047] (5) The data from the soil nailing wall training database, the cement-soil gravity retaining wall training database, the steel-cement-soil mixing wall training database, the retaining structure training database, and the diaphragm wall training database are respectively put into the soil nailing wall-machine learning model, the cement-soil gravity retaining wall-machine learning model, the cement-soil mixing wall-machine learning model, the retaining structure-machine learning model, and the diaphragm wall-machine learning model for training, so as to obtain the soil nailing wall-machine model, the cement-soil gravity retaining wall-machine model, the cement-soil mixing wall-machine model, the retaining structure-machine model, and the diaphragm wall-machine model.

[0048] (6) Input the data of the newly built foundation pit into the corresponding machine model and output the deformation of the foundation pit. Specifically, for different support structures, the data of the new foundation pit is input into the corresponding machine model for calculation.

[0049] In some embodiments, when constructing various machine learning models (soil nailing wall - machine learning model, cement-soil gravity retaining wall - machine learning model, cement-soil mixing wall - machine learning model, retaining structure - machine learning model, diaphragm wall - machine learning model), the size of the foundation pit, the depth of the foundation pit, the shape of the foundation pit, the surrounding environmental conditions, and the safety level of the foundation pit are used as variables, and the weights of the variables are continuously adjusted to obtain the best machine learning model.

[0050] Similarly, when setting up the training database, the size, depth, shape, surrounding environmental conditions, and safety level of the foundation pit are used as variables to train the machine learning model.

[0051] In some embodiments, groundwater level and temperature information in the geological conditions are also used as variables, and the weights of groundwater level and temperature are continuously adjusted when constructing the machine learning model. That is to say, the geological conditions in this invention not only include stratum information (such as sand layers, clay layers, pebble layers, etc.), but also groundwater level and temperature information.

[0052] Similarly, when setting up the training database, groundwater level information and temperature information are used as variables to train the machine learning model.

[0053] In some embodiments, soil moisture content in geological conditions is also included as a variable, and its weight is continuously adjusted when constructing the machine learning model. That is to say, the geological conditions in this invention also include soil moisture content.

[0054] Similarly, when setting up the training database, soil moisture content is also used as a variable to train the machine learning model.

[0055] In some embodiments, the method further includes the following steps: separately constructing machine learning models for the support capacity (e.g., support capacity includes bearing capacity, bending moment resistance, anti-slip capacity, anti-overturning capacity, etc.), soil load (soil lateral pressure and soil pressure), and deformation of the foundation pit support structure, and training them to obtain a support capacity-load-machine model, wherein the support capacity and soil load are used as variables; inputting the support capacity and soil load corresponding to the support structure of the new foundation pit into the support capacity-load-machine model to obtain the corresponding deformation of the foundation pit. The deformation obtained in this step is used to play an auxiliary role. However, if any deformation in this step or any deformation obtained in step (6) exceeds a set value (e.g., the total deformation does not exceed 30 mm, and the deformation in any 24-hour period does not exceed 2 mm), an alarm will be triggered.

[0056] The setting value for the deformation amount is clear and understandable to those skilled in the art, and will not be elaborated here.

[0057] Since the support capacity is closely related to the size, depth, shape, and geological conditions of the foundation pit, a support capacity-load-machine model is constructed to intuitively obtain the relationship between the support capacity of the support structure, soil load, and deformation. This facilitates the design of support structures for new foundation pits.

[0058] In the actual real-time process, to ensure the stability of the foundation pit support structure, its support capacity was amplified during the design process (e.g., the safety factor K was set). Preferably, when constructing the support capacity-load-machine model, the support capacity / safety factor of the support structure was used as a variable to construct the support capacity-load-machine model. That is to say, the support capacity-load-machine model was constructed using the data after the actual support capacity / safety factor of the support structure and the magnitude of the soil load.

[0059] Because the support capacity of the foundation pit support structure is relatively large, the data changes drastically when expanded according to the safety factor K. Therefore, in order to eliminate the influence of the safety factor K and to facilitate the construction of the machine model, the support capacity is used as a variable to eliminate the influence of the safety factor K.

[0060] The present invention provides a multi-parameter variable-based early warning and monitoring method for foundation pit deformation. This method utilizes existing foundation pit data and current foundation pit deformation monitoring data. By adjusting the weights of different variables, a corresponding machine model is constructed. This machine model is then used to predict the deformation of the foundation pit. Compared to existing technologies that passively monitor deformation using sensors, the deformation prediction and early warning method of this invention is proactive. It not only guides the design of foundation pit support structures and facilitates their optimization, but also provides strong guidance on foundation pit excavation and risk points during the excavation process, thereby improving the safety of foundation pit support and excavation.

[0061] Furthermore, this invention can guide the determination of the installation location of sensors used for subsequent foundation pit monitoring, thereby making the placement of sensors for subsequent data collection more accurate and improving the accuracy of subsequent deformation monitoring.

Claims

1. A method for early warning and monitoring of foundation pit deformation based on multi-parameter variables, characterized in that, Includes the following steps: (1) Collect data information of previous foundation pits and record monitoring data information of the deformation of the foundation pit. The data information of the foundation pit includes: the shape and size of the foundation pit, the depth of the foundation pit, the geological conditions, the type of support structure of the foundation pit, the support capacity of the support structure, the surrounding environmental conditions, and the safety level of the foundation pit; the monitoring data information includes: the total deformation and the deformation collected by the sensor at each time point, as well as the location information of the sensor. (2) Classify the data and monitoring data of the foundation pit according to the geological conditions of the foundation pit and the type of support structure of the foundation pit; and classify them into at least soil nailing wall database, cement-soil gravity retaining wall database, steel-cement-soil mixing wall database, retaining structure database, and underground continuous wall database. (3) Construct machine learning models using data from the soil nailing wall database, cement-soil gravity retaining wall database, steel-cement-soil mixing wall database, retaining structure database, and diaphragm wall database respectively; that is, construct a soil nailing wall-machine learning model using data from the soil nailing wall database; construct a cement-soil gravity retaining wall-machine learning model using data from the cement-soil gravity retaining wall database; construct a steel-cement-soil mixing wall-machine learning model using data from the steel-cement-soil mixing wall database; construct a retaining structure-machine learning model using data from the retaining structure database; and construct a diaphragm wall-machine learning model using data from the diaphragm wall database. (4) Collect the training database again, which includes at least the soil nailing wall training database, the cement-soil gravity retaining wall training database, the steel-cement-soil mixing wall training database, the retaining structure training database, and the underground continuous wall training database. (5) The data from the soil nailing wall training database, the cement-soil gravity retaining wall training database, the steel-cement-soil mixing wall training database, the retaining structure training database, and the diaphragm wall training database are respectively put into the soil nailing wall-machine learning model, the cement-soil gravity retaining wall-machine learning model, the cement-soil mixing wall-machine learning model, the retaining structure-machine learning model, and the diaphragm wall-machine learning model for training, so as to obtain the soil nailing wall-machine model, the cement-soil gravity retaining wall-machine model, the cement-soil mixing wall-machine model, the retaining structure-machine model, and the diaphragm wall-machine model; (6) Input the data information of the newly built foundation pit into the corresponding machine model and output the deformation amount of the foundation pit; it also includes building a machine learning model for the support capacity, soil load and deformation amount of the support structure of the foundation pit separately and training it to obtain the support capacity-load-machine model, wherein the support capacity and soil load are variables; input the support capacity and soil load of the support structure of the new foundation pit into the support capacity-load-machine model to obtain the deformation amount of the foundation pit; when building the support capacity-load-machine model, the support capacity / safety factor of the support structure is used as a variable to build the support capacity-load-machine model.

2. The method for early warning and monitoring of foundation pit deformation based on multiple parameter variables according to claim 1, characterized in that, When constructing various machine learning models, the size, depth, shape, surrounding environmental conditions, and safety level of the foundation pit are used as variables, and the weights of the variables are continuously adjusted to obtain the best machine learning model.

3. The method for early warning and monitoring of foundation pit deformation based on multiple parameter variables according to claim 1, characterized in that, Groundwater level and temperature information from geological conditions are also used as variables, and their weights are continuously adjusted when building the machine learning model.

4. The method for early warning and monitoring of foundation pit deformation based on multiple parameter variables according to claim 1, characterized in that, Soil moisture content in geological conditions is also treated as a variable, and its weight is continuously adjusted when constructing the machine learning model.

Images