A high-pressure jet grouting pile construction optimization control method and system

Abstract

The application discloses a high-pressure rotary jet grouting pile construction optimization control method and system, and belongs to the technical field of pile foundation construction. The method predicts the pile strength based on an XGBoost regression model, and a HSIO hybrid swarm intelligence optimization algorithm is combined with an artificial bee colony algorithm and a beluga whale optimization algorithm to dynamically adjust the construction parameter combination, so that the optimal matching of the target strength and the real-time construction state is realized. The application solves the problems of parameter setting deviation and uneven pile quality caused by the dependence of traditional construction on manual experience, is especially suitable for high-precision construction requirements under complex geological conditions (such as karst areas and deep silt layers), significantly improves the pile strength stability, construction efficiency and quality controllability, and reduces material waste and human error risks.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of pile foundation construction, and particularly relates to a high-pressure jet grouting pile construction optimization control method and system. BACKGROUND

[0002] Traditional methods such as throwing stone, following with steel casing, and drilling and grouting are often difficult to effectively solve problems when facing deep silt layer and complex karst area, resulting in low construction efficiency, high cost, high quality risk, and even causing safety accidents. High-pressure jet grouting pile technology forms a strong reinforcement body in soft ground by high-pressure jet grouting cement slurry, effectively fills the cave, and enhances the bearing capacity of the foundation and the stability of the pile foundation. By using the combination of pile circumferential jet grouting pile and inter-pile jet grouting pile, a closed reinforcement system is formed, further improving the stability of the pile foundation and avoiding the problems of cavities or uneven reinforcement that may occur in traditional construction. During construction, the operation progress of drilling and grouting is strictly controlled to ensure continuous operation of each pile, and real-time monitoring is combined to ensure controllable construction quality. During the construction of jet grouting piles, jet pressure, jet volume, drill pipe lifting speed, and drill pipe rotation speed are key parameters that affect pile quality. Traditional construction methods mainly rely on manual experience to adjust these parameters. However, the traditional method is difficult to ensure the accuracy and consistency of the parameters, and is easily affected by human factors, resulting in unstable construction quality. Manual adjustment is easily disturbed by subjective factors, and the parameters lack precision and consistency. It is difficult to dynamically adjust the jet parameters according to real-time construction data, and it is difficult to adapt to complex and variable construction environments, which may lead to uneven jetting and insufficient pile strength. In addition, the construction environment is complex, and changes in factors such as mud density and soil characteristics further increase the uncertainty of construction.

[0003] Although there are some automatic control methods in the current industry, their optimization ability is limited, and they cannot be dynamically adjusted according to real-time construction data, resulting in limited improvement in construction efficiency and quality. Existing automatic control methods and systems are mainly applied to simple construction environments, and are difficult to meet the needs of construction projects with complex geological conditions and high precision requirements, and cannot be flexibly adapted to unexpected situations. In addition, there is a lack of closed-loop feedback mechanism, and the correlation between construction parameters and target strength cannot be real-time corrected, making it difficult to guarantee the stability and uniformity of the pile quality. Therefore, an intelligent control parameter high-pressure jet grouting pile method and system is urgently needed to meet the needs of modern foundation construction. SUMMARY

[0004] In view of the deficiencies of the prior art, one of the purposes of the present application is to provide a high-pressure jet grouting pile construction optimization control method to achieve real-time dynamic adjustment and optimization of construction parameters, reduce manual intervention, eliminate human error, and improve construction consistency and efficiency.

[0005] The second object of the present application is to provide a system for implementing the high-pressure rotary jet pile construction optimization control method.

[0006] The present application provides a high-pressure rotary jet pile construction optimization control method, comprising the following steps:

[0007] S1. Collecting historical construction data of high-pressure rotary jet piles; preprocessing the collected real-time construction data of high-pressure rotary jet piles to obtain pretreatment data;

[0008] S2. Based on a machine learning algorithm, using the pretreatment data as training data, a pile strength prediction model is constructed;

[0009] S3. Based on the hybrid swarm optimization algorithm (HSIO), combining the artificial bee colony algorithm and the beluga whale optimization algorithm, using the pile strength prediction model, the optimal parameter setting is found, and the control instruction is generated according to the optimal parameter setting, and the high-pressure rotary jet pile construction control is completed;

[0010] S4. Continuously acquiring real-time construction effect parameters of high-pressure rotary jet piles, and comparing with the prediction results, according to the comparison results, updating the pile strength prediction model parameters and the hybrid swarm optimization algorithm parameters, and completing the high-pressure rotary jet pile construction optimization control.

[0011] In step S1, the real-time construction data of high-pressure rotary jet piles includes high-pressure slurry pump pressure, slurry delivery volume, jet start and stop time, drill rod lifting speed, and drill rod rotating speed.

[0012] The pretreatment specifically includes: in the data collection process, combining data filtering and outlier rejection method to avoid the influence of instantaneous pulsation, flow fluctuation or pressure pulse on accuracy; using median filtering, mean filtering or Kalman filtering and other signal processing methods to improve the stability and reliability of the data.

[0013] Further, the pretreatment specifically includes the following steps:

[0014] From the real-time construction data of high-pressure rotary jet piles, the jet pressure, slurry flow, drill rod lifting speed and drill rod rotating speed are extracted, and the abnormal values with pressure or flow of 0 are rejected;

[0015] 5-point or 7-point moving average filtering is used to smooth the data fluctuation;

[0016] The features are scaled to a distribution with a mean of 0 and a variance of 1 by Z-score standardization to obtain pretreatment data.

[0017] In step S2, the following steps are specifically included:

[0018] Based on the XGBoost regression model, an initial pile strength prediction model is constructed;

[0019] Based on historical construction parameters and pile strength data, an initial pile strength prediction model is trained to obtain the pile strength prediction model.

[0020] Step S2 specifically comprises:

[0021] Based on the XGBoost regression model, an initial pile strength prediction model is constructed, which is represented by the following formula:

[0022]

[0023] Wherein, Strength is the model output, indicating the predicted pile strength value; P is the jetting pressure; F is the slurry flow; is the drill pipe lifting speed; is the drill pipe rotation speed; is the model error; is the XGBoost regression model processing;

[0024] Based on historical construction parameters and pile strength data, an initial pile strength prediction model is trained, the model generalization ability is evaluated by using the root mean square error and the determination coefficient, the model hyperparameters are adjusted by using random search, the hyperparameters include the number of trees, the maximum depth of the tree, the learning rate, and the sample sampling ratio, and the regression tree is constructed by iteratively optimizing the objective function in the training process.

[0025] Through random search cross-validation, the model automatically evaluates the effect of different hyperparameter combinations through K-fold cross-validation, selects the optimal hyperparameters, and obtains the pile strength prediction model.

[0026] The calculation formula of the root mean square error RMSE is as follows:

[0027]

[0028] Wherein, is the true pile strength corresponding to the i-th historical construction data; is the pile strength value predicted by the model based on the i-th historical construction data; n is the total number of historical construction parameters and pile strength data.

[0029] The calculation formula of the determination coefficient R2 is as follows:

[0030]

[0031] Wherein, is the average value of the pile strength in the historical construction parameter and pile strength data.

[0032] Step S3 comprises the following steps:

[0033] ​Based on the hybrid swarm optimization algorithm, the construction parameters of the high-pressure jet grouting pile are adjusted by combining the artificial bee colony algorithm and the beluga whale optimization algorithm, and the optimal parameter setting is obtained by optimizing the combination of the pile body strength prediction model.

[0034] The optimal parameter setting is converted into a control instruction to complete the real-time control of the high-pressure jet grouting pile construction.

[0035] Step S3 is specifically:

[0036] The hybrid swarm optimization algorithm is used to take the vector composed of the jetting pressure, the slurry flow, the drill pipe lifting speed and the drill pipe rotating speed as a solution, and the optimal parameter setting is obtained by combining the pile body strength prediction model.

[0037] The objective function of the hybrid swarm optimization algorithm is to minimize the prediction error, so that the difference between the predicted pile body strength obtained by the optimal parameter setting and the target strength is minimized, and the objective function is expressed by the following formula:

[0038]

[0039] Wherein, The error between the predicted pile body strength and the target strength; The predicted pile body strength obtained by the pile body strength prediction model; The target strength of the pile body;

[0040] The specific process of the hybrid swarm optimization algorithm is as follows:

[0041] Initialize a random solution space, wherein each solution represents a vector composed of four parameters of jetting pressure, slurry flow, drill pipe lifting speed and drill pipe rotating speed;

[0042] Set the constraint conditions of each parameter to meet the actual production requirements;

[0043] Randomly generate solutions within the given constraint condition range;

[0044] Use the pile body strength prediction model to predict each solution and calculate its fitness value; the fitness value represents the contribution of each solution to the objective function;

[0045] Optimize and iterate each solution using the artificial bee colony algorithm and the beluga whale optimization algorithm;

[0046] The optimization and iteration of each solution based on the artificial bee colony algorithm is specifically: each solution is locally optimized, a new solution is generated by neighborhood search, the fitness is updated and selected, the number of trials of the solution is updated, and the unimproved solution is reinitialized;

[0047] The optimization iteration of each solution based on the beluga whale optimization algorithm is specifically: taking the local optimal solution found in the last iteration as the leader solution; using the surrounding and spiral updating strategy of the beluga whale optimization algorithm to perform global search on the solution; evaluating and updating the position of the solution to further optimize the fitness;

[0048] A formation-adaptive artificial bee colony algorithm and a beluga whale optimization algorithm switching ratio are introduced to dynamically adjust the global and local search weights;

[0049] The formation-adaptive artificial bee colony algorithm and the beluga whale optimization algorithm switching ratio The following formula is used to represent:

[0050]

[0051] Wherein, is a real-time formation complexity index; k is an adjustment factor; is the standard deviation of the injection pressure; is the standard deviation of the slurry flow; is the average value of the drill pipe lifting speed; is the average value of the drill pipe rotating speed;

[0052] The first iteration of the iteration is an iteration optimization based on the artificial bee colony algorithm, and the remaining iterations are optimized based on the beluga whale optimization algorithm;

[0053] The optimal parameter setting is obtained by combining the artificial bee colony algorithm and the beluga whale optimization algorithm, and the high-pressure rotary jet pile construction parameters are adjusted according to the optimal parameter setting to complete the real-time control of the high-pressure rotary jet pile construction.

[0054] The control instruction is specifically converted as follows: using a PID control algorithm, based on real-time construction parameter data and optimal parameter settings, an adjustment value is calculated, and based on the adjustment value, the high-pressure rotary jet pile construction parameters are adjusted to complete the real-time control of the high-pressure rotary jet pile construction.

[0055] The PID control algorithm is represented by the following formula:

[0056]

[0057] Wherein, is the output of the PID controller; is the proportional parameter; is the integral parameter; is the integral gain parameter; is the error of the control parameter at time t, which is calculated using the following formula: ; wherein, is the optimal parameter value obtained by the hybrid swarm intelligence optimization algorithm; Real-time measurement values acquired for the sensor.

[0058] Step S4 comprises the following steps:

[0059] Real-time construction parameter data is acquired;

[0060] Based on the pile body strength prediction model, a real-time pile body strength prediction value is calculated, and error judgment is performed on the real-time pile body strength prediction value and the preset target pile body strength;

[0061] When the error is less than the threshold value, it is determined to be up to standard;

[0062] When the error is greater than the threshold value, it is determined to be abnormal, and abnormal analysis is triggered;

[0063] When a certain parameter setting significantly improves the pile body strength, the weight of the parameter setting is enhanced by a hybrid swarm intelligent optimization algorithm, so that it is preferentially called under similar stratum conditions;

[0064] When a certain parameter setting triggers abnormal analysis, a new solution is searched by an artificial bee colony algorithm, the parameter constraint range is adjusted and updated to the model training set, and the prediction accuracy is improved.

[0065] The abnormal analysis specifically comprises:

[0066] It is checked whether the stratum has a mutation and whether the equipment has a fault, and the checking result is recorded and transmitted to the user;

[0067] The standard deviation of the jetting pressure and the slurry flow fluctuation is calculated, and if it exceeds the threshold value, it is determined to be over jetting or under jetting, and the determination result is recorded and transmitted to the user;

[0068] The construction efficiency is calculated, it is determined whether there is a coordination problem between the drill rod lifting speed and the rotation speed, and the determination result is recorded and transmitted to the user.

[0069] The construction efficiency is calculated using the following formula:

[0070]

[0071] Wherein, is a first weight coefficient; is a second weight coefficient; is the construction efficiency, indicating the cooperation efficiency of the drill rod lifting speed and the rotation speed, and the greater the value, the better the coordination.

[0072] The application also provides a system for implementing the high-pressure rotary jet pile construction optimization control method, comprising an environmental parameter acquisition module, an optimization control module and an execution layer control module.

[0073] The environmental parameter acquisition module acquires construction data of each area in real time, in the data acquisition process, in combination with data filtering and outlier rejection method, influences of instantaneous pulsation, flow fluctuation or pressure pulse on system control accuracy are avoided, signal processing methods such as median filtering, mean filtering or Kalman filtering are adopted, and the stability and reliability of data are improved;

[0074] The optimization control module is used for receiving construction data from the environmental parameter acquisition module, analyzing and predicting the construction data based on an algorithm, obtaining a prediction result, and optimizing control decisions of construction parameters in the high-pressure jet grouting pile control system based on the prediction result, and generating control instructions for the execution mechanism;

[0075] The execution layer control module is used for receiving the control instructions and dynamically adjusting the construction parameters of the high-pressure jet grouting device according to the control instructions, so as to complete the high-pressure jet grouting pile construction optimization control.

[0076] The application discloses a high-pressure jet grouting pile construction optimization control method and system, solves the problems of parameter setting deviation and uneven pile quality caused by traditional construction relying on manual experience, is especially suitable for high-precision construction requirements under complex geological conditions (such as karst areas and deep silt layers), significantly improves the stability of pile strength, construction efficiency and quality controllability, and reduces material waste and human error risk. BRIEF DESCRIPTION OF DRAWINGS

[0077] Figure 1 The application discloses a high-pressure jet grouting pile construction optimization control method and system, solves the problems of parameter setting deviation and uneven pile quality caused by traditional construction relying on manual experience, is especially suitable for high-precision construction requirements under complex geological conditions (such as karst areas and deep silt layers), significantly improves the stability of pile strength, construction efficiency and quality controllability, and reduces material waste and human error risk.

[0078] Figure 2 The application discloses a high-pressure jet grouting pile construction optimization control method and system, solves the problems of parameter setting deviation and uneven pile quality caused by traditional construction relying on manual experience, is especially suitable for high-precision construction requirements under complex geological conditions (such as karst areas and deep silt layers), significantly improves the stability of pile strength, construction efficiency and quality controllability, and reduces material waste and human error risk. DETAILED DESCRIPTION

[0079] The application provides a high-pressure jet grouting pile construction optimization control method, and a flowchart as shown in the figure comprises the following steps: Figure 1

[0080] S1. Collect high-pressure jet grouting pile historical construction data; pretreat the collected high-pressure jet grouting pile real-time construction data to obtain pretreated data;

[0081] In step S1, the high-pressure jet grouting pile real-time construction data comprises high-pressure slurry pump pressure, slurry delivery volume, jet start and stop time, drill rod lifting speed and drill rod rotating speed.

[0082] The pretreatment is specifically: in the data acquisition process, in combination with data filtering and outlier rejection method, influences of instantaneous pulsation, flow fluctuation or pressure pulse on precision are avoided; signal processing methods such as median filtering, mean filtering or Kalman filtering are adopted, and the stability and reliability of data are improved.

[0083] ​Further, the preprocessing specifically includes the following steps:

[0084] From the real-time construction data of high-pressure rotary jet pile, the jetting pressure, slurry flow, drill pipe lifting speed and drill pipe rotating speed are extracted, and the abnormal values with pressure or flow of 0 are removed;

[0085] The data fluctuation is smoothed by 5-point or 7-point moving average filtering;

[0086] Each feature is scaled to a distribution with a mean of 0 and a variance of 1 by Z-score standardization to obtain preprocessed data.

[0087] S2. Based on the machine learning algorithm, the preprocessed data is used as training data to construct a pile strength prediction model;

[0088] In step S2, the following steps are specifically included:

[0089] Based on the XGBoost regression model, an initial pile strength prediction model is constructed;

[0090] Based on the historical construction parameters and pile strength data, the initial pile strength prediction model is trained to obtain the pile strength prediction model.

[0091] Step S2 is specifically:

[0092] Based on the XGBoost regression model, an initial pile strength prediction model is constructed, which is represented by the following formula:

[0093]

[0094] Where Strength is the model output, representing the predicted pile strength value; P is the jetting pressure; F is the slurry flow; is the drill pipe lifting speed; is the drill pipe rotating speed; is the model error; is the XGBoost regression model processing;

[0095] Based on the historical construction parameters and pile strength data, the initial pile strength prediction model is trained, the model generalization ability is evaluated by root mean square error and determination coefficient; the model hyperparameters are adjusted by random search; the hyperparameters include: the number of trees, the maximum depth of the tree, the learning rate, and the sample sampling ratio; the regression tree is constructed by iteratively optimizing the objective function in the training process;

[0096] Through random search cross-validation, the model automatically evaluates the effect of different hyperparameter combinations through K-fold cross-validation, selects the optimal hyperparameters, and obtains the pile strength prediction model;

[0097] The calculation formula of the root mean square error RMSE is as follows:

[0098]

[0099] wherein, is the real pile strength corresponding to the i-th historical construction data; is the pile strength value predicted based on the i-th historical construction data model; n is the total number of historical construction parameters and pile strength data;

[0100] The decision coefficient The calculation formula is as follows:

[0101]

[0102] wherein, is the average pile strength in the historical construction parameter and pile strength data.

[0103] S3. Based on the hybrid swarm optimization algorithm, combining the artificial bee colony algorithm and the beluga optimization algorithm, the pile strength prediction model is used to find the optimal parameter setting, and control instructions are generated according to the optimal parameter setting to complete the high-pressure jet grouting pile construction control;

[0104] Step S3 includes the following steps:

[0105] Based on the hybrid swarm optimization algorithm, combining the artificial bee colony algorithm and the beluga optimization algorithm, the high-pressure jet grouting pile construction parameters are adjusted, and then combined with the pile strength prediction model to obtain the optimal parameter setting;

[0106] The optimal parameter setting is converted into control instructions to complete the real-time control of high-pressure jet grouting pile construction.

[0107] Step S3 is specifically:

[0108] The hybrid swarm optimization algorithm is used to form a vector composed of four parameters of jetting pressure, slurry flow, drill pipe lifting speed and drill pipe rotating speed as a solution, and then combined with the pile strength prediction model to find the optimal parameter setting;

[0109] The objective function of the hybrid swarm optimization algorithm is to minimize the prediction error, so that the pile prediction strength obtained by the optimal parameter setting after optimization through the pile strength prediction model is the minimum difference from the target strength, and the objective function is expressed using the following formula:

[0110]

[0111] wherein, The error between the pile prediction strength and the target strength; is the pile prediction strength obtained through the pile strength prediction model; is the target strength of the pile;

[0112] The specific process of the mixed group optimization algorithm is as follows:

[0113] Initialize a random solution space, wherein each solution represents a vector composed of four parameters of jet pressure, slurry flow, drill pipe lifting speed and drill pipe rotating speed;

[0114] Set the constraint conditions of each parameter to meet the actual production requirements;

[0115] Randomly generate solutions within the given constraint condition range;

[0116] Use the pile body strength prediction model to predict each solution and calculate its fitness value; the fitness value represents the contribution of each solution to the objective function;

[0117] Use the artificial bee colony algorithm and the whale optimization algorithm to optimize and iterate each solution;

[0118] The optimization and iteration of each solution based on the artificial bee colony algorithm are as follows:

[0119] Locally optimize each solution, generate a new solution through neighborhood search, update the fitness and select, update the number of trials of the solution, and reinitialize the solution that is not improved;

[0120] The optimization and iteration of each solution based on the whale optimization algorithm are as follows:

[0121] Take the locally optimal solution found in the last iteration as the leader solution; use the surrounding and spiral update strategy of the whale optimization algorithm to perform global search on the solution; evaluate and update the position of the solution to further optimize the fitness;

[0122] Introduce a formation-adaptive artificial bee colony algorithm and whale optimization algorithm switching ratio to dynamically adjust the global and local search weights;

[0123] The formation-adaptive artificial bee colony algorithm and whale optimization algorithm switching ratio Use the following formula to represent:

[0124]

[0125] Wherein, is a real-time formation complexity index; k is a regulation factor; is the standard deviation of the jet pressure; is the standard deviation of the slurry flow; is the average of the drill pipe lifting speed; is the average of the drill pipe rotating speed;

[0126] The first iteration of the iteration is the iteration optimization based on the artificial bee colony algorithm, and the remaining iterations are optimized based on the whale optimization algorithm.

[0127] The optimal parameter setting is obtained by combining the artificial bee colony algorithm and the beluga whale optimization algorithm, and the high-pressure rotary jet pile construction parameters are adjusted according to the control instruction converted from the optimal parameter setting to complete real-time control of high-pressure rotary jet pile construction.

[0128] The control instruction is specifically converted as follows: using a PID control algorithm, based on real-time construction parameter data and optimal parameter settings, an adjustment value is calculated, and based on the adjustment value, the high-pressure rotary jet pile construction parameters are adjusted to complete real-time control of high-pressure rotary jet pile construction.

[0129] The PID control algorithm is expressed by the following formula:

[0130]

[0131] wherein, is the output of the PID controller; is the proportional parameter; is the integral parameter; is the integral gain parameter; is the error of the control parameter at time t, which is calculated using the following formula: ; wherein, is the optimal parameter value obtained by the hybrid swarm intelligence optimization algorithm; is the real-time measurement value obtained by the sensor.

[0132] S4. Continuously obtain the real-time construction effect parameters of the high-pressure rotary jet pile, and compare them with the predicted results, and according to the comparison results, update the pile strength prediction model parameters and the hybrid swarm optimization algorithm parameters to complete the optimization control of the high-pressure rotary jet pile construction.

[0133] Step S4 includes the following steps:

[0134] Real-time construction parameter data is obtained;

[0135] Based on the pile strength prediction model, the real-time pile strength prediction value is calculated, and an error judgment is made between the real-time pile strength prediction value and the preset target pile strength;

[0136] When the error is less than the threshold, it is determined to be up to standard;

[0137] When the error is greater than the threshold, it is determined to be abnormal, triggering abnormal analysis;

[0138] When a certain parameter setting significantly improves the pile strength, the weight of the parameter setting is enhanced by the hybrid swarm intelligence optimization algorithm, so that it is preferentially called under similar stratum conditions;

[0139] When a certain parameter setting triggers abnormal analysis, a new solution is searched by the artificial bee colony algorithm, the parameter constraint range is adjusted and updated to the model training set to improve the prediction accuracy.

[0140] The anomaly analysis is specifically:

[0141] checking whether the stratum mutates or the equipment malfunctions, and recording and delivering the checking result to the user;

[0142] calculating the standard deviation of the jetting pressure and the slurry flow fluctuation, and determining over jetting or under jetting if the standard deviation exceeds a threshold value, and recording and delivering the determination result to the user;

[0143] calculating the construction efficiency, determining whether there is a coordination problem between the drill rod lifting speed and the rotating speed, and recording and delivering the determination result to the user.

[0144] The construction efficiency is calculated using the following formula:

[0145]

[0146] wherein, is a first weight coefficient; is a second weight coefficient; is the construction efficiency, indicating the matching efficiency of the drill rod lifting speed and the rotating speed, and the greater the value, the better the coordination.

[0147] The application also provides a system for realizing the high-pressure rotary jet pile construction optimization control method, a structure diagram of which is shown in the figure, comprising an environmental parameter acquisition module, an optimization control module and an execution layer control module. Figure 2

[0148] The environmental parameter acquisition module acquires the construction data of each region in real time, and in the data acquisition process, the influence of instantaneous pulsation, flow fluctuation or pressure pulse on the control accuracy of the system is avoided by combining data filtering and outlier rejection methods, and the stability and reliability of the data are improved by using signal processing methods such as median filtering, mean filtering or Kalman filtering.

[0149] The environmental parameter acquisition module is distributed in different regions of the high-pressure rotary jet device, including the high-pressure pump outlet, the end of the jetting pipe (the front section of the drill rod or the nozzle), the outlet pipe of the grouting pump, the sectional timer and other devices.

[0150] The environmental parameter acquisition module comprises a pressure sensor, a flow meter, an encoder, a frequency converter and a timer, which are used to acquire the construction data of each region in real time, and the construction data comprises high-pressure slurry pump pressure, slurry delivery volume, jetting start and stop time, drill rod lifting speed, drill rod rotating speed and the like.

[0151] ​The pressure sensor can be a piezoresistive, strain gauge or capacitive pressure sensor, and should be able to withstand the maximum pressure within the flow valve operating range to avoid sensor damage or excessive measurement error. The flow measurement can be performed using a differential pressure flowmeter, electromagnetic flowmeter or ultrasonic flowmeter, and the specific selection can be determined according to the pipe diameter and the required measurement accuracy.

[0152] Each sensor interacts with the controller through wireless communication for subsequent calculation and adjustment. Through the above steps, the data of flow, pressure, drill rod lifting speed and drill rod rotating speed can be accurately and real-time acquired, providing accurate and reliable basic data for subsequent prediction and intelligent adjustment.

[0153] The optimization control module is used to receive construction data from the environmental parameter acquisition module, analyze and predict the construction data based on an algorithm, obtain a prediction result, and optimize the control decision of the construction parameter in the high-pressure jet grouting pile control system based on the prediction result to generate a control instruction for the actuator.

[0154] From the real-time data of the sensors in each area of the high-pressure jet grouting pile construction system, pressure, flow, lifting speed, rotating speed and other parameter data are obtained, and are processed and predicted through a machine learning model to generate optimal control parameters of the jetting amount, pressure and time in each construction process.

[0155] After removing the abnormal values with zero pressure or flow from the sensor data, a 5-point moving average filter is used to smooth the data fluctuations (5 points are empirical values that need to be adjusted according to the data characteristics. If the data noise is large (such as electromagnetic interference), a 7-point window can be tried), and each feature is scaled to a distribution with a mean of 0 and a variance of 1 through Z-score standardization. Normalization can eliminate the dimensional differences of MPa and cm / min, and avoid the model biasing towards large numerical features caused by the difference in numerical magnitude.

[0156] By real-time collection of sensor data (such as pressure, flow, etc.) and combination of the predicted pile strength through a regression model, the HSIO optimization algorithm can dynamically adjust the jetting pressure and flow in real time to ensure that the pressure and flow of each jetting are optimal, thereby ensuring that the strength of the pile body reaches the expected target. When the predicted pile strength is lower than the target value, the HSIO optimization algorithm automatically adjusts the jetting parameters to dynamically optimize the pressure, flow, lifting speed and rotating speed, etc. to ensure that the pile strength during construction meets the expected standard, avoiding the lag and human error of manual intervention.

[0157] The execution layer control module is used to receive the control instruction and dynamically adjust the construction parameters of the high-pressure jet grouting device according to the control instruction to complete the high-pressure jet grouting pile construction optimization control.

[0158] By controlling the execution layer equipment, adjusting the injection pressure, flow rate, lifting speed and rotation speed, ensuring the smooth construction, wherein the execution layer equipment includes control valves (electric or pneumatic) for adjusting the injection pressure and flow rate. Control valves for adjusting the injection pressure and flow rate. Control valves receive control signals from the PID control system, adjust the valve opening in real time, so as to accurately control the injection pressure and flow rate. Hydraulic actuators adjust the lifting and rotation actions of the equipment according to the PID control signals, to ensure the accuracy of each operation during pile construction. As the core control unit of the system, the PLC or SCADA system is responsible for receiving the construction parameters output by the optimization control module, and converting these parameters into electrical signals through the PID control algorithm, and sending them to the actuators. The PLC or SCADA system is also responsible for continuously monitoring the running state of the execution equipment, to ensure that all parameters during construction always meet the optimization requirements.

[0159] The execution layer control module receives control instructions through wireless communication, which includes target construction parameters generated by the HSIO optimization algorithm, such as target injection pressure, injection flow rate, lifting speed and rotation speed, etc. After the control instructions are parsed by the microprocessor built-in the execution layer control module, they are converted into operable control signals, and sent to the electric regulating valve, hydraulic actuator and variable frequency circulating pump in each area. These control signals will ensure that each construction parameter is adjusted according to the target value output by the HSIO optimization algorithm.

[0160] The execution layer control module is to convert the target parameters (such as injection pressure, injection flow rate, lifting speed, rotation speed, etc.) output by the HSIO optimization algorithm into specific control signals that can be recognized by the execution equipment. The electric regulating valve is installed in the shotcrete pipeline, which automatically adjusts the opening of the valve according to the target injection pressure and flow rate, accurately controls the pressure and flow rate of the shotcrete system, and ensures that the injection parameters meet the preset standards during construction. At the same time, the hydraulic actuator adjusts the running state of the lifting equipment and the rotation equipment according to the target lifting speed and rotation speed, to ensure the stability and accuracy of pile construction.

[0161] During construction, the execution layer control module monitors the key parameters of the shotcrete system in real time through the equipped sensing unit, including injection pressure, flow rate, lifting speed, rotation speed, etc. The sensing unit feeds back these real-time data to the control module for real-time evaluation.

[0162] According to these feedback data, the execution layer control module calculates the deviation between the current construction effect and the target by comparing with the target parameters provided by the HSIO optimization algorithm, and generates new control instructions. These new control instructions will be transmitted to the execution equipment again for dynamic adjustment and optimization of construction parameters. The dynamic adjustment of the execution layer control module relies on the PID control algorithm. According to the error between the real-time measured parameters and the target parameters given by the HSIO optimization algorithm, the PID control algorithm calculates the adjustment amount. Specifically, the proportional control (P) adjusts in time according to the current deviation, the integral control (I) eliminates the steady-state error of the system, and the derivative control (D) predicts the trend of the system according to the deviation rate, thereby suppressing overshoot and fluctuations. Through PID control, the injection pressure, flow rate, lifting speed and rotation speed of the execution equipment can be accurately adjusted to ensure that all parameters during construction are always in the optimal state. During the adjustment of the execution equipment, the system will continuously monitor the feedback data to ensure that each operation minimizes the deviation, so that the construction parameters such as injection pressure, flow rate, lifting speed and rotation speed are as close as possible to the target values given by the HSIO optimization algorithm.

[0163] The whole process forms a closed-loop control system, and the execution layer control module continuously adjusts the equipment state and evaluates the construction effect in real time to ensure that each control adjustment meets the expected target, and finally realizes the stability and strength requirements of the pile body.

[0164] The following is an embodiment of the algorithm of the present optimization control module:

[0165] Based on the same construction data set, the present method is compared with PSO+linear regression and random forest in terms of accuracy and reliability.

[0166] The three models all input control parameters (P, F, , ), and output predicted pile strength. PSO+linear regression optimizes the coefficients through PSO iteration, outputs the optimal coefficients, and generates the prediction results. Random forest directly uses the default parameters to train the random forest model, and outputs the predicted pile strength; the present method first pre-trains XGBoost, optimizes the HSIO parameters, outputs the optimal parameter combination, and generates the prediction results using the optimized parameter combination.

[0167] The comparison indicators of the prediction results of the three models are shown in Table 1.

[0168] Table 1 Comparison indicators of prediction results of three models

[0169]

[0170] The simulation result shows that the method has the smallest mean square error and the smallest pressure fluctuation value, has better stability and reliability, reduces the risk of over-spraying or under-spraying, and ensures that the strength of the pile body is closer to the design target.

Claims

1. A high-pressure jet grouting pile construction optimization control method, characterized in that, The method comprises the following steps: S1. Collecting high-pressure jet grouting pile historical construction data; preprocessing the collected real-time construction data of the high-pressure jet grouting pile to obtain pretreated data; S2. Based on a machine learning algorithm, taking the pretreated data as training data to construct a pile strength prediction model; S3. Based on a hybrid swarm optimization algorithm, combining an artificial bee colony algorithm and a beluga whale optimization algorithm, using the pile strength prediction model to find the optimal parameter setting, and generating a control instruction according to the optimal parameter setting to complete the high-pressure jet grouting pile construction control; S4. Continuously acquiring real-time construction effect parameters of the high-pressure jet grouting pile, and comparing them with the prediction results, updating the pile strength prediction model parameters and the hybrid swarm optimization algorithm parameters according to the comparison results to complete the high-pressure jet grouting pile construction optimization control; Step S3 comprises the following steps: Based on the hybrid swarm optimization algorithm, combining the artificial bee colony algorithm and the beluga whale optimization algorithm, adjusting the high-pressure jet grouting pile construction parameters, and then combining them with the pile strength prediction model to obtain the optimal parameter setting; The optimal parameter setting is converted into a control instruction to complete the real-time control of the high-pressure jet grouting pile construction; Step S3 specifically comprises: Using the hybrid swarm optimization algorithm to take a vector composed of the jetting pressure, the slurry flow, the drill rod lifting speed and the drill rod rotating speed as a solution, and then combining it with the pile strength prediction model to find the optimal parameter setting; The objective function of the hybrid swarm optimization algorithm is to minimize the prediction error, so that the pile prediction strength obtained by the optimal parameter setting through the pile strength prediction model is closest to the target strength, and the objective function is represented by the following formula: wherein, the error between the pile body predicted strength and the target strength; the pile body predicted strength obtained through the pile body strength prediction model; the target strength of the pile body; and n is the total number of samples. The specific process of the hybrid swarm optimization algorithm is as follows: Initialize a random solution space, wherein each solution represents a vector composed of the jetting pressure, the slurry flow, the drill rod lifting speed and the drill rod rotating speed; Set the constraint conditions of each parameter to meet the actual production requirements; Randomly generate a solution within the given constraint condition range; Use the pile strength prediction model to predict each solution and calculate its fitness value; the fitness value represents the contribution of each solution to the objective function; Use the artificial bee colony algorithm and the beluga whale optimization algorithm to optimize and iterate each solution; The artificial bee colony algorithm optimizes and iterates each solution specifically as follows: locally optimize each solution, generate a new solution through neighborhood search, update the fitness and select, update the number of trials of the solution, and reinitialize the solution that is not improved; The beluga whale optimization algorithm optimizes and iterates each solution specifically as follows: take the locally optimal solution found in the last iteration as the leader solution; use the surrounding and spiral update strategies of the beluga whale optimization algorithm to perform global search on the solution; evaluate and update the position of the solution to further optimize the fitness; Introduce a stratum-adaptive artificial bee colony algorithm and beluga whale optimization algorithm switching ratio to dynamically adjust the global and local search weights; The switching proportion of the stratum-adaptive artificial bee colony algorithm and the beluga whale optimization algorithm Is expressed using the following equation: wherein, is a real-time formation complexity index; k is an adjustment factor; is a standard deviation of the injection pressure; is a standard deviation of the slurry flow rate; is a mean value of the drill pipe hoisting speed; is a mean value of the drill pipe rotation speed; The iterations are preceded by a pre-iteration The pre-iteration is an artificial bee colony algorithm-based iteration optimization, and the remaining iterations are sperm whale optimization algorithm-based optimization. Combine the artificial bee colony algorithm and the beluga whale optimization algorithm to obtain the optimal parameter setting, convert it into a control instruction to adjust the high-pressure jet grouting pile construction parameters, and complete the real-time control of the high-pressure jet grouting pile construction; The control instruction is specifically transformed as follows: using a PID control algorithm, based on real-time construction parameter data and optimal parameter settings, an adjustment value is calculated, and based on the adjustment value, the construction parameters of the high-pressure jet grouting pile are adjusted to complete real-time control of the high-pressure jet grouting pile construction; The PID control algorithm is expressed by the following formula: wherein, is the output of the PID controller; is the proportional parameter; is the integral parameter; is the integral gain parameter; is the error of the control parameter at time t, calculated using the following equation: wherein, is the optimal parameter value obtained by the hybrid swarm intelligence optimization algorithm; is the real-time measurement value obtained by the sensor.

2. The high-pressure jet grouting pile construction optimization control method according to claim 1, characterized in that, In step S1, the real-time construction data of the high-pressure jet grouting pile includes high-pressure slurry pump pressure, slurry delivery volume, jet start and stop time, drill rod lifting speed, and drill rod rotation speed. The preprocessing is specifically as follows: in the data acquisition process, data filtering and outlier rejection methods are combined to avoid the influence of instantaneous pulsation, flow fluctuation or pressure pulse on accuracy; median filtering, mean filtering or Kalman filtering signal processing methods are used to improve the stability and reliability of the data.

3. The high-pressure jet grouting pile construction optimization control method according to claim 2, characterized in that, The preprocessing specifically includes the following steps: From the real-time construction data of the high-pressure jet grouting pile, the jet pressure, slurry flow, drill rod lifting speed and drill rod rotation speed are extracted, and the abnormal values with pressure or flow of 0 are rejected; 5-point or 7-point moving average filtering is used to smooth data fluctuations; Through Z-score standardization, each feature is scaled to a distribution with a mean of 0 and a variance of 1 to obtain preprocessed data.

4. The high-pressure jet grouting pile construction optimization control method of claim 1, wherein, In step S2, the following steps are specifically included: Based on the XGBoost regression model, an initial pile strength prediction model is constructed; Based on historical construction parameter and pile strength data, the initial pile strength prediction model is trained to obtain the pile strength prediction model.

5. The high-pressure jet grouting pile construction optimization control method according to claim 4, characterized in that, Step S2 is specifically as follows: Based on the XGBoost regression model, an initial pile strength prediction model is constructed, which is expressed by the following formula: wherein, Strength is the model output, representing the predicted pile strength value; P is the jetting pressure; F is the slurry flow rate; is the drill pipe lifting speed; is the drill pipe rotating speed; is the model error; is the XGBoost regression model processing; Based on historical construction parameter and pile strength data, the initial pile strength prediction model is trained, and the model generalization ability is evaluated using root mean square error (RMSE) and coefficient of determination (R2); random search is used to adjust the model hyperparameters; the hyperparameters include the number of trees, the maximum depth of the tree, the learning rate, and the sample sampling ratio; the regression tree is constructed by iteratively optimizing the objective function in the training process; Through random search cross-validation, the model automatically evaluates the effects of different hyperparameter combinations through K-fold cross-validation, selects the optimal hyperparameters, and obtains the pile strength prediction model; The calculation formula of the root mean square error (RMSE) is as follows: wherein, is the real pile strength corresponding to the i-th historical construction data; is the pile strength value predicted based on the i-th historical construction data model; n is the total number of historical construction parameter and pile strength data samples; The decision coefficient The calculation formula is as follows: wherein, is the average pile strength in the pile strength data for historical construction parameters.

6. The high-pressure jet grouting pile construction optimization control method of claim 1, wherein, Step S4 includes the following steps: Real-time construction parameter data is obtained; Based on the pile strength prediction model, real-time pile strength prediction values are calculated, and error judgment is performed with the preset target pile strength; When the error is less than the threshold, it is determined to be up to standard; When the error is greater than the threshold, it is determined to be abnormal, triggering abnormal analysis; When a certain parameter setting significantly improves the pile strength, the weight of the parameter setting is enhanced through a hybrid swarm intelligence optimization algorithm, so that it is preferentially called in similar stratum conditions; When a certain parameter setting triggers abnormal analysis, a new solution is searched through an artificial bee colony algorithm, the parameter constraint range is adjusted, and it is updated to the model training set.

7. The high-pressure jet grouting pile construction optimization control method according to claim 6, characterized in that, The abnormal analysis is specifically as follows: Check whether the stratum has mutated or the equipment has failed, and record and deliver the check results to the user; Calculate the standard deviation of the jet pressure and slurry flow fluctuation, and if it exceeds the threshold, determine that it is overjetting or underjetting, and record and deliver the determination results to the user; Calculate construction efficiency, determine whether there is a synergy problem between the lifting speed and the rotating speed of the drill rod, and record and pass the determination result to the user; The construction efficiency is calculated using the following formula: wherein, is a first weight coefficient; is a second weight coefficient; is a drill pipe hoisting speed; is a drill pipe rotation speed; is a construction efficiency, indicating the matching efficiency of the drill pipe hoisting speed and the rotation speed, the larger the value, the better the synergy.

8. A system for implementing the high-pressure jet grouting pile construction optimization control method of any one of claims 1-7, characterized in that, It comprises an environmental parameter acquisition module, an optimization control module, and an execution layer control module. The environmental parameter acquisition module acquires the construction data of each region in real time. During the data acquisition process, the data is first processed by combining data filtering and outlier rejection methods, and is processed by using a median filtering, mean filtering, or Kalman filtering signal processing method. The optimization control module is used to receive the construction data from the environmental parameter acquisition module, analyze and predict the construction data based on an algorithm, obtain a prediction result, and optimize the control decision of the construction parameters in the high-pressure rotary jet pile control system based on the prediction result to generate a control instruction for the execution mechanism. The execution layer control module is used to receive the control instruction and dynamically adjust the construction parameters of the high-pressure rotary jet device according to the control instruction to complete the optimization control of the high-pressure rotary jet pile construction.

Images