Water-rich soft soil layer large rectangular surface pipe jacking working well control system and method thereof
Patent Information
- Application Number
- CN202610823053.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2026-06-09
- Publication Date
- 2026-09-15
- Estimated Expiration
- 2046-06-09
AI Technical Summary
[0005]本发明针对现有技术中缺乏背土效应实时感知手段、控制依赖人工经验导致响应滞后与干预盲目、缺乏分级防控机制的技术问题,提供一种富水软土地层大矩形面顶管工作井控制系统及其方法
相较于现有技术,本发明首先在顶管机壳体和后续管节顶部以及工作井基坑侧壁支护结构上布设管土接触压力与土体位移传感器,构建背土效应实时感知网络,解决了传统室内模拟无法在真实施工中获取动态数据的问题。其次,基于传感数据判别背土效应所处阶段及工作井基坑侧壁支护的受影响程度,当处于竖向扩展阶段时,自动定位注浆干预区域执行增强注浆,实现了从被动经验判断到主动精准注浆的转变。再次,引入分级预警机制:发展指数超过第一阈值时启动机械阻隔装置,阻隔板伸出切断背土效应竖向扩展通道;超过第二阈值时生成预警并调整顶进参数。本发明通过“感知-判别-注浆-阻隔-预警”五级联动控制,形成了针对背土效应与基坑侧壁支护协同安全的全链条主动防控体系,满足了工程对高防渗、高安全的施工要求,保障了工作井周边土体稳定与基坑侧壁支护结构安全。
Smart Images

Figure CN122359571B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of foundation pit construction technology, specifically to a control system and method for a large rectangular jacking well in water-rich soft soil strata. Background Technology
[0002] In the construction of pipe jacking projects such as water conveyance corridors, river underpasses, and flood control and drainage pipe corridors, the back soil effect can easily cause embankment slippage, surface subsidence, and damage to seepage prevention structures. At best, it can lead to instability of the sidewall support structure of the working well foundation pit, and at worst, it can cause major safety and environmental risks such as river leakage and instability of surrounding facilities. Especially in water-rich soft soil strata, the soil has poor self-stability and significant seepage, making the engineering hazards caused by the back soil effect even more prominent.
[0003] Existing technologies largely focus on indoor simulation tests, such as observing soil deformation in scaled-down models using DIC equipment or measuring the range of backsoil effect using multi-point displacement gauges. Both are post-hoc analysis methods and cannot perceive the development of the backsoil effect in real time during actual construction, nor do they have the ability to proactively intervene. In actual engineering projects, the control of the backsoil effect still relies on manual experience for grouting drag reduction, which suffers from shortcomings such as delayed response, blind intervention, insufficient seepage prevention protection, lack of graded control mechanisms, and lack of linkage monitoring for working well support. It is difficult to effectively suppress the evolution of the backsoil effect and cannot meet the construction requirements of high seepage prevention and high safety in engineering projects.
[0004] Therefore, there is an urgent need for a control system and method that can sense, intelligently predict and actively control the backfill effect in real time, and simultaneously ensure the safety of the working well foundation pit. Summary of the Invention
[0005] This invention addresses the technical problems in existing technologies, such as the lack of real-time sensing methods for back soil effect, reliance on manual experience leading to delayed response and blind intervention, and the lack of a hierarchical prevention and control mechanism. It provides a control system and method for a large rectangular jacking well in water-rich soft soil strata.
[0006] The technical solution of the present invention to solve the above-mentioned technical problems is as follows: In a first aspect, the present invention provides a method for controlling a large rectangular surface jacking well in water-rich soft soil strata, comprising: The sensor monitoring data on the casing of the pipe jacking machine, the top of the subsequent pipe sections, and the side wall support structure of the working well pit are loaded. The sensor monitoring data includes pipe-soil contact pressure data and soil displacement data. Based on the soil-to-pipe contact pressure data and the soil displacement data, the stage of the back soil effect and the degree of impact on the sidewall support of the working well pit are determined. Based on the stage of the back soil effect and the degree of impact, the back soil effect development index is calculated. The stage of the back soil effect includes the vertical expansion stage. When the back soil effect is in the vertical expansion stage, the development location of the back soil effect is identified and the grouting intervention area is obtained. The grouting device at the corresponding location is then activated to perform enhanced grouting. If the back soil effect development index exceeds the first warning threshold, the mechanical blocking device at the tail of the pipe jacking machine casing is activated, so that the blocking plate of the mechanical blocking device extends to the top of the pipe section between the overlying soil. If the back soil effect development index exceeds the second warning threshold, a warning signal is generated and the jacking parameters are adjusted. The second warning threshold is greater than the first warning threshold.
[0007] Secondly, this invention provides a control system for a large rectangular surface pipe-jacking well in water-rich soft soil strata, comprising: The data loading module is used to load sensor monitoring data from the casing of the pipe jacking machine, the top of the subsequent pipe sections, and the side wall support structure of the working well pit. The sensor monitoring data includes pipe-soil contact pressure data and soil displacement data. The effect discrimination module is used to determine the stage of the back soil effect and the degree of impact on the sidewall support of the working well pit based on the soil contact pressure data and the soil displacement data. The back soil effect development index is calculated based on the stage of the back soil effect and the degree of impact. The stage of the back soil effect includes the vertical expansion stage. The grouting control module is used to identify the development location of the back soil effect and obtain the grouting intervention area when the back soil effect is in the vertical expansion stage, and to start the grouting device at the corresponding location to perform enhanced grouting. The first execution module is used to activate the mechanical blocking device at the tail of the pipe jacking machine housing if the back soil effect development index exceeds the first warning threshold, so that the blocking plate of the mechanical blocking device extends to the top of the pipe section between the overlying soil. The second execution module is used to generate an early warning signal and adjust the jacking parameters if the back soil effect development index exceeds the second early warning threshold, wherein the second early warning threshold is greater than the first early warning threshold.
[0008] The beneficial effects of this invention are: Compared to existing technologies, this invention firstly deploys soil-to-pipe contact pressure and soil displacement sensors on the casing of the pipe jacking machine, the top of subsequent pipe sections, and the sidewall support structure of the working well pit, constructing a real-time sensing network for the back soil effect. This solves the problem that traditional indoor simulations cannot obtain dynamic data in actual construction. Secondly, based on the sensor data, it determines the stage of the back soil effect and the degree of impact on the sidewall support of the working well pit. When it is in the vertical expansion stage, it automatically locates the grouting intervention area and performs enhanced grouting, realizing the transformation from passive experience-based judgment to proactive and precise grouting. Thirdly, it introduces a graded early warning mechanism: when the development index exceeds the first threshold, a mechanical blocking device is activated, and the blocking plate extends to cut off the vertical expansion channel of the back soil effect; when it exceeds the second threshold, an early warning is generated and the jacking parameters are adjusted. This invention, through a five-level linkage control of "sensing-discrimination-grouting-blocking-early warning," forms a full-chain proactive prevention and control system for the coordinated safety of the back soil effect and the sidewall support of the pit, meeting the engineering requirements for high seepage prevention and high safety, and ensuring the stability of the soil around the working well and the safety of the sidewall support structure of the pit. Attached Figure Description
[0009] Figure 1 A flowchart illustrating a method for controlling a large rectangular surface pipe jacking well in water-rich soft soil strata, as provided by this invention. Figure 2 This is a schematic diagram of the structure of a large rectangular surface pipe jacking well control system for water-rich soft soil strata provided by the present invention.
[0010] In the attached diagram, the components represented by each number are as follows: Data loading module 11, effect discrimination module 12, grouting control module 13, first execution module 14, second execution module 15. Detailed Implementation
[0011] Example 1, as Figure 1 As shown in the figure, this invention provides a method for controlling a large rectangular surface jacking well in water-rich soft soil strata, including: S10: Load the sensor monitoring data on the casing of the pipe jacking machine, the top of the subsequent pipe sections, and the side wall support structure of the working well pit. The sensor monitoring data includes pipe-soil contact pressure data and soil displacement data. During rectangular pipe jacking construction, the pipe jacking machine casing and subsequent pipe sections come into contact and rub against the surrounding soil during jacking. The interaction between the top of the pipe section and the overlying soil is particularly critical for the backsoil effect. Simultaneously, the working shaft, as the core node for both pipe jacking initiation and reception, bears additional loads from the backsoil effect during jacking. When the backsoil effect causes the overlying soil to be dragged and moved, the soil displacement is transmitted through the strata to the perimeter of the working shaft, causing lateral compression or tensile cracking of the pit sidewall support structure. Especially when traversing levees, riverbeds, or water-rich soft soil areas with high groundwater levels, seepage further exacerbates the weakening of lubrication and pressure fluctuations at the pipe-soil contact surface, making the triggering of the backsoil effect more subtle and its development more rapid, and its impact on the working shaft pit sidewall support structure even more significant.
[0012] The backsoil effect refers to a soil movement phenomenon caused by friction between the top of the pipe section and the overlying soil during rectangular pipe jacking construction. Specifically, as the pipe jacking machine advances, the pipe section moves relative to the surrounding soil, generating friction on the contact surface between the top of the pipe section and the overlying soil. As the jacking length increases, the contact area between the pipe section and the soil gradually increases, and the accumulated total frictional resistance also increases. When the total frictional resistance exceeds the shear strength of the overlying soil itself and the restraining capacity of the surrounding soil, the overlying soil will be driven forward by the frictional force at the top of the pipe section, moving forward with the pipe section. This results in overall shear failure of the overlying soil, leading to surface heave or subsidence. Surface deformation can directly damage the impermeable core of embankments, cause lateral seepage in rivers, and even threaten the foundation safety of nearby pumping stations, sluice gates, and other hydraulic structures.
[0013] The hazards of the backsoil effect are mainly manifested in the formation of tensile cracking zones behind the pipe sections when the overlying soil is dragged and moved, leading to surface cracks, subsidence, and even uplift. In severe cases, the overlying soil may undergo overall shear failure, causing surface collapse and threatening construction safety and the surrounding environment. This is especially true in water-rich soft soil strata, such as floodplain soft soil and lacustrine sedimentary soft soil, where the soil has poor self-stability and fluctuating frictional resistance. Furthermore, the near-horizontal top surface of rectangular pipe jacking, with its large contact area with the soil, exacerbates the backsoil effect, posing a serious threat to the stability of the working shaft and the safety of the surrounding environment. Therefore, real-time monitoring of the development of the backsoil effect and proactive control measures are crucial in rectangular pipe jacking construction.
[0014] In order to perceive the development of the back soil effect in real time, it is necessary to install sensing and monitoring devices on the casing of the pipe jacking machine, the top of the subsequent pipe sections, and the side wall support structure of the working well pit, and continuously load the sensing and monitoring data collected by the sensors.
[0015] Specifically, the sensor monitoring data includes two categories: pipe-soil contact pressure data and soil displacement data. Pipe-soil contact pressure data reflects the magnitude and trend of the contact pressure between the top of the pipe section and the overlying soil; when the backsoil effect develops, the pipe-soil contact pressure increases significantly. Soil displacement data reflects the vertical or horizontal displacement of the soil above the pipe section; when the backsoil effect causes shear failure, the soil displacement rate accelerates significantly. By loading these two types of data, real-time and continuous quantitative evidence can be provided for subsequent identification of the stage of the backsoil effect. The starting point of this step is to transform the traditional backsoil effect control, which relies on manual experience, into proactive perception based on real-time sensor data, providing a data foundation for tiered prevention and control.
[0016] Specifically, the sensor monitoring data loaded onto the casing of the pipe jacking machine, the top of subsequent pipe sections, and the sidewall support structure of the working shaft pit includes: The collected values of multiple sets of pipe-soil contact pressure sensors, which are arranged at intervals along the longitudinal direction of the pipe jacking machine casing and along the extension direction of the working well pit sidewall support structure, are added to the pipe-soil contact pressure data. The collected values of multiple sets of soil displacement sensors, which are arranged at intervals along the longitudinal direction of the pipe jacking machine casing and along the extension direction of the working well pit sidewall support structure, are added to the soil displacement data.
[0017] First, the data collected by multiple sets of soil-to-pipe contact pressure sensors spaced at intervals along the longitudinal direction of the pipe jacking machine casing are obtained, and the soil-to-pipe contact pressure data is added. The soil-to-pipe contact pressure sensors are typically earth pressure cells or thin-film pressure sensors, installed on the longitudinal axis of the top of the pipe jacking machine casing and the top of subsequent pipe sections. Multiple sets of sensors are spaced at intervals along the longitudinal direction, with the interval distance set according to the length of the pipe jacking machine and monitoring accuracy requirements, for example, one set every 2 to 5 meters. Each set of sensors can contain multiple measuring points, evenly distributed along the transverse direction of the pipe section. In addition, the data collected by multiple sets of soil-to-pipe contact pressure sensors spaced at intervals along the extension direction of the working shaft pit sidewall support structure are also obtained, and the soil-to-pipe contact pressure data is added. The sensor spacing on the support structure is set according to the size of the support structure and monitoring accuracy requirements, for example, one set every 3 to 6 meters. Two to three measuring points can be arranged vertically along the support structure to monitor the stress characteristics at the top, middle, and bottom of the support structure.
[0018] The soil-to-pipe contact pressure sensors collect values in kilopascals or megapascals, reflecting in real time the contact pressure between the top of the pipe section and the overlying soil, as well as the contact pressure between the support structure and the surrounding soil. When the backsoil effect develops, the soil-to-pipe contact pressure increases significantly, and the area with the fastest pressure increase is usually where the backsoil effect is most concentrated. The collected values from each soil-to-pipe contact pressure sensor are recorded according to time and location, forming a soil-to-pipe contact pressure data sequence. This data sequence is used to determine the stage and development location of the backsoil effect, and to assess the degree of impact on the sidewall support of the working well pit.
[0019] Secondly, the data collected by multiple sets of soil displacement sensors spaced at intervals along the longitudinal direction of the pipe jacking machine casing are added to the soil displacement data. The soil displacement sensors can be vibrating wire displacement gauges, tie rod displacement gauges, or fiber optic displacement sensors, and are deployed on the top of the pipe jacking machine casing and the top of subsequent pipe sections, adjacent to or staggered from the pipe-soil contact pressure sensors. One end of the soil displacement sensor is fixed to the top of the pipe section, and the other end is fixed to the overlying soil through an anchor head. When the soil undergoes vertical or horizontal displacement, the sensor can measure the displacement in real time, in millimeters. Simultaneously, the data collected by multiple sets of soil displacement sensors spaced at intervals along the extension direction of the working shaft pit sidewall support structure are added to the soil displacement data. The soil displacement sensors on the support structure are used to monitor the displacement and deformation of the support structure itself, and their deployment positions correspond to those of the pressure sensors, with multiple measuring points also arranged vertically along the support structure.
[0020] Multiple sets of sensors are deployed at intervals along the longitudinal direction, with the intervals consistent with those of the pressure sensors. When the overlying soil is dragged forward due to the backsoil effect, the soil displacement rate increases significantly, and the displacement continues to increase. The data collected by each soil displacement sensor is recorded according to time and location to form a soil displacement data sequence. This sequence is cross-referenced with the soil-pipe contact pressure data to accurately determine the stage of the backsoil effect and the degree of deformation of the working well pit sidewall support structure affected by the backsoil effect.
[0021] By simultaneously loading the above two sets of data, we can fully understand the soil-to-pipe contact state, soil deformation characteristics, and stress-deformation state of the support structure, providing a complete data foundation for subsequent calculation of the back soil effect development index, identification of development stages, and judgment of the degree of support impact.
[0022] S20: Based on the soil-to-pipe contact pressure data and the soil displacement data, determine the stage of the back soil effect and the degree of impact on the sidewall support of the working well pit. Calculate the back soil effect development index based on the stage of the back soil effect and the degree of impact. The stage of the back soil effect includes the vertical expansion stage. Specifically, after acquiring the pipe-soil contact pressure data and soil displacement data, the real-time sensing data needs to be processed and analyzed to determine the development stage of the back soil effect and simultaneously determine the degree of impact of the back soil effect on the working well foundation pit sidewall support structure. Based on the stage of the back soil effect and the degree of impact on the support, the back soil effect development index is calculated, and the current stage of the back soil effect is determined.
[0023] Specifically, the backfill effect is a process that gradually evolves with the increase of the jacking length. In water-rich soft soil strata, based on the friction between the top of the pipe section and the overlying soil and the soil response characteristics, the development process of the backfill effect can be divided into four sequentially evolving stages, including: Pre-start stage: The frictional force between the top of the pipe section and the overlying soil has not yet exceeded the shear strength of the soil itself, and there is no tendency for relative slippage between the soil and the pipe section. During this stage, the pipe-soil contact pressure is in a stable low range, and the soil displacement is minimal and shows no obvious trend.
[0024] Vertical Expansion Stage: Local soil at the top of the pipe section begins to drag forward with the pipe section, and the disturbance gradually develops vertically from the top of the pipe section upwards into the deeper layers of overlying soil. During this stage, the contact pressure between the pipe and the soil shows a continuous upward trend, but the disturbance has not yet affected the horizontal direction; soil displacement is mainly concentrated in the vertical area above the pipe section. This stage is the critical window period for the backsoil effect to transition from its inception to expansion.
[0025] Horizontal expansion stage: The vertical disturbance has reached the surface or encountered stable strata and cannot continue to expand upwards. The disturbance then spreads horizontally forward and backward. In this stage, the range of increase in soil-to-soil contact pressure expands significantly, the soil displacement rate accelerates, and the disturbed area expands horizontally.
[0026] Overall backing stage: The soil covering the top of the entire pipe section has formed a continuous drag zone, and the ground surface shows obvious uplift or settlement. In this stage, the pipe-soil contact pressure and soil displacement both reach their peak values, and the backing effect has fully manifested.
[0027] The backsoil effect development index is a quantitative indicator that comprehensively reflects the current severity of the backsoil effect and its impact on the safety of the working well pit sidewall support structure. A higher backsoil effect development index indicates a more severe backsoil effect, a stronger tendency for the soil to be dragged and moved, and a greater risk of additional loads and deformation to the pit sidewall support structure. It should be noted that in water-rich soft soil strata, due to high groundwater levels, the effective stress of the soil decreases, shear strength weakens, and seepage accelerates the deterioration of the frictional characteristics of the soil-pipe contact surface. Therefore, the transition of the backsoil effect from the vertical expansion stage to the horizontal expansion stage is usually faster than in conventional strata. Consequently, the identification window for the vertical expansion stage is shorter, requiring higher standards for real-time sensing and rapid response.
[0028] The starting point of this step is that the backfilling effect at different development stages requires differentiated control measures. Accurately identifying the stage of the backfilling effect is a prerequisite for achieving graded and progressive proactive prevention and control. For example, in the vertical expansion stage, the backfilling effect is still within a controllable range and can be effectively suppressed by enhancing grouting. If this window of opportunity is missed and the effect enters the horizontal expansion stage or the overall backfilling stage, higher-level intervention measures such as activating mechanical barrier devices and adjusting jacking parameters are required. By identifying the stage, it is possible to avoid the risk of accidents due to insufficient intervention or the waste of resources due to excessive intervention.
[0029] In addition, before determining the stage of the soil-carrying effect, it is also necessary to train a soil-carrying effect development prediction model: Collect a dataset of back soil effect records from historical rectangular pipe jacking projects, and extract records of pipe-soil friction coefficient, jacking length, overburden pressure, soil displacement rate, and the degree of impact on the working well foundation pit sidewall support and the stage of back soil effect corresponding to each set of records. Using the impact of the working well foundation pit sidewall support and the stage of the back soil effect as supervision, and the recorded data of the pipe-soil friction coefficient, the recorded data of the jacking length, the recorded data of the overburden pressure and the recorded data of the soil displacement rate as input, a back soil effect development prediction model based on physical information neural network is trained. Based on the current rectangular pipe jacking project model, the back soil effect development prediction model is invoked. The pipe-soil friction coefficient, jacking length, overburden pressure, and soil displacement rate under the current construction state are used as input feature variables to obtain the stage of the back soil effect and the degree of impact on the sidewall support of the working well pit.
[0030] Before determining the current stage of the backsoil effect, a predictive model for its development needs to be constructed. Specifically, the evolution of the backsoil effect exhibits significant lag and suddenness. When obvious anomalies occur in the pipe-soil contact pressure or soil displacement, the backsoil effect has often already progressed to a more severe stage, leaving extremely limited response time for control measures. Relying solely on threshold triggering based on real-time sensor data is insufficient for proactive early warning and intervention of the backsoil effect. Therefore, a predictive model for the development of the backsoil effect needs to be constructed to establish a mapping relationship between construction parameters and the development state of the backsoil effect, enabling forward-looking predictions of its future development trend.
[0031] Specifically, the function of this backfill effect development prediction model is to establish a mapping relationship between construction parameters and the development state of the backfill effect, thereby achieving a forward-looking prediction of the development trend of the backfill effect. Its construction and training steps include: First, a dataset of backsoil effect records from historical rectangular pipe jacking projects was collected. This dataset comes from multiple completed or constructed rectangular pipe jacking projects, each recording various parameters at different times during the jacking process. Five types of data were extracted from the backsoil effect record dataset: pipe-soil friction coefficient data, reflecting the frictional characteristics between the pipe section and the soil contact surface; jacking length data, reflecting the total distance jacked by the pipe jacking machine; overburden pressure data, reflecting the vertical pressure exerted on the pipe section by the soil above it; and soil displacement rate data, reflecting the displacement change of the overburden soil per unit time. In addition, data corresponding to the above sets of records in time and location, including the degree of impact on the working shaft pit sidewall support and the stage of the backsoil effect, were extracted as target labels for model training. The data on the degree of impact on the working shaft pit sidewall support is a quantitative assessment of the degree of influence of the backsoil effect on the support structure in historical projects, and the data on the stage of the backsoil effect is the category of the backsoil effect determined in historical projects.
[0032] Optionally, the degree of impact on the sidewall support of the working well pit can be determined as follows: the degree of impact is divided into three levels: slight impact, moderate impact, and severe impact. The quantitative judgment criteria for each level are as follows: Slight impact can be defined as an increase in the pipe-soil contact pressure on the support structure of less than 20%, and a displacement deformation rate of less than 1 mm / d. In this case, the support structure is in an elastic working state with a small additional load, and the quantitative value is 0.2. Moderate impact can be defined as an increase in the pipe-soil contact pressure on the support structure of between 20% and 50%, or a displacement deformation rate of between 1 mm / d and 3 mm / d. In this case, the additional load borne by the support structure is significantly increased, and the quantitative value is 0.6. Severe impact can be defined as an increase in the pipe-soil contact pressure on the support structure of greater than 50%, or a displacement deformation rate of greater than 3 mm / d. In this case, the support structure is already bearing a large additional load, and the deformation rate exceeds the normal range, and the quantitative value is 1.0.
[0033] The calculation method for the increase in contact pressure between the pipe and the soil is as follows: the difference between the measured contact pressure value at the current moment and the reference contact pressure value in the initial state of the support structure before pipe jacking, divided by the reference contact pressure value, and then multiplied by 100%. The calculation method for the displacement deformation rate is as follows: the difference in displacement between two adjacent sampling moments divided by the sampling time interval. If the levels corresponding to the two indicators, contact pressure increase and displacement deformation rate, are inconsistent, the higher level is taken as the final degree of support impact, and the corresponding quantitative value is adopted.
[0034] The stage of the backsoil effect can be determined by dividing it into four progressively evolving stages: the initial stage, the vertical expansion stage, the horizontal expansion stage, and the overall backsoil stage. The initial stage can be defined as when the friction between the top of the pipe section and the overlying soil has not yet exceeded the soil's shear strength, the pipe-soil contact pressure is in a stable low range, and the soil displacement is minimal and shows no significant trend. The quantification value for this stage is 0.1. The vertical expansion stage refers to the stage where local soil at the top of the pipe section begins to drag forward with the pipe section, and the disturbance gradually develops vertically into the deeper layers of the overlying soil. The pipe-soil contact pressure shows a continuous upward trend, and the soil displacement is mainly concentrated in the vertical range above the pipe section. This stage is the critical window period for the backsoil effect to transition from initiation to expansion, and the quantification value is 0.4. The horizontal expansion stage refers to the stage where the vertical disturbance has developed to the surface or encountered stable strata and cannot continue to expand upwards. The disturbance then spreads horizontally to the front and back sides, the range of increase in pipe-soil contact pressure significantly expands, and the soil displacement rate accelerates. The quantification value is 0.7. The overall backing soil stage refers to the stage where the soil covering the top of the entire pipe section has formed a continuous drag zone, and the ground surface shows obvious uplift or settlement. The pipe-soil contact pressure and soil displacement have both reached their peak values, with a quantitative value of 0.9. The determination of each stage is verified and confirmed by engineering technicians based on actual monitoring data, and is recorded as a label for the stage of the backing soil effect. The corresponding quantitative values are then used in model training.
[0035] Furthermore, using the recorded data on the degree of impact on the sidewall support of the working well foundation pit and the recorded data on the stage of the back soil effect as supervision, and the recorded data on the pipe-soil friction coefficient, jacking length, overburden pressure and soil displacement rate as input, a back soil effect development prediction model based on physical information neural network is trained.
[0036] Physical information neural networks (PINs) are machine learning models that incorporate physical equations as constraints into neural network training. Their advantage lies in their ability to learn mapping relationships that conform to physical laws under limited data samples. In this approach, the loss function of the PSN includes not only the data fitting error between predicted and true values but also constraints from physical governing equations governing the development of the backsoil effect, such as the soil-to-pipe friction equilibrium equation and the soil displacement continuity equation. This allows the model to learn both data features and physical laws simultaneously during training, thereby improving prediction accuracy and generalization ability. After training, the model can establish a nonlinear mapping relationship from input feature variables to the backsoil effect development index and the stage of the backsoil effect.
[0037] For example, the back soil effect development prediction model adopts a physical information neural network architecture, which consists of an input layer, multiple fully connected hidden layers, an output layer, and a physical constraint loss term.
[0038] The input layer has four nodes, corresponding to four input feature variables: pipe-soil friction coefficient, jacking length, overburden pressure, and soil displacement rate. Each input feature variable needs to be standardized before input, mapping its numerical range to the interval 0 to 1 to eliminate the influence of dimensional differences on model training. For example, the pipe-soil friction coefficient ranges from 0.2 to 0.8, the jacking length from 0 to 100 meters, the overburden pressure from 0 to 200 kPa, and the soil displacement rate from 0 to 50 mm / min.
[0039] The hidden layers employ a 5-layer fully connected structure, with 128 neurons per layer. Each neuron uses the hyperbolic tangent activation function to introduce nonlinear transformation capabilities. The output value of the hyperbolic tangent activation function ranges from -1 to +1, exhibiting zero-centrosymmetry, which is beneficial for the stability of model training. Each fully connected layer is followed by a batch normalization layer to accelerate model convergence and suppress overfitting. Between the last hidden layer and the output layer, a physical constraint layer is introduced to calculate the physical equation residuals under the current input feature variables.
[0040] The output layer has two branches: the first branch outputs the degree of impact on the sidewall support of the working well pit. This branch uses a linear activation function, with output values ranging from 0 to 1. An output value of 0.2 corresponds to a slight impact, 0.6 to a moderate impact, and 1.0 to a severe impact. Values in between indicate a transitional state between adjacent levels. The second branch outputs the probability distribution of the backfill effect stage. This branch uses a Softmax activation function and outputs the probability values of four nodes, corresponding to the probabilities of the non-starting stage, the vertical expansion stage, the horizontal expansion stage, and the overall backfill stage, respectively. The sum of the four probability values is 1, and the stage corresponding to the node with the highest probability is taken as the stage of the backfill effect output by the model.
[0041] The physical constraint loss term is constructed based on the soil-to-pipe friction equilibrium equation. According to soil mechanics principles, the frictional force per unit length at the top of the pipe section equals the soil-to-pipe friction coefficient multiplied by the effective vertical stress at the top of the pipe section. This frictional force must be balanced with the shear strength of the soil. The physical constraint loss term is defined as the sum of squared deviations between the frictional force corresponding to the backsoil effect development index predicted by the model and the theoretical frictional force calculated by the physical equation. This deviation is added as an additional loss term to the total loss function to guide the model to learn a mapping relationship that conforms to physical laws.
[0042] During training, the Adam optimizer was used, with an initial learning rate of 0.001. The learning rate was decayed by multiplying by 0.9 every 50 training epochs. The number of training epochs was set to 500, and the batch size was 32. The total loss function consisted of two parts: data fitting loss and physical constraint loss. The data fitting loss used mean squared error to measure the deviation between the model output and the true label, while the physical constraint loss used the aforementioned friction balance deviation. For example, the weight of the data fitting loss was set to 0.7, and the weight of the physical constraint loss was set to 0.3. The weighted sum of the two yielded the total loss value. Training was stopped when the total loss value on the validation set no longer decreased for 20 consecutive epochs, resulting in a completed backsoil effect development prediction model.
[0043] Finally, the backfill effect prediction model is invoked based on the current rectangular pipe jacking project model. Different pipe jacking machine models have different dimensions, weights, and structural characteristics, resulting in varying impacts on the backfill effect. Therefore, it is necessary to invoke the corresponding prediction model from the pre-trained backfill effect prediction model library according to the pipe jacking machine model used in the current construction. The backfill effect prediction model library is constructed as follows: For each common pipe jacking machine model, backfill effect record datasets from multiple historical projects of that model are collected, and a backfill effect prediction model corresponding to that model is independently trained using the aforementioned method. All trained models are categorized and stored according to the pipe jacking machine model, forming the backfill effect prediction model library.
[0044] Using the current construction conditions as the soil-to-pipe friction coefficient, jacking length, overburden pressure, and soil displacement rate as input characteristic variables, these variables are input into the back soil effect development prediction model. After forward calculation, the back soil effect development prediction model outputs the stage of the back soil effect and the degree of impact on the working well pit sidewall support.
[0045] The four input characteristic variables are obtained as follows: The pipe-soil friction coefficient is calculated based on the ratio of the pipe-soil contact pressure data to the effective vertical stress at the top of the pipe section. The pipe-soil contact pressure data comes from the pipe-soil contact pressure sensor deployed in step S10, and the effective vertical stress at the top of the pipe section is calculated by subtracting the pore water pressure from the overburden pressure. The jacking length is directly read from the jacking stroke sensor of the pipe jacking machine, which records the total distance jacked by the machine. The overburden pressure is calculated based on the overburden thickness and soil weight above the top of the pipe section. The overburden thickness can be obtained from the construction design documents or on-site measurements, and the soil weight is determined based on the soil layer parameters in the geological survey report. The soil displacement rate is obtained by differentiating the displacement value collected by the soil displacement sensor deployed in step S10 with respect to time. Specifically, it is calculated by dividing the displacement difference between two adjacent sampling times by the sampling time interval.
[0046] In summary, through the training and application of the aforementioned back soil effect development prediction model, the development trend of the back soil effect can be quickly predicted based on real-time construction parameters during the jacking process, and the degree of impact on the sidewall support structure of the working shaft foundation pit can be assessed simultaneously without waiting for significant changes in soil displacement or pressure before making a judgment. This achieves a forward-looking assessment of the back soil effect and support safety, and buys response time for the implementation of subsequent graded prevention and control measures.
[0047] Finally, after obtaining the stage of the backsoil effect and the degree of impact on the working well pit sidewall support, the backsoil effect development index is calculated using the following formula: The backsoil effect development index equals the quantitative value of the backsoil effect stage multiplied by the first weighting coefficient, plus the quantitative value of the degree of impact on the working well pit sidewall support multiplied by the second weighting coefficient. The first weighting coefficient reflects the contribution of the backsoil effect itself to the comprehensive index, and the second weighting coefficient reflects the contribution of the degree of impact on the support to the comprehensive index. For example, based on the correlation characteristics between the backsoil effect and support safety in pipe jacking construction, the first weighting coefficient can be set to 0.7, and the second weighting coefficient can be set to 0.3. The backsoil effect development index ranges from 0 to 1; the higher the index value, the more severe the backsoil effect, and the greater the additional load and deformation risk on the pit sidewall support structure.
[0048] S30: When the back soil effect is in the vertical expansion stage, identify the development position of the back soil effect and obtain the grouting intervention area, and start the grouting device at the corresponding position to perform enhanced grouting; Specifically, when the backfilling effect development prediction model determines that the current stage of the backfilling effect is the vertical expansion stage, it indicates that the backfilling effect has entered a critical window period of deep expansion into the overlying soil from its initial stage. In this stage, the disturbance is limited to a localized area at the top of the pipe section and has not yet spread horizontally. At this time, if enhanced grouting intervention is implemented, high-pressure grout can be injected between the top of the pipe section and the overlying soil. This fills the gaps between the soil and the pipe section, forming a lubricating film, thereby reducing the pipe-soil friction coefficient; it also cuts off the friction transmission path, preventing the soil from being dragged forward by the pipe section. By intervening in the vertical expansion stage in a timely manner, the backfilling effect can be nipped in the bud, preventing it from developing into a wider and more difficult-to-control horizontal expansion stage or an overall backfilling stage.
[0049] Specifically, the location of the backsoil effect development and the grouting intervention area are identified, including: The location of the sensor with the largest pressure increase in the soil-pipe contact pressure data is selected as the first core development location. The location of the sensor with the largest displacement rate in the soil displacement data is selected as the second core development location; The first core development position and the second core development position are merged and extended forward and backward by a preset distance along the jacking direction to obtain the grouting intervention area.
[0050] First, the location of the sensor with the highest pressure increase in the pipe-soil contact pressure data was selected as the first core development location. During the development of the backsoil effect, the contact pressure between the top of the pipe section and the overlying soil will increase with the increase of the jacking length. The location with the larger the pressure increase indicates that the pipe-soil friction increases more violently and the soil is dragged more strongly. Therefore, this location is one of the areas where the backsoil effect is most concentrated.
[0051] Secondly, the location of the sensor with the highest displacement rate in the soil displacement data was selected as the second core development location. Soil displacement rate reflects the speed at which the overlying soil is dragged and moved; a higher displacement rate indicates more severe soil disturbance at that location, representing another concentrated area of the backsoil effect. Due to slight differences in time and space between pressure and displacement responses, the first and second core development locations may overlap or be adjacent. Both point to the core area of the backsoil effect.
[0052] Furthermore, the first and second core development positions are merged, and the grouting intervention area is obtained by extending a predetermined distance forward and backward along the jacking direction. Specifically, the spatial union area of the first and second core development positions is taken as the reference point. A predetermined distance is extended forward and backward along the jacking direction of the pipe jacking machine. This predetermined distance is determined based on the arrangement spacing of the grouting device array and the effective diffusion radius of the grout, and is typically 1 to 2 times the spacing between adjacent grouting nozzles. A grouting nozzle spacing refers to the center distance between two adjacent grouting nozzles longitudinally arranged along the top of the pipe jacking machine shell and subsequent pipe sections. This distance is determined during the design phase based on the effective diffusion radius of the grout and geological conditions, and is typically 0.5 meters to 1.5 meters.
[0053] Specifically, the reason for setting up this extension step is that the back soil effect is not an isolated phenomenon, but a region with continuous distribution characteristics along the jacking direction. The core development location is the point where the back soil effect is most intense, but the transmission of friction and soil displacement will affect the surrounding areas. The continuous grouting area formed after the extension can cover the entire range of the back soil effect, avoiding the continuous accumulation of local friction caused by grouting omissions, and ensuring the continuity and effectiveness of the drag reduction effect.
[0054] There are two reasons for expanding outwards in steps with the grout nozzle spacing: First, expanding the grout nozzle spacing by one step allows the next set of grout nozzles to be included in the intervention area, ensuring that there is a grouting device that can be started immediately within the newly added range; Second, using a fixed spacing as the unit step facilitates the system to achieve a discretized and repeatable boundary expansion process, avoiding the problem of excessive control complexity caused by continuous expansion.
[0055] Furthermore, after obtaining the grouting intervention area, the grouting device at the corresponding location is activated to perform enhanced grouting.
[0056] Specifically, the grouting device at the corresponding location is activated to perform enhanced grouting, including: From the array of grouting devices arranged along the casing of the pipe jacking machine and the top of the subsequent pipe sections, select the grouting nozzle located in the grouting intervention area, increase the grouting pressure of the selected grouting nozzle to the preset enhanced grouting pressure ratio range, and continue grouting until the pipe-soil contact pressure in the grouting intervention area drops below the conventional pressure threshold.
[0057] After obtaining the grouting intervention area, grouting nozzles located within this intervention area are selected from the grouting device array deployed along the casing of the pipe jacking machine and the top of subsequent pipe sections. The grouting pressure of the selected nozzles is increased to a preset enhanced grouting pressure ratio range, and grouting continues until the pipe-soil contact pressure within the grouting intervention area drops below the conventional pressure threshold. The enhanced grouting pressure ratio range is typically set to 1.5 to 2.0 times the conventional grouting pressure, and the conventional pressure threshold can be determined based on the average pipe-soil contact pressure at each sensor measuring point within the grouting intervention area under normal jacking conditions.
[0058] By increasing the grouting pressure to within the enhanced grouting pressure ratio range, the high-pressure grout can more fully fill the voids between the top of the pipe section and the overlying soil, forming a continuous and dense lubricating film. This effectively reduces the pipe-soil friction coefficient and cuts off the friction transmission path. Grouting continues until the pipe-soil contact pressure drops below the conventional pressure threshold, indicating that the back-soil effect has been effectively suppressed. The conventional pressure threshold refers to the upper limit of the pipe-soil contact pressure allowed at each sensor point within the grouting intervention area under normal jacking conditions; it represents the critical pressure level where pipe-soil friction is within a controllable range. This conventional pressure threshold is determined based on historical data of pipe-soil contact pressure from each sensor within the grouting intervention area during recent stable jacking periods. For example, the average pressure value of each sensor point over the past 30 minutes plus one standard deviation can be used as the conventional pressure threshold. When the pipe-soil contact pressure within the grouting intervention area drops below this threshold, it indicates that the enhanced grouting has effectively reduced pipe-soil friction, the back-soil effect has been suppressed, and enhanced grouting can be stopped.
[0059] In addition, after the initial delineation of the grouting intervention area is completed, the step of adaptively expanding the grouting intervention area is also included.
[0060] Specifically, it also includes: Centered on the current grouting intervention area, continuously monitor the changing trend of the pipe-soil contact pressure on the outer side of the front boundary and the outer side of the rear boundary of the grouting intervention area along the jacking direction; if the increase in the pipe-soil contact pressure on either the outer side of the front boundary or the outer side of the rear boundary exceeds the preset multiple range of the historical average increase rate on that side, then dynamically expand the boundary of that side outward along the jacking direction by a grouting nozzle spacing to form an adaptively expanded grouting intervention area until the increase in the pipe-soil contact pressure on the outer side of that boundary falls back to the range of the historical average increase rate on that side.
[0061] During enhanced grouting, it is necessary to continuously monitor the changes in pipe-soil contact pressure along the outer edge of the front and rear boundaries of the grouting intervention area, centered on the current grouting intervention area. The front boundary refers to the leading edge of the grouting intervention area along the grouting direction, and the rear boundary refers to the trailing edge. The outer edge of the boundary refers to the section immediately beyond the boundary that is not yet covered by the current grouting intervention area. Pipe-soil contact pressure sensors are also deployed in this section to collect pipe-soil contact pressure data in real time.
[0062] If the rate of increase of the pipe-soil contact pressure on either the outer side of the front boundary or the outer side of the rear boundary exceeds the preset multiple of the historical average rate of increase on that side, the boundary of that side will be dynamically expanded outward by a grouting nozzle spacing along the jacking direction to form an adaptively expanded grouting intervention area until the rate of increase of the pipe-soil contact pressure on the outer side of that boundary falls back to the range of the historical average rate of increase on that side.
[0063] The historical average growth rate refers to the average rate of increase in the soil-to-tube contact pressure in the outer section of the boundary during normal jacking before the current grouting intervention area is formed. This average growth rate can be calculated based on historical data from sensors in that section during recent stable jacking periods, for example, the average pressure growth rate over the past 10 minutes. The preset multiplier range can be set based on engineering experience, for example, 1.5 to 2.0 times. When the soil-to-tube contact pressure growth rate on the outer side of the boundary exceeds 1.5 times the historical average growth rate on that side, it indicates that the backsoil effect is spreading outside the current grouting intervention area, and the current grouting range is insufficient to suppress the expansion of the backsoil effect.
[0064] At this point, starting from this side boundary, move outward along the jacking direction by the distance of one adjacent grouting nozzle, incorporating the newly added area into the grouting intervention zone, and activate the corresponding grouting nozzle to perform supplementary grouting. A grouting nozzle spacing refers to the center-to-center distance between two adjacent grouting nozzles longitudinally arranged along the top of the pipe jacking machine casing and subsequent pipe sections, typically ranging from 0.5 meters to 1.5 meters.
[0065] This adaptive expansion process continues. After each expansion, the pressure growth rate outside the new boundary is monitored. If it still exceeds the preset multiple range of the historical average growth rate, the boundary is expanded outward by one grouting nozzle spacing until the pressure growth rate outside the boundary falls back to the range of the historical average growth rate on that side. This indicates that the back soil effect has been effectively suppressed and the grouting intervention area has covered its actual influence range.
[0066] Specifically, the starting point of this step is that the actual impact range of the back soil effect is not fixed, but continuously evolves with the advancement of the jacking process and the dynamic changes in soil conditions. If a fixed boundary is used for the grouting intervention area, insufficient grouting may cause the back soil effect to spread beyond the boundary, or excessive grouting may result in grout waste. Through boundary pressure increase monitoring and an adaptive expansion mechanism, the grouting intervention area can dynamically expand along with the actual development of the back soil effect, ensuring the integrity and timeliness of drag reduction coverage, while avoiding over-grouting.
[0067] It should be noted that the enhanced grouting intervention performed during the vertical expansion stage of this method not only reduces the pipe-soil friction coefficient and cuts off the friction force transmission path, but also effectively reduces the additional load transmitted to the working well pit sidewall support structure due to the back soil effect. By intervening with grouting in the early stage of the back soil effect, it is possible to avoid the accumulation of additional loads that could lead to excessive deformation or instability of the support structure, thus achieving coordinated control of pipe jacking construction and pit sidewall support safety.
[0068] In summary, through the above steps, during the critical window period of the vertical expansion phase, the development location of the backsoil effect can be accurately pinpointed. A continuous grouting zone is formed outwards from the core development location, and enhanced grouting is then performed. The injected high-pressure grout fills the gaps between the soil and the pipe section, forming a lubricating film, thereby reducing the pipe-soil friction coefficient, cutting off the friction force transmission path, and preventing the soil from being further dragged forward by the pipe section. At this point, the backsoil effect disturbance is limited to a localized area at the top of the pipe section and has not yet spread horizontally. Timely intervention can nip the backsoil effect in the bud, preventing it from developing into a wider and more difficult-to-control horizontal expansion phase or an overall backsoil stage.
[0069] If this stage is missed, the backfilling effect will enter the horizontal expansion stage or the overall backfilling stage. At this point, grouting alone will be insufficient to suppress it, and higher-level mechanical barriers and parameter adjustment measures will be required. Therefore, identifying the development location of the backfilling effect and implementing enhanced grouting during the vertical expansion stage is the first proactive preventative line of defense for graded control.
[0070] S40: If the back soil effect development index exceeds the first warning threshold, activate the mechanical blocking device at the tail of the pipe jacking machine housing, so that the blocking plate of the mechanical blocking device extends to the top of the pipe section between the overlying soil. Furthermore, if the backfilling effect development index continues to rise and exceeds the first warning threshold after enhanced grouting is performed, it indicates that enhanced grouting is insufficient to suppress the further development of the backfilling effect. At this point, a higher level of active intervention measures, namely mechanical barrier devices, need to be initiated.
[0071] The mechanical barrier device is a physical isolation mechanism installed at the rear of the pipe jacking machine casing. Initially, its barrier plate is retracted into a groove at the top of the casing, not protruding beyond the casing surface. When the back soil effect intensifies and grouting drag reduction is insufficient, the drive mechanism pushes the barrier plate vertically outward, inserting it between the top of the pipe section and the overlying soil, directly cutting off the frictional force transmission path between the back soil and the pipe section. This mechanical barrier device, acting as a second line of defense after grouting drag reduction fails, physically prevents the soil from dragging forward with the pipe section, avoiding the continued deterioration of the back soil effect and preventing sudden surface collapse.
[0072] The first warning threshold is used to determine whether the back soil effect has developed to a critical value that enhanced grouting cannot effectively control. This threshold can be calibrated based on historical engineering data and field tests, for example, it can be set to 0.6. When the back soil effect development index reaches 0.6, it indicates that the rate of increase in pipe-soil contact pressure and soil displacement rate has significantly exceeded the normal range, the drag reduction effect of grouting is limited, and mechanical isolation needs to be initiated.
[0073] In addition, before activating the mechanical barrier device, the procedure includes: when the back soil effect development index continues to rise and exceeds the first warning threshold after the enhanced grouting is performed, before activating the mechanical barrier device, a jacking speed reduction operation is first performed to reduce the jacking speed to a preset reduction ratio range of the normal jacking speed. If the back soil effect development index still does not decrease after the speed reduction, the mechanical barrier device is then activated.
[0074] Specifically, if the backsoil effect development index continues to rise and exceeds the first warning threshold after enhanced grouting, it indicates that enhanced grouting is insufficient to suppress the further development of the backsoil effect. At this point, before activating the mechanical barrier device, a speed reduction operation should be performed. Jacking speed refers to the distance the pipe jacking machine advances per unit time, usually measured in millimeters per minute. The conventional jacking speed is a standard speed value preset under normal construction conditions based on geological conditions, pipe section size, and equipment capacity. The specific method for speed reduction is to reduce the jacking speed to a preset reduction ratio within the conventional jacking speed range, for example, reducing the conventional jacking speed to 50% to 70%.
[0075] The principle behind the ability of slowing down the jacking speed to suppress the backsoil effect is that the development of the backsoil effect is directly related to the friction between the top of the pipe section and the overlying soil, and the magnitude of the friction is positively correlated with the relative speed between the pipe section and the soil. When the jacking speed is reduced, the relative speed between the pipe section and the soil decreases, and the accumulation rate of the pipe-soil friction decreases accordingly, thereby slowing down the development of the backsoil effect and allowing more time for grouting and drag reduction to take effect.
[0076] If the back soil effect development index begins to decrease after the speed reduction, it indicates that the speed reduction operation has effectively suppressed the back soil effect, and there is no need to activate the mechanical blocking device. At this time, the jacking speed after the speed reduction can be maintained to continue construction. After the back soil effect development index falls back to a safe range, the speed can be gradually restored to the normal jacking speed.
[0077] If the back soil effect development index does not decrease after the speed is reduced, but continues to rise or remains at a high level, it indicates that simply reducing the speed is no longer sufficient to suppress the back soil effect. At this point, the mechanical barrier device should be activated to extend the barrier plate between the top of the pipe section and the overlying soil, physically severing the contact surface between the pipe section and the soil, and completely blocking the transmission path of friction.
[0078] Specifically, the activation and subsequent retrieval of mechanical barriers require additional construction procedures, and the prolonged extension of the barrier plate increases jacking resistance and may cause wear on the pipe sections. By prioritizing the use of a non-invasive intervention method—deceleration—and avoiding the activation of mechanical barriers whenever possible, mechanical wear and construction complexity can be reduced, while also lowering the risk of equipment failure. Mechanical barriers are only activated when deceleration proves ineffective, achieving a progressive procedure from soft to hard intervention in tiered control. This ensures the reliability of backfill effect control while avoiding the negative impacts of overuse of mechanical devices.
[0079] In summary, when the backfill effect develops to a certain extent, grouting alone is insufficient to suppress its spread. This step employs a mechanical barrier device to directly sever the contact between the pipe section and the soil through physical means, effectively eliminating the transmission of frictional force. After the barrier plate extends, the frictional force at the top of the pipe section is significantly reduced, the overlying soil is no longer dragged forward, and the vertical expansion path of the backfill effect is completely blocked. This measure, as a second line of defense after enhanced grouting, allows time for subsequent adjustments to the jacking parameters, preventing the backfill effect from further deteriorating into the overall backfill stage.
[0080] S50: If the back soil effect development index exceeds the second warning threshold, a warning signal is generated and the jacking parameters are adjusted. The second warning threshold is greater than the first warning threshold.
[0081] When the overburden effect development index continues to rise, exceeding the first warning threshold and requiring the initiation of enhanced grouting, speed reduction operations, and even mechanical isolation devices, if the development index still cannot be effectively controlled and continues to climb, exceeding the second warning threshold, it indicates that the overburden effect has developed to an extremely severe level. The overlying soil may have undergone overall shear failure, and the risk of surface subsidence has increased dramatically. Grouting, speed reduction, and mechanical isolation alone are no longer sufficient to reverse the situation. At this point, the highest level of emergency response measures must be activated immediately.
[0082] The second warning threshold, being higher than the first, is the critical value for determining whether the soil backfilling effect has entered an uncontrollable state. This threshold can be determined based on the critical development index corresponding to significant surface subsidence caused by the soil backfilling effect in historical engineering projects. For example, the first warning threshold can be set at 0.6, and the second warning threshold at 0.85. When the development index exceeds 0.85, it indicates that the soil backfilling effect has approached or reached the overall soil backfilling stage, and emergency measures must be taken immediately.
[0083] At this point, an early warning signal is generated. This signal can be sent to construction site management personnel and the remote monitoring center via audible and visual alarms, monitoring platform pop-ups, and SMS push notifications. The early warning signal should include the backsoil effect development index, the current stage, the core development location, the intervention measures already taken, and the recommended emergency response level. The purpose of this signal is to remind on-site personnel to immediately cease routine construction operations and activate the emergency plan.
[0084] Simultaneously, adjust the jacking parameters. Adjustments to the jacking parameters include, but are not limited to: immediately stopping jacking operations to prevent further accumulation of friction; reducing the jacking thrust to decrease the squeezing effect of the pipe sections on the soil; adjusting the jacking direction to avoid further advancement into areas with severe backsoil effects; increasing grouting pressure and volume, and performing full-section high-pressure grouting; and, if necessary, taking soil reinforcement measures, such as surface grouting or high-pressure jet grouting to reinforce the overlying soil. The specific adjustment range of the jacking parameters should be determined comprehensively based on the degree to which the backsoil effect development index exceeds the second warning threshold and the site geological conditions.
[0085] The starting point of this step is that when the backfilling effect develops to an uncontrollable state, the highest level of emergency response measures must be taken to ensure construction safety and the safety of the surrounding environment. Generating early warning signals ensures that relevant personnel are informed of the danger immediately and that emergency plans can be activated in a timely manner; adjusting the jacking parameters can reduce the driving force of the backfilling effect at its source and curb its further deterioration. Thus, this method forms a complete hierarchical prevention and control chain from "sensing—judgment—grouting—deceleration—isolation—early warning—parameter adjustment", realizing active control of the backfilling effect throughout the entire process.
[0086] In addition, it also includes: A three-dimensional geological model was established based on the geological survey data of the construction site; During the jacking process, the three-dimensional geological model is dynamically corrected using the real-time updated sensor monitoring data to obtain a digital twin geological model. The development index of the soil backing effect and the stage of the soil backing effect are displayed in three dimensions in the digital twin geological model.
[0087] Traditional soil backlash monitoring data is presented in numerical tables or two-dimensional curves, lacking spatial intuitiveness, making it difficult for construction managers to quickly understand the spatial distribution and development trend of soil backlash. By constructing a digital twin geological model and displaying it in three dimensions, abstract sensor data can be transformed into intuitive spatial information, improving on-site personnel's ability to perceive the development trend of soil backlash and providing more intuitive visual support for tiered prevention and control decisions.
[0088] First, a three-dimensional geological model is established based on the geological survey data of the construction site. Before the pipe jacking construction begins, information such as the distribution of soil layers, groundwater level, and physical and mechanical parameters of the soil at the construction site is obtained through geological survey. A three-dimensional geological model reflecting the actual geological conditions of the site is then constructed using three-dimensional modeling software. This model, based on three-dimensional spatial coordinates, presents the distribution range, layer thickness, and interface morphology of different soil layers in a three-dimensional form, providing a foundational map for subsequent jacking process simulation and backfill effect analysis.
[0089] Secondly, during the jacking process, the 3D geological model is dynamically corrected using real-time updated sensor monitoring data to obtain a digital twin geological model. During jacking, pipe-soil contact pressure sensors and soil displacement sensors continuously collect real-time data, reflecting the soil's actual response to the jacking construction. This sensor monitoring data is then integrated with the 3D geological model to dynamically correct for uncertainties in the soil parameters. For example, if the measured pipe-soil contact pressure in a certain area is higher than the model's predicted value, the soil strength parameter for that area is adjusted upwards; if the measured soil displacement is higher than the model's predicted value, the soil compression modulus for that area is adjusted downwards. Through dynamic correction, the 3D geological model gradually evolves into a digital twin geological model synchronized with the actual construction process, more accurately reflecting the soil response characteristics under the current construction conditions.
[0090] Furthermore, the backsoil effect development index and its current stage are visualized in a three-dimensional digital twin geological model. Specifically, the backsoil effect development index calculated in step S20 and the identified backsoil effect stage are overlaid into the digital twin geological model using color coding or dynamic cloud maps. For example, areas with lower backsoil effect development indices are displayed in green, while areas with higher indices are displayed in a yellow-to-red gradient, with different development stages marked using different textures or icons. Simultaneously, information such as the first core development location, the second core development location, the grouting intervention area, and the status of mechanical barrier devices are also marked in the model. Construction managers can intuitively observe the spatial distribution, development trend, and correlation with geological conditions of the backsoil effect through a three-dimensional visualization interface, aiding in decision-making regarding when and what intervention measures to initiate.
[0091] In addition, it also includes the source analysis steps for the propagation path of the back soil effect: When the back soil effect development index exceeds the second warning threshold, the change sequence of soil-to-pipe contact pressure data at each sensor location within the preset time window is traced back. The location where the pressure wave first appears abnormal is taken as the source point, and the direction of the pressure peak propagation to the current core development location is taken as the propagation path. A back soil effect source trajectory map is generated in the digital twin geological model. The initial triggering area of the back soil effect is located based on the source trajectory map.
[0092] When the soil backing effect development index exceeds the second warning threshold, it indicates that the soil backing effect has developed into a severely uncontrollable state. At this point, not only are emergency measures required, but it is also necessary to trace the initial location and propagation pattern of the soil backing effect to summarize experience and optimize subsequent construction plans. The specific steps of the source tracing analysis are as follows: First, the time sequence of changes in soil-to-pipe contact pressure data at each sensor location within a preset time window is traced back. The preset time window can be set according to the jacking speed and sensor spacing; for example, it can be set to trace back from the current moment to the past 2 hours.
[0093] Pressure waves refer to the dynamic changes in soil-to-tube contact pressure propagating along the jacking direction after the backsoil effect is triggered. When the soil at a certain location begins to be dragged forward with the tube section, the pressure at that location will rise first, followed by pressure anomalies at adjacent locations, forming a propagation sequence resembling a waveform. Based on this principle, the pressure time-series data of each sensor is played back frame by frame along the time axis from the current moment to the past to find the sensor location where the first abnormal increase in pressure occurs. The criteria for determining an abnormal increase in pressure can be determined based on the statistical characteristics of the sensor's historical data; for example, an anomaly is defined as a pressure value exceeding the historical average plus twice the standard deviation. The location where the first anomaly occurs is the initial trigger point of the backsoil effect.
[0094] After identifying the source point, the propagation trajectory of the pressure peak is traced along the timeline, starting from the moment the anomaly occurs at the source point. The pressure peak refers to the maximum pressure value reached by each sensor during the development of the backsoil effect. Connecting the moments when each sensor reaches its pressure peak in chronological order, the direction from the source point to the current core development location represents the propagation path of the backsoil effect.
[0095] Furthermore, a backsoil effect source trajectory map is generated in the digital twin geological model. Using timestamps as a guide, the trajectory map connects the moments of the first pressure anomalies at each sensor location in chronological order, forming a 3D visual marker with arrows. The starting point of the trajectory line is the location where the pressure wave first appeared, i.e., the initial triggering area; the arrows point to the current core development location of the backsoil effect. Trajectories from different time periods are distinguished by different colors; for example, early propagation trajectories are in blue, and recent propagation trajectories are in red.
[0096] The initial triggering area of the backfill effect can be located using a source trajectory map. Through this map, construction personnel can visually observe where the backfill effect originates, in what direction it propagates, and where it has currently expanded to, thus quickly locating the initial triggering area. This initial triggering area may be a local contact point between the top of the pipe section and the overlying soil, or it may be an area with specific geological anomalies.
[0097] Specifically, this step begins with the understanding that when the backsoil effect develops to an uncontrollable state, simply knowing its current severity is insufficient to guide subsequent construction optimization; it is also necessary to understand the root causes and propagation patterns of the backsoil effect. Through source tracing analysis, the initial location and triggering conditions of the backsoil effect can be identified, thus providing a basis for optimizing subsequent construction plans. For example, if source tracing reveals that the backsoil effect is always triggered from the same geological interface, enhanced grouting measures can be taken in advance at that interface during subsequent construction; if source tracing reveals that the backsoil effect is related to abnormal fluctuations in a certain jacking parameter, the setting range of that parameter can be optimized. Furthermore, the source trajectory diagram can also serve as an important basis for accident investigation and liability determination.
[0098] In summary, the embodiments of this application have at least the following technical effects: This invention first achieves real-time perception of the backsoil effect development process by loading soil contact pressure data and soil displacement data on the casing of the pipe jacking machine, the top of subsequent pipe sections, and the sidewall support structure of the working shaft pit. This solves the problem that traditional methods can only perform post-event analysis and cannot monitor in real time. Second, based on the sensor monitoring data, the stage of the backsoil effect and the degree of impact on the support are determined, and a backsoil effect development index is obtained. When it is in the vertical expansion stage, the development location is identified and enhanced grouting is initiated, realizing graded proactive intervention in the backsoil effect and support safety, avoiding the lag and blindness of manual experience judgment. Third, when the backsoil effect development index exceeds the first warning threshold, a mechanical blocking device is activated to extend the blocking plate between the top of the pipe section and the overlying soil, physically cutting off the vertical expansion path of the backsoil effect. When the development index exceeds the second warning threshold, a warning signal is generated and the jacking parameters are adjusted, forming a multi-level progressive prevention and control system.
[0099] This invention achieves real-time perception, intelligent prediction, and active control of the back soil effect through the synergistic effect of sensor monitoring, grouting intervention, mechanical isolation, and parameter adjustment. It simultaneously ensures the stability of the working well foundation pit sidewall support structure throughout the jacking process, suppresses the development of the back soil effect in rectangular pipe jacking construction in water-rich soft soil strata, meets the actual engineering requirements for high seepage prevention and high safety, and ensures the stability of the soil around the working well, the safety of the foundation pit sidewall support, and the integrity of the embankment seepage prevention structure.
[0100] Example 2, as Figure 2 As shown, based on the same inventive concept as the control method for a large rectangular surface pipe jacking well in water-rich soft soil strata provided in Embodiment 1, this embodiment of the invention also provides a control system for a large rectangular surface pipe jacking well in water-rich soft soil strata, comprising: Data loading module 11 is used to load sensor monitoring data on the casing of the pipe jacking machine, the top of the subsequent pipe sections, and the side wall support structure of the working well pit. The sensor monitoring data includes pipe-soil contact pressure data and soil displacement data. The effect discrimination module 12 is used to determine the stage of the back soil effect and the degree of impact on the sidewall support of the working well pit based on the soil contact pressure data and the soil displacement data. The back soil effect development index is calculated based on the stage of the back soil effect and the degree of impact. The stage of the back soil effect includes the vertical expansion stage. The grouting control module 13 is used to identify the development location of the back soil effect and obtain the grouting intervention area when the back soil effect is in the vertical expansion stage, and to start the grouting device at the corresponding location to perform enhanced grouting. The first execution module 14 is used to activate the mechanical blocking device at the tail of the pipe jacking machine housing if the back soil effect development index exceeds the first warning threshold, so that the blocking plate of the mechanical blocking device extends to the top of the pipe section between the overlying soil. The second execution module 15 is used to generate an early warning signal and adjust the jacking parameters if the back soil effect development index exceeds the second early warning threshold, wherein the second early warning threshold is greater than the first early warning threshold.
[0101] Specifically, the data loading module 11 is used for: Specifically, the sensor monitoring data loaded onto the casing of the pipe jacking machine, the top of subsequent pipe sections, and the sidewall support structure of the working shaft pit includes: The collected values of multiple sets of pipe-soil contact pressure sensors, which are arranged at intervals along the longitudinal direction of the pipe jacking machine casing and along the extension direction of the working well pit sidewall support structure, are added to the pipe-soil contact pressure data. The collected values of multiple sets of soil displacement sensors, which are arranged at intervals along the longitudinal direction of the pipe jacking machine casing and along the extension direction of the working well pit sidewall support structure, are added to the soil displacement data.
[0102] Specifically, the effect discrimination module 12 is used for: Based on the soil-to-pipe contact pressure data and the soil displacement data, the stage of the back soil effect is determined, and the back soil effect development index and the stage of the back soil effect are obtained. The stage of the back soil effect includes the vertical expansion stage.
[0103] In addition, before determining the stage of the soil-carrying effect, it is also necessary to train a soil-carrying effect development prediction model: Collect a dataset of back soil effect records from historical rectangular pipe jacking projects, and extract records of pipe-soil friction coefficient, jacking length, overburden pressure, soil displacement rate, and the degree of impact on the working well foundation pit sidewall support and the stage of back soil effect corresponding to each set of records. Using the impact of the working well foundation pit sidewall support and the stage of the back soil effect as supervision, and the recorded data of the pipe-soil friction coefficient, the recorded data of the jacking length, the recorded data of the overburden pressure and the recorded data of the soil displacement rate as input, a back soil effect development prediction model based on physical information neural network is trained. Based on the current rectangular pipe jacking project model, the back soil effect development prediction model is invoked. The pipe-soil friction coefficient, jacking length, overburden pressure, and soil displacement rate under the current construction state are used as input feature variables to obtain the stage of the back soil effect and the degree of impact on the sidewall support of the working well pit.
[0104] The grouting control module 13 is specifically used for: Specifically, the location of the backsoil effect development and the grouting intervention area are identified, including: The location of the sensor with the largest pressure increase in the soil-pipe contact pressure data is selected as the first core development location. The location of the sensor with the largest displacement rate in the soil displacement data is selected as the second core development location; The first core development position and the second core development position are merged and extended forward and backward by a preset distance along the jacking direction to obtain the grouting intervention area.
[0105] Specifically, the grouting device at the corresponding location is activated to perform enhanced grouting, including: From the array of grouting devices arranged along the casing of the pipe jacking machine and the top of the subsequent pipe sections, select the grouting nozzle located in the grouting intervention area, increase the grouting pressure of the selected grouting nozzle to the preset enhanced grouting pressure ratio range, and continue grouting until the pipe-soil contact pressure in the grouting intervention area drops below the conventional pressure threshold.
[0106] Specifically, it also includes: Centered on the current grouting intervention area, continuously monitor the changing trend of the pipe-soil contact pressure on the outer side of the front boundary and the outer side of the rear boundary of the grouting intervention area along the jacking direction; if the increase in the pipe-soil contact pressure on either the outer side of the front boundary or the outer side of the rear boundary exceeds the preset multiple range of the historical average increase rate on that side, then dynamically expand the boundary of that side outward along the jacking direction by a grouting nozzle spacing to form an adaptively expanded grouting intervention area until the increase in the pipe-soil contact pressure on the outer side of that boundary falls back to the range of the historical average increase rate on that side.
[0107] Specifically, the first execution module 14 is used for: If the back soil effect development index exceeds the first warning threshold, the mechanical blocking device at the tail of the pipe jacking machine casing is activated, so that the blocking plate of the mechanical blocking device extends to the top of the pipe section between the overlying soil.
[0108] In addition, before activating the mechanical barrier device, the procedure includes: when the back soil effect development index continues to rise and exceeds the first warning threshold after the enhanced grouting is performed, before activating the mechanical barrier device, a jacking speed reduction operation is first performed to reduce the jacking speed to a preset reduction ratio range of the normal jacking speed. If the back soil effect development index still does not decrease after the speed reduction, the mechanical barrier device is then activated.
[0109] The second execution module 15 is specifically used for: If the back soil effect development index exceeds the second warning threshold, a warning signal is generated and the jacking parameters are adjusted. The second warning threshold is greater than the first warning threshold.
[0110] In addition, it also includes: A three-dimensional geological model was established based on the geological survey data of the construction site; During the jacking process, the three-dimensional geological model is dynamically corrected using the real-time updated sensor monitoring data to obtain a digital twin geological model. The development index of the soil backing effect and the stage of the soil backing effect are displayed in three dimensions in the digital twin geological model.
[0111] In addition, it also includes the source analysis steps for the propagation path of the back soil effect: When the back soil effect development index exceeds the second warning threshold, the change sequence of soil-to-pipe contact pressure data at each sensor location within the preset time window is traced back. The location where the pressure wave first appears abnormal is taken as the source point, and the direction of the pressure peak propagation to the current core development location is taken as the propagation path. A back soil effect source trajectory map is generated in the digital twin geological model. The initial triggering area of the back soil effect is located based on the source trajectory map.
[0112] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
[0113] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A control system for a large rectangular surface pipe jacking well in water-rich soft soil strata, characterized in that, include: The data loading module is used to load sensor monitoring data from the casing of the pipe jacking machine, the top of the subsequent pipe sections, and the side wall support structure of the working well pit. The sensor monitoring data includes pipe-soil contact pressure data and soil displacement data. The effect discrimination module is used to determine the stage of the back soil effect and the degree of impact on the sidewall support of the working well pit based on the soil contact pressure data and the soil displacement data. The back soil effect development index is calculated based on the stage of the back soil effect and the degree of impact. The stage of the back soil effect includes the vertical expansion stage. The vertical expansion stage is characterized by the local soil at the top of the pipe section beginning to be dragged forward with the pipe section, and the disturbance gradually developing in the vertical direction from the top of the pipe section upwards into the deeper layers of the overlying soil. The grouting control module is used to identify the development location of the back soil effect and obtain the grouting intervention area when the back soil effect is in the vertical expansion stage, and to start the grouting device at the corresponding location to perform enhanced grouting. The first execution module is used to activate the mechanical blocking device at the tail of the pipe jacking machine housing if the back soil effect development index exceeds the first warning threshold, so that the blocking plate of the mechanical blocking device extends to the top of the pipe section between the overlying soil. The second execution module is used to generate an early warning signal and adjust the jacking parameters if the back soil effect development index exceeds the second early warning threshold, wherein the second early warning threshold is greater than the first early warning threshold. Before determining the stage of the soil-carrying effect, the process also includes training a model to predict the development of the soil-carrying effect. Collect a dataset of back soil effect records from historical rectangular pipe jacking projects, and extract records of pipe-soil friction coefficient, jacking length, overburden pressure, soil displacement rate, and the degree of impact on the working well foundation pit sidewall support and the stage of back soil effect corresponding to each set of records. Using the impact of the working well foundation pit sidewall support and the stage of the back soil effect as supervision, and the recorded data of the pipe-soil friction coefficient, the recorded data of the jacking length, the recorded data of the overburden pressure and the recorded data of the soil displacement rate as input, a back soil effect development prediction model based on physical information neural network is trained. Based on the current rectangular pipe jacking project model, the back soil effect development prediction model is invoked. The pipe-soil friction coefficient, jacking length, overburden pressure, and soil displacement rate under the current construction state are used as input feature variables to obtain the stage of the back soil effect and the degree of impact on the sidewall support of the working well pit.
2. The control system for a large rectangular surface pipe jacking well in water-rich soft soil strata as described in claim 1, characterized in that, Load sensor monitoring data from the casing of the pipe jacking machine, the top of subsequent pipe sections, and the sidewall support structure of the working shaft pit, including: The collected values of multiple sets of pipe-soil contact pressure sensors, which are arranged at intervals along the longitudinal direction of the pipe jacking machine casing and along the extension direction of the working well pit sidewall support structure, are added to the pipe-soil contact pressure data. The collected values of multiple sets of soil displacement sensors, which are arranged at intervals along the longitudinal direction of the pipe jacking machine casing and along the extension direction of the working well pit sidewall support structure, are added to the soil displacement data.
3. The control system for a large rectangular surface pipe jacking well in water-rich soft soil strata as described in claim 1, characterized in that, Identify the location of the backfill effect and obtain the grouting intervention area, including: The location of the sensor with the largest pressure increase in the soil-pipe contact pressure data is selected as the first core development location. The location of the sensor with the largest displacement rate in the soil displacement data is selected as the second core development location; The first core development position and the second core development position are merged and extended forward and backward by a preset distance along the jacking direction to obtain the grouting intervention area.
4. The control system for a large rectangular surface pipe jacking well in water-rich soft soil strata as described in claim 1, characterized in that, Also includes: Centered on the current grouting intervention area, continuously monitor the changing trend of the pipe-soil contact pressure on the outer side of the front boundary and the outer side of the rear boundary of the grouting intervention area along the jacking direction; if the increase in the pipe-soil contact pressure on either the outer side of the front boundary or the outer side of the rear boundary exceeds the preset multiple range of the historical average increase rate on that side, then dynamically expand the boundary of that side outward along the jacking direction by a grouting nozzle spacing to form an adaptively expanded grouting intervention area until the increase in the pipe-soil contact pressure on the outer side of that boundary falls back to the range of the historical average increase rate on that side.
5. The control system for a large rectangular surface pipe jacking well in water-rich soft soil strata as described in claim 1, characterized in that, Activate the grouting device at the corresponding location to perform enhanced grouting, including: From the array of grouting devices arranged along the casing of the pipe jacking machine and the top of the subsequent pipe sections, select the grouting nozzle located in the grouting intervention area, increase the grouting pressure of the selected grouting nozzle to the preset enhanced grouting pressure ratio range, and continue grouting until the pipe-soil contact pressure in the grouting intervention area drops below the conventional pressure threshold.
6. The control system for a large rectangular surface pipe jacking well in water-rich soft soil strata as described in claim 1, characterized in that, Also includes: If the back soil effect development index continues to rise and exceeds the first warning threshold after the enhanced grouting is performed, the jacking speed is reduced to a preset reduction ratio range of the normal jacking speed before the mechanical barrier device is activated. If the back soil effect development index still does not decrease after the speed reduction, the mechanical barrier device is activated.
7. The control system for a large rectangular surface pipe jacking well in water-rich soft soil strata as described in claim 1, characterized in that, Also includes: A three-dimensional geological model was established based on the geological survey data of the construction site; During the jacking process, the three-dimensional geological model is dynamically corrected using the real-time updated sensor monitoring data to obtain a digital twin geological model. The development index of the soil backing effect and the stage of the soil backing effect are displayed in three dimensions in the digital twin geological model.
8. The control system for a large rectangular surface pipe jacking well in water-rich soft soil strata as described in claim 1, characterized in that, It also includes the source analysis steps for tracing the propagation path of the back soil effect: When the back soil effect development index exceeds the second warning threshold, the change sequence of soil-to-pipe contact pressure data at each sensor location within the preset time window is traced back. The location where the pressure wave first appears abnormal is taken as the source point, and the direction of the pressure peak propagation to the current core development location is taken as the propagation path. A back soil effect source trajectory map is generated in the digital twin geological model. The initial triggering area of the back soil effect is located based on the source trajectory map.
9. A method for controlling a large rectangular surface pipe jacking well in water-rich soft soil strata, characterized in that, The control system for a large rectangular face pipe jacking well in water-rich soft soil strata, as described in any one of claims 1-8, comprises: The sensor monitoring data on the casing of the pipe jacking machine, the top of the subsequent pipe sections, and the side wall support structure of the working well pit are loaded. The sensor monitoring data includes pipe-soil contact pressure data and soil displacement data. Based on the soil-to-pipe contact pressure data and the soil displacement data, the stage of the back soil effect and the degree of impact on the sidewall support of the working well pit are determined. Based on the stage of the back soil effect and the degree of impact, the back soil effect development index is calculated. The stage of the back soil effect includes the vertical expansion stage. When the back soil effect is in the vertical expansion stage, the development location of the back soil effect is identified and the grouting intervention area is obtained. The grouting device at the corresponding location is then activated to perform enhanced grouting. If the back soil effect development index exceeds the first warning threshold, the mechanical blocking device at the tail of the pipe jacking machine casing is activated, so that the blocking plate of the mechanical blocking device extends to the top of the pipe section between the overlying soil. If the back soil effect development index exceeds the second warning threshold, a warning signal is generated and the jacking parameters are adjusted. The second warning threshold is greater than the first warning threshold.
Citation Information
Patent Citations
Control method for open caisson back soil deformation caused by pipe-jacking construction
CN104947649A
Method for judging dynamic back soil damage of rectangular jacking pipe
CN117763652A
Pipe-soil blocking device for rectangular pipe jacking and construction method of pipe-soil blocking device
CN120946850A