An automatic speed regulating and grouting adjusting system for deep mixing pile construction in a reclamation area

Abstract

The application relates to an automatic speed regulation and grouting adjustment system for deep mixing pile construction in a reclamation area, and the main steps include the following: firstly, a multi-modal sensing system is configured and installed (S101) to monitor key parameters of a construction site in real time, including torque, energy consumption, grouting pressure and stratum permeability; a data acquisition and fusion module is constructed (S102) to integrate multi-sensor data and form a real-time mapping of stratum-construction parameters; a relationship model (S103) of stratum input and stratum response is established through an extended Kalman filter (EKF) and a model predictive control (MPC) algorithm to intelligently evaluate stratum compactness and grouting quality; a feedback closed-loop regulation HFB system (S104) is based to realize real-time self-adaptive adjustment of mixing speed, grouting amount and construction pressure and realize precise control; through construction data storage and self-learning optimization (S105), the system records all construction data and improves long-term construction precision through deep reinforcement learning optimization control algorithm.

Description

TECHNICAL FIELD

[0001] The present application relates to deep mixing pile construction technology, in particular to an automatic speed regulation and grouting adjustment system for deep mixing pile construction in reclamation areas. BACKGROUND

[0002] Currently, in the deep mixing pile construction in reclamation areas and other complex strata, the construction environment is often full of uncertainty and challenges. Due to the unevenness of strata compactness, there are a large number of variables in the construction process, making it difficult to guarantee the construction quality of the mixing pile. The traditional deep mixing pile construction method usually relies on manual experience for operation, and the construction parameters such as mixing speed, feeding speed and grouting quality often rely on manual adjustment. This way not only is difficult to achieve precise control, but also is limited by human factors, which easily leads to low construction efficiency, serious material waste and unstable pile foundation quality, especially in complex geological conditions, the traditional method is difficult to cope with the influence of strata changes.

[0003] In addition, the traditional construction technology is based on fixed construction parameters, ignoring real-time adjustment and feedback in the construction process, so that the construction parameters cannot be optimized in real time according to the strata characteristics during the construction process, causing uneven construction quality and low efficiency. These problems seriously affect the pile foundation construction quality in reclamation projects and other complex geological areas, increase the project cost and prolong the construction period.

[0004] Therefore, how to realize real-time monitoring and automatic adjustment of the construction process under complex strata conditions has become the key to improving construction quality and efficiency. In view of this problem, it is particularly important to develop an intelligent construction control system. The system can real-time detect the compactness, permeability and other key parameters of the strata, and dynamically adjust the mixing speed, grouting amount and other construction parameters based on real-time data, so as to ensure the precise control of the construction process and reduce human operation errors.

[0005] The present application is based on the above-mentioned needs, and proposes an automatic speed regulation and grouting adjustment system for deep mixing pile construction. Through real-time data monitoring and intelligent control algorithm, the system can automatically optimize the construction parameters according to the construction environment and strata characteristics, significantly improve the construction efficiency, reduce material waste and ensure the stability of the pile foundation quality. Compared with the traditional method, the system of the present application not only improves the construction precision, but also has strong adaptability and controllability, which can optimize itself according to different geological conditions, thereby providing a more efficient and reliable solution for pile foundation construction in reclamation areas and other complex strata. SUMMARY

[0006] The present application aims to solve the problems in the prior art, and provides an automatic speed regulation and grouting adjustment system for deep mixing pile construction in a reclamation area, which solves the technical problem that in reclamation area construction, due to complex and difficult-to-accurately-identify stratum changes, traditional methods cannot effectively ensure mixing sufficiency and grouting quality, thereby affecting the overall strength and uniformity of the pile body. To this end, the present application proposes an intelligent system that can monitor the torque and wear of the drill bit in real time and automatically adjust the construction parameters to cope with changes in different strata, ensuring the stability and consistency of construction quality.

[0007] The technical scheme of the present application includes an automatic speed regulation and grouting adjustment system for deep mixing pile construction in a reclamation area, the main steps of which are as follows:

[0008] S101: Install a multi-modal sensing system to monitor key parameters in real time during construction;

[0009] S102: Data acquisition and fusion module, time tagging of key parameters through high-precision synchronization module, and downsampling of high-frequency data for fusion of key parameters monitored by multiple sensors;

[0010] S103: Establish a stratum response model to predict stratum density, set constraints and optimize construction parameters through extended Kalman filter EKF and model predictive control MPC algorithms, and ensure that the soil density reaches the expected target;

[0011] The objective function is:

[0012]

[0013] Wherein, is the deviation of the density, is the prediction step, is the density predicted at time , and is the expected stratum density;

[0014] The constraint conditions include setting hard limits for mixing speed and grouting pressure threshold ;

[0015] S104: Feedback-based closed-loop control HFB system, real-time adjustment of mixing speed and grouting amount through fuzzy PID controller to ensure construction quality;

[0016] S105: Construction data storage and self-learning optimization module, optimization of control strategy through deep reinforcement learning algorithm, and data management and storage using time series database.

[0017] In step S101, the specific operation is as follows:

[0018] S11: According to the key factors of deep mixing pile construction, torque sensor, energy monitoring module, grouting pressure sensor and formation permeability probe are selected, wherein:

[0019] The high-precision dynamic torque sensor is selected for monitoring the rotating resistance of the mixing pile;

[0020] The energy monitoring module integrates current transformers and voltage sensors to calculate the power of the mixing motor in real time;

[0021] The grouting pressure sensor adopts a corrosion-resistant pressure resistance sensor to monitor the pressure change during grouting;

[0022] The formation permeability probe is installed with a miniature pore water pressure gauge to indirectly evaluate the formation permeability coefficient based on the static sounding principle to evaluate the grouting effect.

[0023] S12: The torque sensor is installed at the output end of the mixing drill rod head, and a set of torque sensors is arranged every 2 meters along the axial direction of the drill rod to form a distributed monitoring network. The energy monitoring module is installed at the power input section of the electric control system. The grouting pressure sensor is installed at the outlet pipeline of the grouting pump. The formation permeability probe is embedded in the mixing blade in a spiral array and synchronously collects formation data with the drill rod rotation.

[0024] S13: Set threshold alarm, trigger early warning when torque exceeds 80% of the set value, and suspend construction when grouting pressure fluctuation amplitude > ± 15%.

[0025] In step S102, the specific operation is as follows:

[0026] S21: Deploy high-precision GPS synchronization module (PPS accuracy ± 10ns) to add uniform time label to all sensor data, and downsample high-frequency data to align with low-frequency data at 10Hz common sampling rate.

[0027] S22: Wavelet threshold denoising is used for torque signal, real-time moving average filtering is used for grouting pressure data, and isolated forest algorithm is used to identify outliers, and linear interpolation is used to fill in missing data.

[0028] S23: Time domain features (mean, variance) and frequency domain features (5-20Hz band energy) are extracted from torque data, and dynamic pressure gradient is calculated according to grouting pressure data, and its equation is:

[0029]

[0030] wherein, is the pressure at time , is the spatial position at time , is the spatial position at time , is the spatial position at time It is a moment The spatial location, This refers to the time step. An extended Kalman filter (EKF) is used to fuse formation permeability, torque, and grouting pressure to generate a 6-dimensional state vector:

[0031] [Formation permeability coefficient, mean torque, pressure gradient, torque variance, frequency band energy, formation density].

[0032] In step S103, the specific operations are as follows:

[0033] S31: By defining key state variables, a system capable of predicting formation compaction can be established. The model. First, define the state vector:

[0034]

[0035] in The formation permeability coefficient, The average torque. For pressure gradient, For torque variance, For frequency band energy, For density.

[0036] The observation equation is then:

[0037]

[0038] This function reflects the influence of parameters such as formation permeability coefficient, mean torque, pressure gradient, torque variance, and frequency band energy on formation density.

[0039] S32: Extended Kalman Filter (EKF) is used for state estimation of nonlinear systems to obtain more accurate formation state information. Through iterative processing in the prediction and update stages, EKF can correct the predicted state variables based on sensor data and process noise, providing real-time formation state estimates. This method is particularly suitable for handling systems with nonlinear relationships and optimizes the accuracy of the estimation process by setting an appropriate noise covariance matrix.

[0040] S33: With the assistance of the Model Predictive Control (MPC) algorithm, the system adjusts control parameters such as stirring speed and grouting volume through rolling time-domain optimization to minimize the deviation in formation compaction. MPC sets the prediction step size... (i.e., in the future) (Optimization prediction within a certain time frame) and control step size (i.e., each) control adjustment once per time), combined with the objective function and constraint conditions, real-time optimization of parameter adjustment is realized, thereby ensuring the stability and efficiency of the construction process. With the advancement of the control cycle, MPC continuously optimizes the control strategy according to new data, ensuring that the soil density reaches the expected target.

[0041] The objective function is as follows:

[0042]

[0043] the deviation of the density, the predicted density at time the expected stratum density.

[0044] The constraint conditions are as follows: set the hard limit of the stirring speed and the threshold of the grouting pressure .

[0045] S34: Online parameter identification is realized through recursive least squares (RLS), which can update model coefficients in real time, ensuring that the model has strong adaptability to the changing construction environment. Coefficients are updated every 5 seconds, gradually improving the prediction accuracy of the model. After completing a certain number of construction piles, offline training and model version update are triggered to ensure that the system is continuously optimized and the latest model is applied to the subsequent construction process.

[0046] In step S104, the specific operation is as follows:

[0047] S41: dynamically adjust the stirring speed through the fuzzy PID controller, the input of which is the density deviation , i.e. the difference between the current density and the target density, and the output is the adjustment amount of the stirring speed , the formula is:

[0048]

[0049] wherein, is the proportional gain, which determines the adjustment amount proportional to the current deviation; is the integral gain, which reflects the influence of past cumulative errors and helps to eliminate steady-state errors; is the derivative gain, which is used to consider the rate of error change and help to reduce oscillation or overshoot.

[0050] If , increase the stirring speed; if , reduce the stirring speed.

[0051] ​S42: Limit the grouting pressure using a dynamic threshold algorithm to prevent overpressure damage to equipment and ensure construction quality.

[0052] Define the safe range of grouting pressure , i.e. the upper and lower threshold values of grouting pressure.

[0053] Calculate the deviation based on real-time pressure data , through the dynamic threshold algorithm, when the grouting pressure exceeds the set dynamic threshold , the control system adjusts the flow of the grouting pump through the PID control algorithm, reducing the grouting amount, thereby reducing the pressure, and by adjusting the speed or power of the servo motor, slowing down the grouting speed, thereby reducing the system pressure.

[0054] When the grouting pressure is below the minimum threshold , the system increases the grouting flow to increase the pressure.

[0055] According to the construction progress and specific circumstances, the dynamic threshold algorithm continuously adjusts the threshold. For example, at different stages of construction, as the grouting amount increases or the pressure changes, the dynamic threshold will change accordingly. This helps to more flexibly respond to pressure fluctuations and prevent equipment from being subjected to excessive pressure.

[0056] In addition, a hard limit for stirring speed is set to prevent equipment damage, as well as an emergency grouting pressure cutoff threshold , once the pressure exceeds the limit, the system automatically cuts off the grouting to avoid irreversible damage to the system caused by excessive pressure, ensuring equipment safety and personnel safety during construction.

[0057] S43: Through real-time feedback control system, according to the data collected by the sensor, the system continuously adjusts the stirring speed, grouting amount and pressure, ensures the continuous optimization of the construction process, maintains the optimal working state. This closed-loop control system ensures construction accuracy and efficiency, adapts to changes in different strata conditions.

[0058] In step S105, the specific operation is as follows:

[0059] S51: In data management and storage, time series database InfluxDB is used to store all structured data collected by sensors, ensuring data query and processing by timestamp. The sampling interval is set to 1 second to ensure high-precision data recording and real-time requirements. At the same time, the video stream of the construction site uses HDFS distributed storage to ensure efficient, reliable storage and management of large amounts of video data, providing a solid data foundation for subsequent analysis and optimization.

[0060] S52: The DDPG algorithm in reinforcement learning training can optimize system decision-making in continuous control environment through Actor-Critic structure, helping to automatically adjust the mixing speed and grouting amount and other key operations. The reward function guides learning through the balance of density and energy consumption, ensuring that the target density meets the requirements while the energy consumption remains within the minimum range. The form of the reward function is:

[0061]

[0062] wherein, is the current density, is the target density, is the maximum energy consumption, is the current energy consumption, and are weight coefficients for balancing the optimization between density and energy consumption.

[0063] Through system training, the model is continuously optimized, and the generated control strategy can cope with changes in different construction environments, improving system adaptability. After each training, the strategy is optimized and updated to achieve more accurate and efficient construction operations, thereby continuously improving efficiency and quality during construction.

[0064] S53: The implementation verification stage will conduct closed-loop testing in typical reclamation strata to verify the actual effect of the system. BRIEF DESCRIPTION OF DRAWINGS

[0065] The accompanying drawings are only used for illustrative purposes and cannot be understood as a limitation of the present scheme; in order to better illustrate the present scheme, some components in the drawings may be omitted, enlarged or reduced, and do not represent the size of the actual product; it is understandable that some well-known structures and their descriptions in the drawings may be omitted for those skilled in the art.

[0066] Figure 1 is a flowchart of an automatic speed regulation and grouting regulation system for reclamation area deep mixing pile construction of the present application;

[0067] Figure 2 is a drill pipe structure and sensor deployment profile;

[0068] Figure 3 is a structural schematic diagram of a mixing drill bit;

[0069] Figure 4 is a stratum response model schematic diagram;

[0070] Figure 5 is a system operation interface schematic diagram.

[0071] The figures in the specification include: tower 001, power head 002, mixing drill pipe 003, torque sensor 004, mixing drill bit 005, base 006, energy monitoring module 007, electric control system 008, grouting pipe 009, high-pressure mud pump 010, grouting pressure sensor 011, mixing blade 012, mixing element 013, formation permeability probe 014. DETAILED DESCRIPTION

[0072] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0073] Embodiment: Reference Figures 1-5

[0074] In particular, as Figure 1 indicated, the present application provides an automatic speed regulation and grouting adjustment system for reclamation area deep mixing pile construction, which aims to improve construction quality and efficiency by real-time dynamic adjustment of construction parameters, adopts a "energy regulation-formation response" dual variable control method, combines with a phased construction regulation strategy, and ensures the accuracy and adaptability of the construction process. The implementation steps are as follows:

[0075] Step S101, the sensor module includes a torque sensor, a wear detection device, a pressure sensor, a flow sensor, and a formation permeability probe. Each sensor collects the mixing speed, grouting pressure, grouting flow, and formation density in real time, and transmits the data to the data acquisition module through a combination of wired and wireless networks.

[0076] The data acquisition and fusion module in step S102 includes a data collector, a signal amplifier, a data converter, and a data fusion algorithm. The original data collected by the sensors is enhanced by the signal amplifier and converted into a standard signal format suitable for transmission and processing by the data converter. Subsequently, the data fusion algorithm integrates and calibrates the multi-source data, and transmits the processed data to the central control system through the data bus for subsequent analysis and decision-making.

[0077] Step S103, the system uses extended Kalman filter (EKF) and model predictive control (MPC) algorithm to build a formation response model. By analyzing the formation and grouting data collected by the sensors, the system establishes a relationship model between the formation input and the formation response. This model evaluates the formation density and grouting quality in real time, intelligently analyzes the formation changes during construction, and provides a basis for construction adjustment, thereby optimizing the construction effect.

[0078] Step S104 includes feedback closed-loop control of the HFB system, which uses the aforementioned formation response model, real-time data, and intelligent algorithms for dynamic control. The fuzzy PID controller adjusts the stirring speed according to the density deviation and adjusts the regulating signal of the grouting pump according to the grouting amount deviation, ensuring accurate and uniform grouting. The dynamic threshold algorithm is used to limit the grouting pressure to avoid damage to the equipment caused by overpressure. The grouting flow rate is adjusted according to the real-time pressure data to maintain the grouting pressure within a safe range. Through this feedback mechanism, the system can adaptively adjust the construction parameters to ensure precise control and dynamic optimization of the construction process.

[0079] Step S105 uses the time series database InfluxDB to store all structured data collected by the sensors, ensuring high-precision data recording and real-time requirements, and uses HDFS distributed storage to manage large amounts of video data. In reinforcement learning training, the DDPG algorithm is used to optimize key operations such as stirring speed and grouting amount, ensuring the accuracy and efficiency of the construction process.

[0080] Through the collaborative work of the multi-modal sensing system, data acquisition and fusion module, formation response model, feedback closed-loop control HFB system, and construction data storage and self-learning optimization module, the invention breaks through the limitations of traditional single parameter optimization methods. The system can adaptively adjust the construction parameters in real time, achieve energy-formation dual variable control and dynamic optimization throughout the process, and significantly improve the safety, uniformity, and construction efficiency of the reclamation area deep mixing pile construction.

[0081] In particular, Figure 2 The overall scheme of the drill pipe structure and sensor deployment and its workflow are shown, and the core system includes sensor configuration, data acquisition, data processing and intelligent algorithm judgment, automatic adjustment and feedback mechanism. In particular, step S101, the sensors are arranged during the drilling operation, including torque sensor 004, formation permeability probe 014, grouting pressure sensor 011, and energy monitoring module 007, which collect torque, formation permeability, and grouting pressure of the mixing drill pipe 003 in real time and transmit them to the data acquisition module through a combination of wired and wireless networks.

[0082] The torque sensor 004 is installed at the output end of the mixing drill pipe 003, wherein a set of torque sensors 004 is arranged every 2m along the drill pipe axis to form a distributed monitoring network, the energy monitoring module 007 is installed at the power input section of the electric control system 008, and the grouting pressure sensor 011 is installed at the outlet pipeline of the grouting pipe 009 of the high-pressure mud pump 010. The formation permeability probe 014 is embedded in the mixing blade 012 in a spiral array and synchronously collects formation data with the rotation of the mixing drill pipe 003.

[0083] As Figure 3As shown, a structural diagram of the mixing drill bit 005 is shown, which mainly includes:

[0084] Mixing drill pipe 003: connecting the power head 002 and the mixing drill bit 005, transmitting the rotating power of the power head 002 to the mixing drill bit 005;

[0085] Mixing blade 012: used for mixing soil, cement slurry or other materials to make them fully mixed;

[0086] Formation permeability probe 014: used for detecting the permeability of the formation to help determine the grouting parameters.

[0087] The formation permeability probe 014 is marked with a green triangle on the structural diagram of the mixing drill bit 005, and the blue dashed line represents the movement trajectory of the formation permeability probe 014 during the rotation or mixing during the construction process. The probe moves with the rotation of the mixing drill bit 005, thereby detecting the formation permeability at different positions and providing more comprehensive formation permeability data.

[0088] In particular, the attached Figure 4 A schematic diagram of the formation response model is shown, and the core system includes data preprocessing, intelligent control algorithm, execution layer adjustment and feedback mechanism. In particular, the data acquisition system containing step S102 includes a data collector, a signal amplifier and a data converter, which collects the original data sent by the sensor and converts it into a signal format suitable for transmission and processing. In particular, containing step S103, the data in the data acquisition module is connected with the central control system through the data bus, and the data processing module in the central control system performs cleaning, denoising, isolation forest outlier detection and other preprocessing work on the received original data, to ensure the accuracy and reliability of the data. The data processed by EKF filtering is sent to the intelligent algorithm analysis module for formation density evaluation, grouting quality analysis, construction state judgment, etc.

[0089] In particular, step S104 is included, and the intelligent control system uses a PID control strategy to regulate the grouting flow and the servo motor speed. Among them, the proportional (P) control adjusts the grouting pressure in real time, the integral (I) control eliminates accumulated errors, and the differential (D) control predicts the trend to prevent overshoot. The intelligent algorithm adjusts the control parameters according to the real-time pressure feedback to maintain stable grouting pressure, and combines the MPC (Model Predictive Control) optimization strategy to predict the system response in a short time, dynamically adjust the construction parameters, and improve the control precision. The hard limit protection mechanism is used to prevent the system from running out of the safe range. The maximum speed of the stirring shaft and the upper limit of the grouting pressure are set as forced constraints. When the system detects that the construction parameters are close to the safety threshold, the intelligent algorithm will adjust the parameters in advance. If it exceeds the safety range, an emergency stop will be triggered immediately. The emergency shutdown mechanism is responsible for protecting construction safety in extreme cases. When the grouting pressure exceeds the maximum grouting pressure or the stirring speed exceeds the maximum speed, the system immediately executes the emergency stop process, cuts off the operation of the grouting pump, and sends a high-priority alarm message to the operator to ensure the safety of the construction site.

[0090] In particular, step 105 is included, and the system uses a dynamic feedback mechanism to record the adjusted construction parameters and optimization results in the database, which is further used for construction quality evaluation and system optimization. Historical data will be used to improve the dynamic threshold calculation model and provide optimization suggestions for subsequent construction, forming a closed-loop control system to realize intelligent and adaptive optimization of the construction process.

[0091] As shown in Figure 4 The key parameters in the construction process are monitored in real time by the sensor module in the environment and delivered to the data acquisition module. The key parameters include torque, formation permeability, grouting pressure, etc. After the received raw data is cleaned, denoised, and isolated forest outlier detection, etc. in the data preprocessing module, the data processed by EKF filtering is sent to the MPC rolling optimization algorithm and PID control algorithm module for formation density evaluation, grouting quality analysis, construction state judgment, etc. The optimal control strategy is generated, and the control signal is generated according to the system error. Then, the PID controller regulation decision module judges whether the system has abnormalities or potential risks according to the state prediction result and the preset threshold. If there are abnormalities, the specific abnormal conditions are output to the staff through the human-computer interaction system. The staff sends manual instructions to the adjustment output module, which sends instructions to the central control system for dynamic adjustment of the output parameters of the system. After receiving the instructions, the central control system transmits instructions to the execution template, and the execution template executes specific operation tasks according to the sent instructions to realize the actual control function of the system.

[0092] In particular, Figure 5The main functions and user interaction parts of the central control system operation interface are displayed, including the contents of steps S105 and S106. The top of the interface is the navigation bar, which spans the entire screen and includes main function options, help, user login, and a notification center for quick access to different function modules. The main function options include:

[0093] Real-time monitoring is used to obtain real-time construction status, construction parameters, and construction state, etc.

[0094] Historical data is used to store various historical data of the construction, so that the system can establish a stratum response model and optimize the control strategy through deep reinforcement learning algorithm for self-learning optimization;

[0095] System settings allow staff to adjust the system display interface and other settings to better suit their usage habits;

[0096] Report generation can automatically generate warning reports based on abnormal detection and risk assessment results. The report content includes detailed description of abnormal situation, risk level and its impact range, suggested handling measures for abnormal situation, such as adjusting construction parameters, increasing monitoring frequency, etc., as well as graphical risk assessment results and key parameter trends in the construction process to help construction personnel and decision-makers quickly understand.

[0097] The left side is the real-time monitoring panel, occupying one-third of the screen. The real-time monitoring panel includes:

[0098] Construction status, used to display real-time construction parameters, including mixing pile number, current depth, mixing speed, grouting pressure, stratum permeability coefficient, torque, and stratum density;

[0099] Real-time chart, used to graphically display risk assessment results and key parameter trends in the construction process;

[0100] The alarm indicator light includes abnormal, suspected abnormal, and normal indicator lights, which allow staff to quickly confirm the construction status through the alarm indicator light.

[0101] The right side is the construction control panel, occupying two-thirds of the screen. The construction control panel includes: manual parameter adjustment module, which can adjust parameters by manually pulling the parameter bar left and right;

[0102] It also includes the operation control button on the left and the mode switching button on the right. The operation control button is divided into start, pause, and stop, and the mode switching button is divided into manual and automatic. Staff can select the current system working state by pressing the button, which is simple and easy to understand;

[0103] The construction control panel also includes a real-time data input box through which a worker can manually enter accurate construction parameter values to adjust the construction state.

[0104] The bottom of the interface is a system status panel that spans the entire screen, displaying system health status indicators, device status monitoring, and a system journal window. Through the top navigation bar, the user can access historical data and report panels, as well as a help and support panel, which are typically presented in pop-up windows or new pages.

Claims

1. An automatic speed regulation and grouting adjustment system for deep mixing pile construction in a reclamation area, characterized in that, The method comprises the following steps: S101: install a multi-modal sensing system for real-time monitoring of key parameters during construction; S102: a data acquisition and fusion module, which time-labels the key parameters through a high-precision synchronization module and down-samples high-frequency data therein for fusing the key parameters monitored by multiple sensors; S103: establish a stratum response model to predict stratum density, set constraints and optimize construction parameters through extended Kalman filter (EKF) and model predictive control (MPC) algorithms, and ensure that the soil density reaches the expected target; Objective function: wherein, is a deviation of the density, is a prediction step, is a time predicted density, is an expected formation density; The constraint includes a hard limit on the setting of the stirring speed and a threshold value for the grouting pressure ; S104: a feedback closed-loop control HFB system for real-time adjustment of stirring speed and grouting amount through a fuzzy PID controller to ensure construction quality; S105: a construction data storage and self-learning optimization module for optimizing control strategies through deep reinforcement learning algorithms and managing and storing data using a time series database; In step S103, the stratum response model optimizes control parameters, including stirring speed and grouting amount, based on real-time stratum state and construction data through model predictive control (MPC) algorithms to minimize the deviation of stratum density.

2. The automatic speed regulating and grouting adjusting system for deep mixing pile construction in a reclamation area according to claim 1, characterized in that, The multi-modal sensing system in step S101 includes a torque sensor, an energy monitoring module, a grouting pressure sensor, and a stratum permeability probe; The torque sensor is used to monitor the rotation resistance of the mixing pile, the energy monitoring module is used to calculate the power of the mixing motor in real time, the grouting pressure sensor is used to monitor the change of grouting pressure, and the stratum permeability probe indirectly evaluates the stratum permeability coefficient through the static sounding principle to evaluate the grouting effect.

3. The automatic speed regulating and grouting adjusting system for deep mixing pile construction in a reclamation area according to claim 2, characterized in that, The torque sensor is installed at the output end of the mixing drill rod head, and a set of sensors is arranged every 2 meters; The grouting pressure sensor is installed at the outlet pipeline of the grouting pump; The stratum permeability probe is embedded in the mixing blade and synchronously collects data with the rotation of the drill rod.

4. The automatic speed regulating and grouting adjusting system for deep mixing pile construction in a reclamation area according to claim 1, characterized in that, The data acquisition and fusion module in step S102 further includes a high-precision GPS synchronization module for time synchronization, and an extended Kalman filter (EKF) algorithm is used to fuse stratum permeability, torque, and grouting pressure data to generate a 6-dimensional state vector: [Stratum permeability coefficient, torque average, pressure gradient, torque variance, frequency band energy, stratum density].

5. The automatic speed regulating and grouting adjusting system for deep mixing pile construction in a reclamation area according to claim 1, characterized in that, The deep reinforcement learning algorithm in step S105 continuously optimizes control strategies based on historical data during construction to improve construction accuracy.

6. The automatic speed regulation and grouting adjustment system for deep mixing pile construction in a reclamation area according to claim 1, characterized in that, The feedback closed-loop control HFB system in step S104 adjusts the stirring speed and grouting amount through a fuzzy PID controller, specifically by adjusting the stirring speed and grouting pump flow in real time to ensure the accuracy and uniformity during construction.

7. The automatic speed regulation and grouting adjustment system for deep mixing pile construction in a reclamation area according to claim 6, characterized in that, The fuzzy PID controller adjusts the stirring speed based on the density deviation, and the grouting pump flow control adjusts the servo motor speed or power in real time to ensure grouting accuracy and uniformity.

8. The automatic speed regulation and grouting adjustment system for deep mixing pile construction in a reclamation area according to claim 7, characterized in that, Step S104 specifically includes: S41: dynamically adjusting the stirring speed by the fuzzy PID controller, the input of which is the compactness deviation , i.e. the difference between the current compactness and the target compactness, and the output is the adjustment amount of the stirring speed : P is a proportional gain that determines the amount of adjustment that is proportional to the current error; Integral gain, reflecting the influence of past cumulative error, helps to eliminate steady-state error; Kd is the derivative gain, used to consider the rate of error change, helping to reduce oscillations or overshoot; If , increase the stirring speed; if , decrease the stirring speed; S42: limit the grouting pressure using a dynamic threshold algorithm to avoid overpressure damage to equipment and ensure construction quality; Defining a safety range for grouting pressure i.e. upper and lower threshold values for grouting pressure; Calculating deviation from real-time pressure data , through a dynamic threshold algorithm, when the grouting pressure exceeds the set dynamic threshold , the control system adjusts the flow of the grouting pump through a PID control algorithm, reduces the grouting amount, thereby reducing the pressure, and adjusts the speed or power of the servo motor to slow down the grouting speed, thereby reducing the system pressure; When the grouting pressure is below the minimum threshold , the system increases the pressure by increasing the grouting flow rate; Hard limit of the stirring speed is set To prevent the equipment from being damaged, and an emergency grouting pressure cut-off threshold When the emergency grouting pressure cut-off threshold is exceeded The system automatically cuts off the grouting to avoid irreversible damage to the system caused by excessive pressure; S43: Through real-time feedback control system, according to the data collected by the sensor, the system continuously adjusts the stirring speed, grouting quantity and pressure to ensure the continuous optimization of the construction process and maintain the optimal working state.

9. The automatic speed regulating and grouting adjusting system for deep mixing pile construction in a reclamation area according to claim 1, characterized in that, The construction data storage and self-learning optimization module in step S105 uses a time series database to store sensor data and optimizes control strategies through reinforcement learning algorithms to generate control strategies for different construction environments, thereby improving construction efficiency and quality.

Images