High-pressure jet grouting pile construction optimization control method and system

Abstract

The invention discloses a high-pressure jet grouting pile construction optimization control method and system, and belongs to the technical field of pile foundation construction. According to the method, the pile body strength is predicted based on the XGBoost regression model, the HSIO hybrid swarm intelligent optimization algorithm is combined with the artificial bee colony algorithm and the seathead optimization algorithm, the construction parameter combination is dynamically adjusted, and optimal matching of the target strength and the real-time construction state is achieved. The method solves the problems of parameter setting deviation and non-uniform pile forming quality caused by the fact that traditional construction depends on artificial experience, is particularly suitable for high-precision construction requirements under complex geological conditions (such as karst areas and deep and thick sludge layers), remarkably improves the pile body strength stability, construction efficiency and quality controllability, and reduces the construction cost. And meanwhile, material waste and personal error risks are reduced.

Description

Technical Field

[0001] This invention belongs to the field of pile foundation construction technology, specifically relating to an optimized control method and system for high-pressure jet grouting pile construction. Background Technology

[0002] Traditional methods such as riprap filling, steel casing, and drilling grouting often fail to effectively address deep silt layers and complex karst areas, leading to low construction efficiency, high costs, significant quality risks, and even safety accidents. High-pressure jet grouting technology, through high-pressure injection of cement grout, forms a strong reinforced body in soft strata, effectively filling karst cavities and enhancing the bearing capacity and stability of the foundation. The combined use of perimeter and inter-piles jet grouting creates a closed reinforcement system, further improving foundation stability and avoiding voids or uneven reinforcement problems that may occur in traditional construction methods. During construction, strict control of drilling and grouting progress ensures continuous operation for each pile, and real-time monitoring guarantees controllable construction quality. In jet grouting construction, injection pressure, injection volume, drill rod lifting speed, and drill rod rotation speed are key parameters affecting pile quality. Traditional methods rely heavily on manual experience to adjust these parameters; however, traditional methods struggle to guarantee parameter accuracy and consistency, making them susceptible to human error and resulting in unstable construction quality. Manual adjustments are susceptible to subjective interference, and parameter settings lack precision and consistency. They cannot be dynamically adjusted based on real-time construction data, making it difficult to adapt to complex and variable construction environments and easily leading to problems such as uneven spraying and insufficient pile strength. Furthermore, the complex construction environment and variations in factors such as mud density and soil characteristics further increase the uncertainty of construction.

[0003] While some automated control methods exist in the industry, their optimization capabilities are limited, failing to dynamically adjust based on real-time construction data, resulting in limited improvements in construction efficiency and quality. Existing automated control methods and systems are mainly applied to simple construction environments, struggling to meet the needs of projects with complex geological conditions and high precision requirements, and unable to flexibly adapt to unforeseen circumstances. Furthermore, the lack of a closed-loop feedback mechanism means that the correlation between construction parameters and target strength cannot be corrected in real time, making it difficult to guarantee the stability and uniformity of pile quality. Therefore, there is an urgent need for a high-pressure jet grouting pile method and system with intelligent control parameters to meet the needs of modern foundation engineering construction. Summary of the Invention

[0004] To address the shortcomings of existing technologies, one of the objectives of this invention is to provide an optimized control method for high-pressure jet grouting pile construction, so as 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 objective of this invention is to provide a system for implementing the optimized control method for high-pressure jet grouting pile construction.

[0006] This invention provides an optimized control method for high-pressure jet grouting pile construction, comprising the following steps:

[0007] S1. Collect historical construction data of high-pressure jet grouting piles; preprocess the collected real-time construction data of high-pressure jet grouting piles to obtain preprocessed data;

[0008] S2. Based on machine learning algorithms, preprocessed data is used as training data to construct a pile strength prediction model;

[0009] S3. Based on the Hybrid Swarm Optimization (HSIO) algorithm, combined with the artificial bee colony algorithm and the humpback whale optimization algorithm, the optimal parameter settings are found using the pile strength prediction model, and control commands are generated according to the optimal parameter settings to complete the construction control of high-pressure jet grouting piles.

[0010] S4. Continuously acquire real-time construction effect parameters of high-pressure jet grouting piles and compare them with the prediction results. Based on the comparison results, update the pile strength prediction model parameters and hybrid group optimization algorithm parameters to complete the optimization control of high-pressure jet grouting pile construction.

[0011] In step S1, the real-time construction data of the high-pressure jet grouting pile includes high-pressure grout pumping pressure, grout delivery volume, jetting start and end time, drill rod lifting speed, and drill rod rotation speed.

[0012] The preprocessing specifically involves: during the data acquisition process, combining data filtering and outlier removal methods to avoid the impact of instantaneous pulsations, flow fluctuations, or pressure pulses on accuracy; and employing signal processing methods such as median filtering, mean filtering, or Kalman filtering to improve the stability and reliability of the data.

[0013] Furthermore, the preprocessing specifically includes the following steps:

[0014] From the real-time construction data of high-pressure jet grouting piles, extract the jetting pressure, grout flow rate, drill rod lifting speed and drill rod rotation speed, and remove outliers with pressure or flow rate of 0;

[0015] Use a 5-point or 7-point moving average filter to smooth data fluctuations;

[0016] Z-score standardization is used to scale each feature to a distribution with a mean of 0 and a variance of 1, resulting in preprocessed data.

[0017] Step S2 specifically includes the following steps:

[0018] An initial pile strength prediction model was constructed based on the XGBoost regression model.

[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 is as follows:

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

[0022] Where Strength is the model output, representing the predicted pile strength value; P is the injection pressure; and F is the grout flow rate. Increase the drill pipe lifting speed; The drill pipe rotation speed; This refers to model error; Processed for XGBoost regression models;

[0023] Based on historical construction parameters and pile strength data, an initial pile strength prediction model is trained, and the generalization ability of the model is evaluated using root mean square error and coefficient of determination. The model hyperparameters are adjusted using random search, including the number of trees, the maximum tree depth, the learning rate, and the sample sampling ratio. A regression tree is constructed by iteratively optimizing the objective function during the training process.

[0024] Through random search cross-validation, the model automatically evaluates the effect of different hyperparameter combinations by K-fold cross-validation, selects the optimal hyperparameters, and obtains the pile strength prediction model.

[0025] The formula for calculating the root mean square error (RMSE) is as follows:

[0026] in, The actual pile strength corresponding to the i-th historical construction data; The value of the pile body is the predicted value based on the i-th historical construction data model; n is the total number of historical construction parameters and pile strength data.

[0027] The coefficient of determination The calculation formula is as follows:

[0028] in, This represents the average pile strength from historical construction parameters and pile strength data.

[0029] Step S3 includes the following steps:

[0030] Based on the hybrid swarm optimization algorithm, combined with the artificial bee colony algorithm and the humpback whale optimization algorithm, the construction parameters of the high-pressure jet grouting pile are adjusted, and then combined with the pile strength prediction model to obtain the optimal parameter settings.

[0031] The optimal parameter settings are converted into control commands to achieve real-time control of high-pressure jet grouting pile construction.

[0032] Step S3 is as follows:

[0033] The hybrid swarm optimization algorithm is used to find the optimal parameter settings by taking the vector composed of four parameters, namely injection pressure, slurry flow rate, drill pipe lifting speed and drill pipe rotation speed, and then combining it with the pile strength prediction model.

[0034] The objective function of the hybrid swarm optimization algorithm is to minimize the prediction error, so that the difference between the predicted pile strength and the target strength obtained by the pile strength prediction model through the optimal parameter settings is minimized. The objective function is expressed by the following formula:

[0035] in, Error between predicted and target strength of pile body; The predicted strength of the pile is obtained through the pile strength prediction model. The target strength of the pile body;

[0036] The specific process of the hybrid swarm optimization algorithm is as follows:

[0037] Initialize a random solution space, where each solution represents a vector consisting of four parameters: injection pressure, slurry flow rate, drill pipe lifting speed, and drill pipe rotation speed;

[0038] Set constraints for each parameter to meet actual production requirements;

[0039] Randomly generate solutions within the given constraints;

[0040] The pile strength prediction model is used to predict each solution and calculate its fitness value; the fitness value represents the contribution of each solution to the objective function.

[0041] The artificial bee colony algorithm and the humpback whale optimization algorithm are used to optimize and iterate for each solution;

[0042] The optimization iteration based on the artificial bee colony algorithm for each solution specifically involves: performing local optimization on each solution, generating new solutions through neighborhood search, updating the fitness and making a selection, updating the number of trials for the solution, and re-initializing unimproved solutions;

[0043] The optimization iteration based on the humpback whale optimization algorithm specifically involves: taking the local optimum found in the previous iteration as the leader solution; using the encirclement and spiral update strategy of the humpback whale optimization algorithm to perform a global search on the solution; evaluating and updating the position of the solution, and further optimizing the fitness;

[0044] The switching ratio between the formation-adaptive artificial bee colony algorithm and the humpback whale optimization algorithm is introduced to dynamically adjust the global and local search weights;

[0045] The switching ratio between the formation-adaptive artificial bee colony algorithm and the humpback whale optimization algorithm Express it using the following formula:

[0046] in, This represents the real-time stratigraphic complexity index; k is an adjustment factor. The standard deviation of the injection pressure; The standard deviation of the slurry flow rate; This represents the average drill pipe lifting speed. This represents the average rotational speed of the drill pipe.

[0047] The iteration before The first iteration is an iterative optimization based on the artificial bee colony algorithm, and the remaining iterations are optimizations based on the humpback whale optimization algorithm;

[0048] By combining the artificial bee colony algorithm and the humpback whale optimization algorithm, the optimal parameter settings are obtained. Based on the optimal parameter settings, control commands are converted into adjustments to the high-pressure jet grouting pile construction parameters, thus achieving real-time control of high-pressure jet grouting pile construction.

[0049] The control command conversion specifically involves: using a PID control algorithm, calculating adjustment values ​​based on real-time construction parameter data and optimal parameter settings, and adjusting the high-pressure jet grouting pile construction parameters based on the adjustment values ​​to complete real-time control of high-pressure jet grouting pile construction;

[0050] The PID control algorithm is expressed using the following formula:

[0051] in, This is the output of the PID controller; This is a proportional parameter; For integration parameters; This is the integral gain parameter; The error of the control parameters at time t is calculated using the following formula: ;in, The optimal parameter values ​​obtained by the hybrid swarm intelligence optimization algorithm; The real-time measurement values ​​acquired by the sensor.

[0052] Step S4 includes the following steps:

[0053] Real-time acquisition of construction parameter data;

[0054] Based on the pile strength prediction model, the real-time pile strength prediction value is calculated, and the error is judged by comparing it with the preset target pile strength.

[0055] When the error is less than the threshold, the standard is deemed met;

[0056] When the error exceeds the threshold, it is judged as an anomaly, and anomaly analysis is triggered;

[0057] If a parameter setting significantly improves the strength of the pile, the weight of that parameter setting is enhanced through a hybrid swarm intelligence optimization algorithm, so that it is preferentially called under similar geological conditions.

[0058] When a parameter setting triggers anomaly analysis, a new solution is locally searched using the artificial bee colony algorithm. The parameter constraint range is adjusted and updated to the model training set to improve prediction accuracy.

[0059] The anomaly analysis specifically refers to:

[0060] Check for abrupt changes in the formation and equipment malfunctions, and record and transmit the inspection results to the user;

[0061] Calculate the standard deviation of the fluctuation between injection pressure and slurry flow rate. If it exceeds the threshold, it is judged as over-spraying or under-spraying. The judgment result is recorded and transmitted to the user.

[0062] Calculate construction efficiency, determine whether there is a coordination problem between drill pipe lifting speed and rotation speed, and record and transmit the determination results to the user.

[0063] The construction efficiency is calculated using the following formula:

[0064] in, This is the first weighting coefficient; This is the second weighting coefficient; For construction efficiency, it represents the coordination efficiency between the drill pipe lifting speed and rotation speed; the larger the value, the better the coordination.

[0065] The present invention also provides a system for implementing the optimized control method for high-pressure jet grouting pile construction, including an environmental parameter acquisition module, an optimization control module, and an execution layer control module;

[0066] The environmental parameter acquisition module collects construction data from various areas in real time. During the data acquisition process, data filtering and outlier removal methods are combined to avoid the impact of instantaneous pulsations, flow fluctuations or pressure pulses on the system control accuracy. Signal processing methods such as median filtering, mean filtering or Kalman filtering are used to improve the stability and reliability of the data.

[0067] The optimization control module receives construction data from the environmental parameter acquisition module, analyzes and predicts the construction data based on algorithms, obtains prediction results, optimizes the construction parameters in the high-pressure jet grouting pile control system based on the prediction results, and generates control commands for the actuators.

[0068] The execution layer control module is used to receive control commands and dynamically adjust the construction parameters of the high-pressure jet grouting device according to the control commands, so as to complete the optimized control of high-pressure jet grouting pile construction.

[0069] This invention discloses an optimized control method and system for high-pressure jet grouting pile construction, which solves the problems of parameter setting deviation and uneven pile quality caused by the reliance on manual experience in traditional construction. It is especially suitable for high-precision construction under complex geological conditions (such as karst areas and deep silt layers), significantly improving the strength stability of the pile, construction efficiency and quality controllability, while reducing material waste and the risk of human error. Attached Figure Description

[0070] Figure 1 This is a schematic flowchart of the method of the present invention;

[0071] Figure 2 This is a schematic diagram of the system of the present invention. Detailed Implementation

[0072] This invention provides an optimized control method for high-pressure jet grouting pile construction, the flowchart of which is shown below. Figure 1 As shown, it includes the following steps:

[0073] S1. Collect historical construction data of high-pressure jet grouting piles; preprocess the collected real-time construction data of high-pressure jet grouting piles to obtain preprocessed data;

[0074] In step S1, the real-time construction data of the high-pressure jet grouting pile includes high-pressure grout pumping pressure, grout delivery volume, jetting start and end time, drill rod lifting speed, and drill rod rotation speed.

[0075] The preprocessing specifically involves: during the data acquisition process, combining data filtering and outlier removal methods to avoid the impact of instantaneous pulsations, flow fluctuations, or pressure pulses on accuracy; and employing signal processing methods such as median filtering, mean filtering, or Kalman filtering to improve the stability and reliability of the data.

[0076] Furthermore, the preprocessing specifically includes the following steps:

[0077] From the real-time construction data of high-pressure jet grouting piles, extract the jetting pressure, grout flow rate, drill rod lifting speed and drill rod rotation speed, and remove outliers with pressure or flow rate of 0;

[0078] Use a 5-point or 7-point moving average filter to smooth data fluctuations;

[0079] Z-score standardization is used to scale each feature to a distribution with a mean of 0 and a variance of 1, resulting in preprocessed data.

[0080] S2. Based on machine learning algorithms, preprocessed data is used as training data to construct a pile strength prediction model;

[0081] Step S2 specifically includes the following steps:

[0082] An initial pile strength prediction model was constructed based on the XGBoost regression model.

[0083] Based on historical construction parameters and pile strength data, an initial pile strength prediction model is trained to obtain the pile strength prediction model.

[0084] Step S2 is as follows:

[0085] Based on the XGBoost regression model, an initial pile strength prediction model is constructed, expressed by the following formula:

[0086] Where Strength is the model output, representing the predicted pile strength value; P is the injection pressure; and F is the grout flow rate. Increase the drill pipe lifting speed; The drill pipe rotation speed; This refers to model error; Processed for XGBoost regression models;

[0087] Based on historical construction parameters and pile strength data, an initial pile strength prediction model is trained, and the generalization ability of the model is evaluated using root mean square error and coefficient of determination. The model hyperparameters are adjusted using random search, including the number of trees, the maximum tree depth, the learning rate, and the sample sampling ratio. A regression tree is constructed by iteratively optimizing the objective function during the training process.

[0088] Through random search cross-validation, the model automatically evaluates the effect of different hyperparameter combinations by K-fold cross-validation, selects the optimal hyperparameters, and obtains the pile strength prediction model.

[0089] The formula for calculating the root mean square error (RMSE) is as follows:

[0090] in, The actual pile strength corresponding to the i-th historical construction data; The value of the pile body is the predicted value based on the i-th historical construction data model; n is the total number of historical construction parameters and pile strength data.

[0091] The coefficient of determination The calculation formula is as follows:

[0092] in, This represents the average pile strength from historical construction parameters and pile strength data.

[0093] S3. Based on the hybrid swarm optimization algorithm, combined with the artificial bee colony algorithm and the humpback whale optimization algorithm, the optimal parameter settings are found using the pile strength prediction model, and control commands are generated according to the optimal parameter settings to complete the construction control of high-pressure jet grouting piles.

[0094] Step S3 includes the following steps:

[0095] Based on the hybrid swarm optimization algorithm, combined with the artificial bee colony algorithm and the humpback whale optimization algorithm, the construction parameters of the high-pressure jet grouting pile are adjusted, and then combined with the pile strength prediction model to obtain the optimal parameter settings.

[0096] The optimal parameter settings are converted into control commands to achieve real-time control of high-pressure jet grouting pile construction.

[0097] Step S3 is as follows:

[0098] The hybrid swarm optimization algorithm is used to find the optimal parameter settings by taking the vector composed of four parameters, namely injection pressure, slurry flow rate, drill pipe lifting speed and drill pipe rotation speed, and then combining it with the pile strength prediction model.

[0099] The objective function of the hybrid swarm optimization algorithm is to minimize the prediction error, so that the difference between the predicted pile strength and the target strength obtained by the pile strength prediction model through the optimal parameter settings is minimized. The objective function is expressed by the following formula:

[0100] in, Error between predicted and target strength of pile body; The predicted strength of the pile is obtained through the pile strength prediction model. The target strength of the pile body;

[0101] The specific process of the hybrid swarm optimization algorithm is as follows:

[0102] Initialize a random solution space, where each solution represents a vector consisting of four parameters: injection pressure, slurry flow rate, drill pipe lifting speed, and drill pipe rotation speed;

[0103] Set constraints for each parameter to meet actual production requirements;

[0104] Randomly generate solutions within the given constraints;

[0105] The pile strength prediction model is used to predict each solution and calculate its fitness value; the fitness value represents the contribution of each solution to the objective function.

[0106] The artificial bee colony algorithm and the humpback whale optimization algorithm are used to optimize and iterate for each solution;

[0107] The optimization and iteration of each solution based on the artificial bee colony algorithm specifically involves:

[0108] For each solution, perform local optimization, generate new solutions through neighborhood search, update the fitness and make a selection, update the number of trials for the solution, and reinitialize unimproved solutions;

[0109] The optimization iteration based on the humpback whale optimization algorithm for each solution specifically involves:

[0110] The local optimum found in the previous iteration is used as the leader solution; the humpback whale optimization algorithm is used to perform a global search for the solution using the encirclement and spiral update strategy; the position of the solution is evaluated and updated, and the fitness is further optimized;

[0111] The switching ratio between the formation-adaptive artificial bee colony algorithm and the humpback whale optimization algorithm is introduced to dynamically adjust the global and local search weights;

[0112] The switching ratio between the formation-adaptive artificial bee colony algorithm and the humpback whale optimization algorithm Express it using the following formula:

[0113] in, This represents the real-time stratigraphic complexity index; k is an adjustment factor. The standard deviation of the injection pressure; The standard deviation of the slurry flow rate; This represents the average drill pipe lifting speed. This represents the average rotational speed of the drill pipe.

[0114] The iteration before The first iteration is an iterative optimization based on the artificial bee colony algorithm, and the remaining iterations are optimizations based on the humpback whale optimization algorithm;

[0115] By combining the artificial bee colony algorithm and the humpback whale optimization algorithm, the optimal parameter settings are obtained. Based on the optimal parameter settings, control commands are converted into adjustments to the high-pressure jet grouting pile construction parameters, thus achieving real-time control of high-pressure jet grouting pile construction.

[0116] The control command conversion specifically involves: using a PID control algorithm, calculating adjustment values ​​based on real-time construction parameter data and optimal parameter settings, and adjusting the high-pressure jet grouting pile construction parameters based on the adjustment values ​​to complete real-time control of high-pressure jet grouting pile construction;

[0117] The PID control algorithm is expressed using the following formula:

[0118] in, This is the output of the PID controller; This is a proportional parameter; For integration parameters; This is the integral gain parameter; The error of the control parameters at time t is calculated using the following formula: ;in, The optimal parameter values ​​obtained by the hybrid swarm intelligence optimization algorithm; The real-time measurement values ​​acquired by the sensor.

[0119] S4. Continuously acquire real-time construction effect parameters of high-pressure jet grouting piles and compare them with the prediction results. Based on the comparison results, update the pile strength prediction model parameters and hybrid group optimization algorithm parameters to complete the optimization control of high-pressure jet grouting pile construction.

[0120] Step S4 includes the following steps:

[0121] Real-time acquisition of construction parameter data;

[0122] Based on the pile strength prediction model, the real-time pile strength prediction value is calculated, and the error is judged by comparing it with the preset target pile strength.

[0123] When the error is less than the threshold, the standard is deemed met;

[0124] When the error exceeds the threshold, it is judged as an anomaly, and anomaly analysis is triggered;

[0125] If a parameter setting significantly improves the strength of the pile, the weight of that parameter setting is enhanced through a hybrid swarm intelligence optimization algorithm, so that it is preferentially called under similar geological conditions.

[0126] When a parameter setting triggers anomaly analysis, a new solution is locally searched using the artificial bee colony algorithm. The parameter constraint range is adjusted and updated to the model training set to improve prediction accuracy.

[0127] The anomaly analysis specifically refers to:

[0128] Check for abrupt changes in the formation and equipment malfunctions, and record and transmit the inspection results to the user;

[0129] Calculate the standard deviation of the fluctuation between injection pressure and slurry flow rate. If it exceeds the threshold, it is judged as over-spraying or under-spraying. The judgment result is recorded and transmitted to the user.

[0130] Calculate construction efficiency, determine whether there is a coordination problem between drill pipe lifting speed and rotation speed, and record and transmit the determination results to the user.

[0131] The construction efficiency is calculated using the following formula:

[0132] in, This is the first weighting coefficient; This is the second weighting coefficient; For construction efficiency, it represents the coordination efficiency between the drill pipe lifting speed and rotation speed; the larger the value, the better the coordination.

[0133] The present invention also provides a system for implementing the optimized control method for high-pressure jet grouting pile construction, the structural schematic diagram of which is shown below. Figure 2 As shown, it includes an environmental parameter acquisition module, an optimization control module, and an execution layer control module;

[0134] The environmental parameter acquisition module collects construction data from various areas in real time. During the data acquisition process, data filtering and outlier removal methods are combined to avoid the impact of instantaneous pulsations, flow fluctuations or pressure pulses on the system control accuracy. Signal processing methods such as median filtering, mean filtering or Kalman filtering are used to improve the stability and reliability of the data.

[0135] The environmental parameter acquisition module is distributed in different areas of the high-pressure jet grouting device, including the high-pressure pump outlet, the end of the grouting pipeline (the front section of the drill rod or nozzle), the grouting pump outlet pipeline, and segmented timing devices.

[0136] The environmental parameter acquisition module includes a pressure sensor, flow meter, encoder, frequency converter, and timer, which are used to collect construction data in various areas in real time. The construction data includes high-pressure slurry pumping pressure, slurry delivery volume, injection start and end time, drill rod lifting speed, drill rod rotation speed, etc.

[0137] Pressure sensors can be piezoresistive, strain gauge, or capacitive, and should be able to withstand the maximum pressure within the operating range of the flow valve to avoid sensor damage or excessive measurement error. Flow rate can be measured using differential pressure flow meters, electromagnetic flow meters, or ultrasonic flow meters; the specific choice depends on the pipe diameter and measurement accuracy requirements.

[0138] Each sensor interacts with the controller wirelessly for subsequent calculations and adjustments. Through these steps, accurate and real-time data on flow rate, pressure, drill pipe lifting speed, and drill pipe rotation speed can be acquired, providing accurate and reliable foundational data for subsequent prediction and intelligent regulation.

[0139] The optimization control module receives construction data from the environmental parameter acquisition module, analyzes and predicts the construction data based on algorithms, obtains prediction results, optimizes the construction parameters in the high-pressure jet grouting pile control system based on the prediction results, and generates control commands for the actuators.

[0140] The system obtains parameters such as pressure, flow rate, lifting speed, and rotation speed from real-time data of sensors in various areas of the high-pressure jet grouting pile construction system. These parameters are then processed and predicted using machine learning models to generate optimal control parameters for jetting volume, pressure, and time during each construction process.

[0141] After removing outliers such as zero pressure or flow rate from the sensor data, a 5-point moving average filter is used to smooth data fluctuations (5 points is an empirical value and needs to be adjusted according to data characteristics. If the data noise is large (such as electromagnetic interference), the window can be increased to 7 points). Z-score standardization is then used to scale each feature to a distribution with a mean of 0 and a variance of 1. Normalization eliminates the difference in unit dimensions such as MPa and cm / min, avoiding model bias towards large numerical features due to differences in numerical magnitude.

[0142] By collecting real-time sensor data (such as pressure and flow rate) and combining it with the pile strength predicted by the regression model, the HSIO optimization algorithm can dynamically adjust the injection pressure and flow rate in real time to ensure that the pressure and flow rate are optimal for each injection, thereby guaranteeing that the pile strength reaches the expected target. When the predicted pile strength is lower than the target value, the HSIO optimization algorithm will automatically adjust the injection parameters, dynamically optimizing parameters such as pressure, flow rate, lifting speed, and rotation speed, to ensure that the pile strength during construction meets the expected standard, avoiding the lag and human error of manual intervention.

[0143] The execution layer control module is used to receive control commands and dynamically adjust the construction parameters of the high-pressure jet grouting device according to the control commands, so as to complete the optimized control of high-pressure jet grouting pile construction.

[0144] By controlling the actuators, the injection pressure, flow rate, lifting speed, and rotation speed are adjusted to ensure smooth construction. The actuators include control valves (electric or pneumatic) for regulating injection pressure and flow rate. These control valves receive control signals from the PID control system and adjust their opening in real time to precisely control the injection pressure and flow rate. Hydraulic actuators adjust the lifting and rotation movements of the equipment according to the PID control signals, ensuring the accuracy of all operations during pile construction. As the core control unit of the system, the PLC or SCADA system receives the construction parameters output from the optimization control module and converts these parameters into electrical signals using a PID control algorithm, sending them to the actuators. The PLC or SCADA system also continuously monitors the operating status of the actuators to ensure that all parameters consistently meet the optimization requirements during construction.

[0145] The execution layer control module receives control commands wirelessly. These commands include target construction parameters generated by the HSIO optimization algorithm, such as target injection pressure, injection flow rate, lifting speed, and rotation speed. After being parsed by the microprocessor built into the execution layer control module, the control commands are converted into operable control signals and sent to the electric regulating valves, hydraulic actuators, and variable frequency circulating pumps in each area. These control signals ensure that all construction parameters are adjusted according to the target values ​​output by the HSIO optimization algorithm.

[0146] The execution layer control module converts the target parameters (such as injection pressure, injection flow rate, lifting speed, and rotation speed) output by the HSIO optimization algorithm into specific control signals that the execution equipment can recognize. An electric regulating valve is installed in the shotcrete pipeline and automatically adjusts its opening based on the target injection pressure and flow rate, precisely controlling the pressure and flow of the shotcrete system to ensure that the injection parameters meet preset standards during construction. Simultaneously, the hydraulic actuator adjusts the operating status of the lifting and rotating equipment according to the target lifting and rotation speeds, ensuring the stability and accuracy of the pile construction.

[0147] During construction, the execution layer control module uses its equipped sensing units to monitor key parameters of the shotcrete system in real time, including spraying pressure, flow rate, lifting speed, and rotation speed. The sensing units feed this real-time data back to the control module for real-time evaluation.

[0148] Based on this feedback data, the execution layer control module calculates the deviation between the current construction effect and the target by comparing it with the target parameters provided by the HSIO optimization algorithm, and generates new control commands. These new control commands are then transmitted to the execution equipment for dynamic adjustment and optimization of construction parameters. The dynamic adjustment of the execution layer control module relies on the PID control algorithm. The PID control algorithm calculates the adjustment amount based on the error between the real-time measured parameters and the target parameters given by the HSIO optimization algorithm. Specifically, proportional control (P) adjusts in a timely manner according to the current deviation, integral control (I) eliminates the steady-state error of the system, and derivative control (D) predicts the trend of the system based on the rate of change of deviation, thereby suppressing overshoot and fluctuations. Through PID control, the injection pressure, flow rate, lifting speed, and rotation speed of the execution equipment can be precisely adjusted to ensure that all parameters during the construction process are always in an optimal state. During the adjustment of the execution equipment, the system continuously monitors the feedback data to ensure that each operation minimizes deviation and makes the construction parameters such as injection pressure, flow rate, lifting speed, and rotation speed as close as possible to the target values ​​given by the HSIO optimization algorithm.

[0149] The entire process forms a closed-loop control system. The execution layer control module continuously adjusts the equipment status and evaluates the construction effect in real time to ensure that every control adjustment meets the expected goals and ultimately achieves the stability and strength requirements of the pile body.

[0150] The algorithm of this optimization control module is further explained below with reference to an embodiment:

[0151] Based on the same construction dataset, the method of this invention is compared with PSO+linear regression and random forest using two metrics: accuracy and reliability.

[0152] All three models are input with control parameters (P, F, ...). , The method outputs the predicted pile strength. PSO+linear regression iteratively optimizes the coefficients through PSO, outputting the optimal coefficients and generating the prediction result. Random forest directly trains the random forest model using default parameters and outputs the predicted pile strength. The method of this invention first pre-trains with XGBoost, then optimizes the parameters using HSIO, outputting the optimal parameter combination, and uses the optimized parameter combination to generate the prediction result.

[0153] The comparison index of the prediction results of the three models is shown in Table 1.

[0154] Table 1. Comparison of prediction results from the three models.

[0155] Simulation results show that the method of the present invention has the smallest mean square error and the smallest pressure fluctuation value, exhibiting better stability and reliability, reducing the risk of over-spraying or under-spraying, and ensuring that the pile strength is closer to the design target.

Claims

1. A method for optimizing and controlling the construction of high-pressure jet grouting piles, characterized in that, The following steps are involved: S1. Collect historical construction data of high-pressure jet grouting piles; preprocess the collected real-time construction data of high-pressure jet grouting piles to obtain preprocessed data; S2. Based on machine learning algorithms, preprocessed data is used as training data to construct a pile strength prediction model; S3. Based on the hybrid swarm optimization algorithm, combined with the artificial bee colony algorithm and the humpback whale optimization algorithm, the optimal parameter settings are found using the pile strength prediction model, and control commands are generated according to the optimal parameter settings to complete the construction control of high-pressure jet grouting piles. S4. Continuously acquire real-time construction effect parameters of high-pressure jet grouting piles and compare them with the prediction results. Based on the comparison results, update the pile strength prediction model parameters and hybrid group optimization algorithm parameters to complete the optimization control of high-pressure jet grouting pile construction.

2. The optimized control method for high-pressure jet grouting pile construction 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 grout pumping pressure, grout delivery volume, jetting start and end time, drill rod lifting speed, and drill rod rotation speed. The preprocessing specifically involves: during the data acquisition process, combining data filtering and outlier removal methods to avoid the impact of instantaneous pulsations, flow fluctuations, or pressure pulses on accuracy; and employing signal processing methods such as median filtering, mean filtering, or Kalman filtering to improve the stability and reliability of the data.

3. The optimized control method for high-pressure jet grouting pile construction according to claim 2, characterized in that, The preprocessing specifically includes the following steps: From the real-time construction data of high-pressure jet grouting piles, extract the jetting pressure, grout flow rate, drill rod lifting speed and drill rod rotation speed, and remove outliers with pressure or flow rate of 0; Use a 5-point or 7-point moving average filter to smooth data fluctuations; Z-score standardization is used to scale each feature to a distribution with a mean of 0 and a variance of 1, resulting in preprocessed data.

4. The optimized control method for high-pressure jet grouting pile construction according to claim 1, characterized in that, Step S2 specifically includes the following steps: An initial pile strength prediction model was constructed based on the XGBoost regression model. Based on historical construction parameters and pile strength data, an initial pile strength prediction model is trained to obtain the pile strength prediction model.

5. The optimized control method for high-pressure jet grouting pile construction according to claim 4, characterized in that, Step S2 is as follows: Based on the XGBoost regression model, an initial pile strength prediction model is constructed, expressed by the following formula: Where Strength is the model output, representing the predicted pile strength value; P is the injection pressure; and F is the grout flow rate. Increase the drill pipe lifting speed; The drill pipe rotation speed; This refers to model error; Processed for XGBoost regression models; Based on historical construction parameters and pile strength data, an initial pile strength prediction model is trained, and the generalization ability of the model is evaluated using root mean square error and coefficient of determination. The model hyperparameters are adjusted using random search, including the number of trees, the maximum tree depth, the learning rate, and the sample sampling ratio. A regression tree is constructed by iteratively optimizing the objective function during the training process. Through random search cross-validation, the model automatically evaluates the effect of different hyperparameter combinations by K-fold cross-validation, selects the optimal hyperparameters, and obtains the pile strength prediction model. The formula for calculating the root mean square error (RMSE) is as follows: in, The actual pile strength corresponding to the i-th historical construction data; The value of the pile body is the predicted value based on the i-th historical construction data model; n is the total number of historical construction parameters and pile strength data samples. The coefficient of determination The calculation formula is as follows: in, This represents the average pile strength from historical construction parameters and pile strength data.

6. The optimized control method for high-pressure jet grouting pile construction according to claim 1, characterized in that, Step S3 includes the following steps: Based on the hybrid swarm optimization algorithm, combined with the artificial bee colony algorithm and the humpback whale optimization algorithm, the construction parameters of the high-pressure jet grouting pile are adjusted, and then combined with the pile strength prediction model to obtain the optimal parameter settings. The optimal parameter settings are converted into control commands to achieve real-time control of high-pressure jet grouting pile construction.

7. The optimized control method for high-pressure jet grouting pile construction according to claim 6, characterized in that, Step S3 is as follows: The hybrid swarm optimization algorithm is used to find the optimal parameter settings by taking the vector composed of four parameters, namely injection pressure, slurry flow rate, drill pipe lifting speed and drill pipe rotation speed, and then combining it with the pile strength prediction model. The objective function of the hybrid swarm optimization algorithm is to minimize the prediction error, so that the difference between the predicted pile strength and the target strength obtained by the pile strength prediction model through the optimal parameter settings is minimized. The objective function is expressed by the following formula: in, Error between predicted and target strength of pile body; The predicted strength of the pile is obtained through the pile strength prediction model. The target strength of the pile body; n is the total number of samples; The specific process of the hybrid swarm optimization algorithm is as follows: Initialize a random solution space, where each solution represents a vector consisting of four parameters: injection pressure, slurry flow rate, drill pipe lifting speed, and drill pipe rotation speed; Set constraints for each parameter to meet actual production requirements; Randomly generate solutions within the given constraints; The pile strength prediction model is used to predict each solution and calculate its fitness value; the fitness value represents the contribution of each solution to the objective function. The artificial bee colony algorithm and the humpback whale optimization algorithm are used to optimize and iterate for each solution; The artificial bee colony algorithm optimizes each solution iteratively as follows: it performs local optimization on each solution, generates new solutions through neighborhood search, updates the fitness and makes selections, updates the number of trials for solutions, and reinitializes unimproved solutions; The humpback whale optimization algorithm optimizes each solution iteratively as follows: the local optimum found in the previous iteration is taken as the leader solution; the encirclement and spiral update strategy of the humpback whale optimization algorithm is used to perform a global search on the solution; the position of the solution is evaluated and updated, and the fitness is further optimized. The switching ratio between the formation-adaptive artificial bee colony algorithm and the humpback whale optimization algorithm is introduced to dynamically adjust the global and local search weights; The switching ratio between the formation-adaptive artificial bee colony algorithm and the humpback whale optimization algorithm Express it using the following formula: in, This represents the real-time stratigraphic complexity index; k is an adjustment factor. The standard deviation of the injection pressure; The standard deviation of the slurry flow rate; This represents the average drill pipe lifting speed. This represents the average rotational speed of the drill pipe. The iteration before The first iteration is an iterative optimization based on the artificial bee colony algorithm, and the remaining iterations are optimizations based on the humpback whale optimization algorithm; By combining the artificial bee colony algorithm and the humpback whale optimization algorithm, the optimal parameter settings are obtained. Based on the optimal parameter settings, control commands are converted to adjust the construction parameters of the high-pressure jet grouting pile, thus completing the real-time control of the high-pressure jet grouting pile construction. The control command conversion specifically involves: using a PID control algorithm, calculating adjustment values ​​based on real-time construction parameter data and optimal parameter settings, and adjusting the high-pressure jet grouting pile construction parameters based on the adjustment values ​​to complete real-time control of high-pressure jet grouting pile construction; The PID control algorithm is expressed using the following formula: in, This is the output of the PID controller; This is a proportional parameter; For integration parameters; This is the integral gain parameter; The error of the control parameters at time t is calculated using the following formula: ;in, The optimal parameter values ​​obtained by the hybrid swarm intelligence optimization algorithm; The real-time measurement values ​​acquired by the sensor.

8. The optimized control method for high-pressure jet grouting pile construction according to claim 1, characterized in that, Step S4 includes the following steps: Real-time acquisition of construction parameter data; Based on the pile strength prediction model, the real-time pile strength prediction value is calculated, and the error is judged by comparing it with the preset target pile strength. When the error is less than the threshold, the standard is deemed met; When the error exceeds the threshold, it is judged as an anomaly, and anomaly analysis is triggered; If a parameter setting significantly improves the strength of the pile, the weight of that parameter setting is enhanced through a hybrid swarm intelligence optimization algorithm, so that it is preferentially called under similar geological conditions. When a parameter setting triggers anomaly analysis, a new solution is locally searched using the artificial bee colony algorithm, the parameter constraint range is adjusted, and the result is updated in the model training set.

9. The optimized control method for high-pressure jet grouting pile construction according to claim 8, characterized in that, The anomaly analysis specifically refers to: Check for abrupt changes in the formation and equipment malfunctions, and record and transmit the inspection results to the user; Calculate the standard deviation of the fluctuation between injection pressure and slurry flow rate. If it exceeds the threshold, it is judged as over-spraying or under-spraying. The judgment result is recorded and transmitted to the user. Calculate construction efficiency, determine whether there is a coordination problem between drill pipe lifting speed and rotation speed, and record and transmit the determination results to the user; The construction efficiency is calculated using the following formula: in, This is the first weighting coefficient; This is the second weighting coefficient; Increase the drill pipe lifting speed; The drill pipe rotation speed; For construction efficiency, it represents the coordination efficiency between the drill pipe lifting speed and rotation speed; the larger the value, the better the coordination.

10. A system for implementing the optimized control method for high-pressure jet grouting pile construction according to any one of claims 1 to 9, characterized in that, It includes an environmental parameter acquisition module, an optimization control module, and an execution layer control module; The environmental parameter acquisition module collects construction data from various areas in real time. During the data acquisition process, it first processes the data by combining data filtering and outlier removal methods, and then uses signal processing methods such as median filtering, mean filtering or Kalman filtering. The optimization control module receives construction data from the environmental parameter acquisition module, analyzes and predicts the construction data based on algorithms, obtains prediction results, optimizes the construction parameters in the high-pressure jet grouting pile control system based on the prediction results, and generates control commands for the actuators. The execution layer control module is used to receive control commands and dynamically adjust the construction parameters of the high-pressure jet grouting device according to the control commands, so as to complete the optimized control of high-pressure jet grouting pile construction.

Images