Pile foundation hole-forming quality intelligent management and control method and system based on drilling rig working parameters

Abstract

The present application relates to the pile foundation construction technical field, especially is based on the pile foundation hole forming quality intelligent management and control method and system of drilling machine work parameter, the method includes the following steps: the data acquisition array is constructed, the key work parameter and the quality index data of pile foundation hole forming of drilling machine are acquired based on the data acquisition array, and the standard time sequence database is established; the dynamic correlation model is constructed based on the standard time sequence database, and the quality determination standard of the pile foundation hole forming is established based on the dynamic correlation model; the quality state of the pile foundation hole forming is obtained according to the dynamic correlation model to the dynamic prediction of the pile foundation hole forming, and the quality deviation result of the pile foundation hole forming is obtained in combination with the quality determination standard; the control strategy of the drilling machine is constructed based on the quality deviation result, and the key work parameter is adjusted according to the control strategy, so that the quality intelligent management and control of the pile foundation hole forming are realized.The present application realizes the intelligent management and control of the pile foundation hole forming by the drilling machine work parameter driving.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of pile foundation construction, in particular to a pile foundation hole forming quality intelligent management and control method and system based on drilling rig working parameters. BACKGROUND

[0002] Pile foundation hole forming quality directly affects the structural safety and durability of buildings. At present, pile foundation hole forming construction mainly relies on manual experience judgment and post-detection, which is difficult to find problems in the construction process in time, and has the disadvantages of reaction lag, low efficiency, high repair cost, etc. In addition, during the pile foundation construction process, the stratum condition is complex and changeable, and the traditional construction monitoring method is difficult to accurately grasp the real-time dynamics of the construction site, resulting in difficulty in effectively controlling the hole forming quality.

[0003] Although the existing technology has introduced drilling rig operation parameter monitoring, most of them are still limited to single parameter or simple linear correlation analysis, and lack effective means for comprehensive dynamic analysis of multiple parameters, resulting in insufficient prediction accuracy. In addition, the traditional model is difficult to adapt to the changes of complex geological conditions, and cannot realize real-time optimization and adjustment of construction parameters. Some existing monitoring technologies often only focus on a single independent parameter in the drilling rig operation process (such as only monitoring torque or rotational speed), or only analyze data for a certain specific state, without considering the mutual influence between multiple key parameters. In addition, the analysis process is not updated or fed back in real time according to the changes of the construction situation, and usually only analyzes the completed data after the event, lacking the ability of immediate response and intelligent dynamic adjustment during the pile foundation hole forming construction process. This way limits the effective response to complex geological and construction condition changes, and makes it difficult to achieve comprehensive control of pile foundation hole forming quality.

[0004] Based on the current situation, the present application provides a pile foundation hole forming quality intelligent management and control method and system based on drilling rig working parameters, which realizes real-time and accurate prediction of hole forming quality by real-time comprehensive analysis of the dynamic correlation between drilling rig key parameters and hole forming quality, and realizes closed-loop real-time feedback adjustment, thereby significantly improving the timeliness, accuracy and economy of pile foundation hole forming quality control, overcoming the shortcomings of the existing technology. SUMMARY

[0005] In view of the defects in the prior art, the present application provides a pile foundation hole forming quality intelligent management and control method and system based on drilling rig working parameters.

[0006] In order to achieve the above-mentioned purpose, in a first aspect, the present application provides a pile foundation hole forming quality intelligent management and control method based on drilling rig working parameters, the method comprising the following steps: constructing a data acquisition array, acquiring key working parameters of a drilling rig and quality index data of pile foundation hole forming based on the data acquisition array, and establishing a standard time sequence database; constructing a dynamic correlation model based on the standard time sequence database, and establishing quality determination criteria for the pile foundation hole forming based on the dynamic correlation model; dynamically predicting the pile foundation hole forming according to the dynamic correlation model to obtain a pile foundation hole forming quality state, and combining the quality determination criteria to obtain a quality deviation result of the pile foundation hole forming; constructing a control strategy for the drilling rig based on the quality deviation result, adjusting the key working parameters according to the control strategy, and realizing intelligent management and control of the quality of the pile foundation hole forming. The present application realizes closed-loop management and control of the quality of pile foundation hole forming through data driving; through real-time acquisition and standardization of multi-source parameters, the present application eliminates manual recording errors and supports dynamic analysis; the dynamic correlation model quantifies the coupling relationship between parameters, the quality determination criteria constructed break through the limitations of traditional experience thresholds, realize accurate positioning of deviations, and effectively avoid quality accidents such as hole deviation and hole collapse; the automatic control strategy based on the model shortens the parameter adjustment cycle, significantly reduces the rework rate and material waste, and saves the comprehensive construction cost.

[0007] Optionally, the constructing a data acquisition array, acquiring key working parameters of a drilling rig and quality index data of pile foundation hole forming based on the data acquisition array, and establishing a standard time sequence database comprises: constructing the data acquisition array based on high-frequency sensors and intelligent sensing devices; performing a trial drilling on a region to be constructed, acquiring initial real-time operating parameters of the drilling rig as historical key working parameters through the data acquisition array, and acquiring historical quality index data of the pile foundation hole forming by using the data acquisition array; establishing a basic database based on the historical key working parameters and the historical quality index data; in a construction process, acquiring real-time key working parameters of the drilling rig based on the data acquisition array to construct a real-time construction database, and establishing a quality index database by collecting real-time quality index data of the pile foundation hole forming through the data acquisition array; and standardizing the basic database, the real-time construction database, and the quality index database to establish the standard time sequence database. The present application realizes all-around acquisition of construction parameters and hole forming quality through a high-frequency intelligent sensor array, and the established historical and real-time double-database architecture not only forms a construction experience benchmark, but also realizes dynamic monitoring; the standardized time sequence database forms a space-time correlation between drilling rig working parameters and hole forming quality index, provides data driving support for construction parameter optimization, quality early warning, and process improvement, significantly improves the fine management and control level of pile foundation construction, and reduces quality risks.

[0008] Optionally, the dynamic correlation model is constructed based on the standard time sequence database, and the quality determination standard of the pile foundation hole forming is established based on the dynamic correlation model, including: a model data set is established based on the basic database, the dynamic correlation model between the key working parameters and the quality index data is constructed according to the model data set, the dynamic correlation model includes a verticality deviation model, a hole position deviation model and a drilling depth model; the key working parameters are used as input parameters of the dynamic correlation model, the quality index data are used as output indexes of the dynamic correlation model, and correlation analysis is performed on the input parameters and the output indexes by a statistical method to obtain a correlation analysis result; the model form of the dynamic correlation model is determined according to the correlation analysis result, and the model form includes a linear model or a nonlinear model; the dynamic correlation model is subjected to model optimization and model verification, and the quality determination standard is established by using the model output of the dynamic correlation model. The quantitative mapping relationship between the parameters and the quality indexes is established by the historical data driving, the interpretable deduction from the construction parameters to the hole forming quality is realized, the parameter coupling law is intelligently identified based on the statistical method, the linear or nonlinear model structure is automatically adapted, the prediction accuracy is improved, the model optimization and verification mechanism ensure the prediction reliability, finally the quality determination standard is constructed, the experience judgment is converted into data-driven decision, the quality control is turned to process control, and the early warning ability of quality abnormality is significantly improved, thereby providing a scientific basis for dynamic adjustment of construction parameters.

[0009] Optionally, the model form of the dynamic correlation model is determined according to the correlation analysis result, and the model form includes a linear model or a nonlinear model, including: the linear model satisfies the following relationship:

[0010]

[0011] wherein, is a dependent variable, is a regression coefficient to be fitted, is an independent variable, is an intercept term; the nonlinear model includes a quadratic polynomial or a cubic polynomial and a nonlinear exponential model: the quadratic polynomial satisfies the following relationship:

[0012]

[0013] the cubic polynomial satisfies the following relationship:

[0014]

[0015] wherein, is a dependent variable, is a regression coefficient to be fitted, is an independent variable, is an intercept term; the nonlinear exponential model satisfies the following relationship:

[0016]

[0017] wherein, is a dependent variable, is a baseline parameter, is an exponential growth rate, is an independent variable, is a bias. The present application realizes accurate selection of model form through correlation analysis, quantifies the synergistic effect of multiple parameters in the form of a simple weighted sum for a linear model, expands the quadratic or cubic polynomial to capture the square term and interaction term effects of the independent variable, and adapts the quality evolution characteristics with exponential law for a nonlinear exponential model, which not only ensures the calculation efficiency of simple working conditions, but also retains the fitting ability of complex scenarios through high-order terms and exponential terms, so that the dynamic correlation model can adapt to the parameter coupling law of different geological conditions and construction stages, and significantly improve the universality and accuracy of quality prediction.

[0018] Optionally, the model optimization and model verification of the dynamic correlation model comprises: obtaining model optimal parameters by performing parameter fitting optimization on the linear model and the nonlinear model through an optimization algorithm; and performing effectiveness test on the dynamic correlation model, wherein the effectiveness test comprises model fitting effect evaluation, residual analysis and cross-validation. The present application improves model accuracy through a parameter optimization algorithm, ensures that the parameter configuration reaches an optimal fitting state, and significantly reduces prediction bias; uses fitting effect evaluation to quantify model explanatory power, residual analysis to diagnose system bias and abnormal points, and cross-validation to guarantee model generalization performance; effectively prevents overfitting, ensures the stability and prediction reliability of the dynamic correlation model under complex working conditions, provides a scientific basis for quality determination standards that have been strictly verified, and significantly improves the accuracy of construction parameter regulation and the confidence of quality early warning.

[0019] Optionally, the dynamic prediction of the pile foundation hole forming according to the dynamic correlation model obtains a pile foundation hole forming quality state, and the quality deviation result of the pile foundation hole forming is obtained in combination with the quality determination standard, including: taking the real-time construction database as a model input of the dynamic correlation model, obtaining a model output by real-time prediction of the dynamic correlation model based on the model input, and the model output is a predicted pile foundation hole forming quality state; and real-time comparison of the pile foundation hole forming quality state with the quality determination standard obtains the quality deviation result. The present application realizes the prediction of the pile foundation hole forming quality by iteratively calculating the model driven by the real-time input construction parameters, forms a closed-loop control of monitoring, prediction and evaluation; the quality deviation result is fed back to the construction state in real time, so that the quality control is changed from passive detection to active intervention, the problem response cycle is significantly shortened, the quality deviation root cause is accurately located by combining the parameter sensitivity analysis capability of the dynamic correlation model, the drilling machine parameter dynamic optimization is guided, the rework rate and material waste are effectively reduced, and at the same time, the accumulated deviation data continuously improves the quality determination standard, and promotes the intelligent iterative upgrading of the construction process.

[0020] Optionally, the real-time construction database is taken as a model input of the dynamic correlation model, a model output is obtained by real-time prediction of the dynamic correlation model based on the model input, and the model output is a predicted pile foundation hole forming quality state, including:

[0021]

[0022] wherein, is a pile foundation hole forming quality prediction index, is time, is a mapping function of the dynamic correlation model, is a real-time key working parameter, is a model parameter of the dynamic correlation model. The present application realizes the prediction of the pile foundation quality state by coupling calculation of the real-time input construction parameters and the optimal model parameters; the introduction of the time variable makes the prediction have the time sequence evolution capability, and can capture the continuity characteristics of the construction process; the mapping function accurately describes the complex correlation between the working parameters and the quality state, provides a quantitative decision basis for the dynamic optimization of the drilling machine parameters, and promotes the evolution of the construction process to the data-driven intelligent mode.

[0023] Optionally, the quality deviation result includes a quality deviation type and a quality deviation degree of the pile foundation hole forming, and the control strategy is constructed according to the quality deviation type and the quality deviation degree, and the control strategy includes verticality overrun control, hole position deviation control and depth error control. The present application realizes accurate intervention of construction parameters through classification and grading of quality deviation, and forms a multi-dimensional correction system including verticality overrun control, hole position deviation control and depth error control, thereby significantly improving the response accuracy of quality abnormalities; the dynamic adjustment mechanism based on the deviation degree can avoid overcorrection, optimize construction energy consumption under the premise of ensuring quality standards, realize early warning and active repair of quality problems, effectively reduce the risk of rework, and continuously optimize the control strategy by accumulating control cases, thereby realizing intelligent iteration of construction technology.

[0024] Optionally, the verticality overrun control includes:

[0025]

[0026] wherein, is a rotation speed adjustment amount of the drilling machine, is a proportional coefficient, is a verticality deviation amount, is a maximum allowable deviation amount of the verticality, is an integral coefficient, is a differential coefficient The present application realizes accurate control of the drilling machine posture through dynamic compensation of the verticality deviation, and the multi-parameter collaborative control mechanism makes the verticality control have self-adaptive ability, thereby significantly improving the hole forming verticality precision and reducing the frequency of manual intervention; meanwhile, the quantitative output of the rotation speed adjustment amount provides accurate control instructions for the drilling machine actuator, thereby forming a complete closed loop from quality monitoring to parameter optimization.

[0027] Secondly, this invention provides an intelligent control system for pile foundation hole formation quality based on drilling rig operating parameters. The system executes the intelligent control method for pile foundation hole formation quality based on drilling rig operating parameters provided by this invention. The system includes input devices, output devices, a processor, and a memory. Its advantages lie in the excellent performance of the integrated hardware facilities. The input devices, output devices, processor, and memory are interconnected, ensuring smooth information transmission between components. Through the interaction of multiple hardware facilities, a highly efficient information processing system is constructed. This invention, through its efficient information processing architecture, ensures parallel processing capabilities for data acquisition, model training, and real-time prediction, enabling the monitoring of drilling rig parameters and hole formation quality. It effectively supports real-time prediction of dynamic correlation models, achieves quality deviation early warning, generates control commands, significantly shortens the control closed-loop cycle, and upgrades quality control from being driven by manual experience to being driven by data intelligence. This greatly improves construction accuracy and anomaly response efficiency while reducing reliance on manual intervention. Attached Figure Description

[0028] Figure 1 This is a flowchart of the intelligent control method for pile foundation hole formation quality based on drilling rig operating parameters according to an embodiment of the present invention.

[0029] Figure 2 This is a framework diagram of an intelligent control system for pile foundation hole formation quality based on drilling rig operating parameters, according to an embodiment of the present invention.

[0030] Figure 3 This is a closed-loop flowchart of the intelligent control system for pile foundation hole formation quality based on drilling rig operating parameters, according to an embodiment of the present invention. Detailed Implementation

[0031] Specific embodiments of the present invention will now be described in detail. It should be noted that the embodiments described herein are for illustrative purposes only and are not intended to limit the invention. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the invention. However, it will be apparent to those skilled in the art that these specific details are not necessary to practice the invention. In other instances, well-known circuits, software, or methods have not been specifically described to avoid obscuring the invention.

[0032] Reference throughout this specification to "one embodiment", "an embodiment", "one example", or "an example", means that a particular feature, structure, or characteristic described in connection with the embodiment or example is included in at least one embodiment of the application. The appearances of the phrases "in one embodiment" or "in an embodiment" or "one example" or "an example" in various places in the specification are not necessarily all referring to the same embodiment or example. Furthermore, the particular features, structures, or characteristics can be combined in any suitable

[0033] Reference will now be made to Figure 1 One embodiment of the present application provides a pile foundation hole-forming quality intelligent control method based on drilling rig working parameters, which comprises the following steps:

[0034] S1, a data acquisition array is constructed, key working parameters of a drilling rig and quality index data of pile foundation hole-forming are acquired based on the data acquisition array, and a standard time sequence database is established.

[0035] In this embodiment, data of key working parameters of a drilling rig related to pile foundation hole-forming quality indexes are collected, including historical construction data of the drilling rig and real-time collected data.

[0036] Before formal construction, trial drilling is carried out in the area to be constructed, key parameters of the drilling rig running under different geological conditions are collected, initial real-time running parameter data in the trial drilling process are taken as historical key working parameters, and corresponding pile foundation hole-forming quality measured data are taken as historical quality index data, so as to establish a basic database, correct the model or function relationship according to the data, and establish the relationship between the drilling rig working parameters and the pile foundation hole-forming quality indexes and the pile foundation quality judgment standard.

[0037] During formal construction, high-frequency sensors installed on the drilling rig are used to collect key running parameters of the drilling rig in real time, and the collection frequency of real-time data is set to ensure the real-time and integrity of all key data during construction.

[0038] Pile foundation hole-forming quality indexes include hole position deviation, perpendicularity, and drilling depth, and front-end intelligent sensing devices installed on the drilling rig are used to collect hole-forming quality index data.

[0039] The historical data provides a stable basis for the establishment of the dynamic correlation model. By reviewing the construction data under various geological conditions, the system can establish a relatively accurate correlation model before entering the formal construction, and these data reflect the influence relationship of the drilling rig working parameters on the pile foundation quality under different environments. Using these historical data, the system can establish a preliminary quality standard to ensure that the subsequent construction data are effectively connected with the standard.

[0040] The introduction of real-time data enables the model to be continuously optimized and updated during the construction process. As the construction proceeds, new real-time construction data are continuously collected and stored, and each new data collected can provide additional correction and adjustment for the model. Based on these data, the model is continuously updated and optimized through learning to ensure that the model can cope with changes in geological environment and construction conditions, so that the parameters can be adjusted more effectively to ensure the stability of the hole quality.

[0041] Specifically, at the pile foundation construction site, the data of the key working parameters of the drilling rig related to the pile foundation hole quality index are collected, the high-frequency sensors installed at the key positions of the drilling rig are used to collect the key operating parameters of the drilling rig in real time, and the collection frequency of the real-time data is set to ensure the real-time and completeness of all key data during the construction process. The specific sensor arrangement is as follows:

[0042] A rotation speed sensor (encoder) is installed on the power head of the drilling rig to measure the rotation speed, which is in units of revolutions per minute, and the collected data can be used to analyze the rotation condition of the drill bit and the construction efficiency; a torque sensor is installed at the torsion part of the drill pipe or drill bit to obtain the resistance torque, which is in units of Newton-meters (N·m); a pressure sensor is arranged on the hydraulic cylinder circuit to convert the push stroke pressure (shaft pressure); a displacement sensor or measuring rope is used to record the lifting and lowering displacement of the drilling tool to calculate the drilling depth and speed; an inclination sensor is installed on the mast of the drilling rig to monitor the verticality of the pile frame, which is in units of degrees (°). The real-time collection frequency is not less than once per second.

[0043] In this embodiment, the characteristic value of the foundation bearing capacity is the design parameter given by the survey and specification after comprehensively considering the factors such as the shear strength, compressibility and underground water of the stratum. It directly reflects the limit pressure per unit area that the stratum can bear, and is closely related to the pile foundation resistance, which is in units of kilopascals (kPa). In this invention, the parameter represents the characteristic value of the foundation bearing capacity of the first layer of the foundation soil, which can be extracted from the engineering geological survey report. The characteristic value of the foundation bearing capacity comprehensively reflects the strength and deformation performance of the soil layer, and can effectively represent the comprehensive resistance of the rock and soil of the layer, serving as the basis for the composition of the rock and soil strength parameter in the model.

[0044] When the drilling hole passes through multiple layers of soil, the equivalent rock and soil strength can be calculated according to the weighted formula, which satisfies the following relationship:

[0045]

[0046] wherein, is the rock strength, is the total amount of rock layers, is the index variable of rock layers, the current enters the first the thickness of the layer of rock, is the current drilling depth, is the first the characteristic value of the bearing capacity of the foundation.

[0047] Through collecting the data of the drilling construction process under various geological conditions in the historical construction projects, a historical construction database is formed. The database content includes the drilling parameters such as the drilling speed, the torque, the axial pressure, the drilling displacement (speed), the drilling rod inclination angle, and the corresponding measured quality indexes such as the pile hole verticality, the hole position deviation, and the drilling depth error. These data will be used for subsequent model training and quality judgment standard establishment.

[0048] Further, the historical construction data is collected, and the historical data of the previous pile foundation construction projects under various typical geological conditions is collected. The specific collected drilling parameters include the speed, the torque, the axial pressure, the drilling displacement, the drilling rod inclination angle, and the corresponding hole forming quality indexes including the pile hole verticality, the hole position deviation, the drilling depth error, and the corresponding pile foundation hole forming quality measured data. The basic database of the historical construction is constructed, the basic data is provided for the model training and verification, the preliminary correlation model between the drilling parameters and the hole forming quality and the quality judgment standard are established through the historical data analysis.

[0049] Further, the real-time construction data is collected, and the key operating parameters of the drilling machine are monitored in real time in the construction site. The drilling machine is installed with sensors such as the speed sensor, the torque sensor, the axial pressure sensor, the displacement sensor, and the drilling rod inclination sensor. The real-time collection frequency is not less than once per second. The real-time construction database is constructed based on the collected real-time data, which is used for the real-time hole forming quality prediction of the model, and is compared and analyzed in real time with the judgment standard established based on the historical construction data, the construction state is monitored in real time, and the pile foundation hole forming quality indexes (verticality, hole position deviation, drilling depth) are collected, and the quality index database is established.

[0050] In this embodiment, all collected data should be stored in the database, and real-time construction data and historical construction data together constitute a data set. Each time new data is collected, it needs to be combined with the historical data set. To ensure the effectiveness of the data, a time series database is used for data storage to ensure the efficiency of data query and update. During data collection, all real-time data needs to be standardized to eliminate the dimensional differences of different sensor data; for example, rotation speed, torque, pressure and other data need to be converted into a unified dimension through standardization method; the standardization method satisfies the following relationship:

[0051]

[0052] wherein, is the standardized data, is the original data, is the mean of the original data, is the standard deviation of the original data.

[0053] S2, based on the standard time series database, a dynamic correlation model is constructed, and based on the dynamic correlation model, a quality judgment standard of the pile foundation hole forming is established.

[0054] In this embodiment, based on the historical construction data, the relationship between the working parameters of the drilling machine and the pile foundation hole forming quality index is analyzed through correlation analysis and statistical modeling method, and the strength and form of the parameter influence on the quality are determined. Then, a linear or nonlinear model is used to perform parameter fitting and optimization on the model through least square method or Levenberg-Marquardt algorithm (LM), and cross validation and residual analysis are performed to ensure the accuracy and reliability of the model prediction. Finally, a reliable dynamic correlation model of the drilling machine operating parameters and the hole forming quality is established, and a pile foundation quality judgment standard is established according to the model output, which is used for real-time construction quality evaluation.

[0055] Specifically, the historical construction data covering different geological conditions, drilling machine parameter changes and corresponding hole forming quality measured indexes are used to establish a dynamic correlation model between the drilling machine working parameters and the pile foundation hole forming quality indexes (verticality, hole position deviation, drilling depth), so as to learn the specific influence law of the drilling machine parameter change on the construction quality.

[0056] The specific implementation steps are as follows:

[0057] S21, based on the principles of physical mechanics and drilling technology, the basic assumptions of the model are proposed: the higher the speed, the faster the drilling speed, but too high speed may increase the swing of the drill pipe, affecting the perpendicularity; the size of the torque is proportional to the properties of the stratum, and too large torque may cause uneven wear of the inner wall of the drill hole, affecting the deviation of the hole position and the perpendicularity; too large axial pressure may cause uneven pressure on the drill pipe, causing the drill pipe to bend, causing the inclination angle of the drill pipe to increase, and thus affecting the perpendicularity and hole position deviation; too fast drilling speed may lead to a decrease in the perpendicularity of the drill hole, as the drill pipe is prone to swing during fast drilling; the inclination angle of the drill pipe is directly related to the perpendicularity of the hole in a linear or nonlinear manner.

[0058] It is assumed that the strength parameters of rock and soil directly affect the resistance of the drill bit and the stability of the drill hole during drilling, and thus affect the response relationship of parameters such as torque and axial pressure. In high-strength strata, insufficient axial pressure may cause the drill bit to slip and the hole to be uneven; in low-strength soil layers, an unreasonable axial pressure to drilling speed ratio may easily cause the drill hole to collapse. Therefore, there is an interaction between rock and soil strength and other construction parameters, which needs to be included as an independent feature in the dynamic correlation modeling to enhance the adaptability and accuracy of the model under complex geological conditions. The value of rock and soil strength takes into account the compressive strength, shear strength, friction of the soil layer, and other stratum characteristics, and is quantitatively represented by the characteristic value of the foundation bearing capacity.

[0059] S22, establish a model data set, collect and organize the drilling machine operating parameters and corresponding pile foundation hole quality indicators under different geological conditions as a training set to establish a dynamic correlation model. The historical data set includes drilling machine operating parameters and pile foundation hole quality indicators: drilling machine operating parameters include speed, torque, pressure, drilling displacement (speed) and drill pipe inclination angle, which affect the quality of the drill hole; pile foundation hole quality indicators include hole position deviation, perpendicularity and drill hole depth error, which are used to measure the quality of the hole. Assuming there are N historical data records, the data set can be represented as:

[0060]

[0061]

[0062]

[0063] wherein, is the drilling machine operating parameter in the th historical data record, is the hole quality indicator in the historical data record, N is the total amount of historical data records, is the index variable of the historical data record, is the speed, is the torque, is the pressure, For drilling displacement, The drill pipe inclination angle. For hole position deviation, For verticality, This represents the drilling depth error.

[0064] S23. Parameter Correlation Analysis: To clarify the influence and relationship between each input parameter (rotation speed, torque, axial pressure, drilling displacement (speed), and drill pipe inclination angle) and the output indicators (verticality, hole position deviation, and drilling depth), statistical methods were used to quantitatively analyze the relationships between parameters. Pearson correlation coefficient and Spearman rank correlation coefficient were used for analysis.

[0065] Pearson correlation coefficient analysis is suitable for datasets where the relationship between parameters is approximately linear and the variables follow a normal distribution; satisfying the following relationship:

[0066]

[0067] in, The Pearson correlation coefficient is used. For the first input parameter One measurement value, The mean of the input parameters. For the output parameter of the first One measurement value, This is the mean of the output parameters.

[0068] Spearman's rank correlation coefficient analysis is suitable for nonlinear relationships or cases where the data is not normally distributed; it satisfies the following relationship:

[0069]

[0070] in, The Spearman rank correlation coefficient. The difference in rank for each data set. The total number of samples.

[0071] Specifically, the criteria for judging the strength of correlation are shown in Table 1:

[0072] Table 1

[0073]

[0074] S24, model form and equation determination, after completing the correlation analysis, determine the mathematical form of the correlation model, and clarify the physical meaning and possible influence path of each parameter, for example: the rotation speed is related to the lateral swing of the drill pipe, which can be assumed to be a nonlinear relationship; torque and shaft pressure have a more direct impact on the bending stiffness of the drill pipe and drilling stability, which can be assumed to be a linear or exponential relationship; the inclination angle of the drill pipe directly reflects the deviation degree of the hole, which can be assumed to be a linear or nonlinear relationship with the perpendicularity; drilling speed is directly related to drilling stability and hole wall quality, which can be linear or nonlinear.

[0075] Based on the physical hypothesis and preliminary analysis, the model form is selected. If the correlation analysis shows that the linear relationship is obvious, the linear model is selected, which satisfies the following relationship:

[0076]

[0077] wherein, is the dependent variable, is the regression coefficient to be fitted, is the independent variable, is the intercept term.

[0078] If there is an obvious nonlinear relationship, a quadratic polynomial or a cubic polynomial and a nonlinear exponential model are used.

[0079] The quadratic polynomial satisfies the following relationship:

[0080]

[0081] The cubic polynomial satisfies the following relationship:

[0082]

[0083] wherein, is the dependent variable, is the regression coefficient to be fitted, is the independent variable, is the intercept term.

[0084] The nonlinear exponential model (physical model oriented) satisfies the following relationship:

[0085]

[0086] wherein, is the dependent variable, is the baseline parameter, is the exponential growth rate, is the independent variable, is the offset.

[0087] Specifically, the dynamic correlation model includes a perpendicularity deviation model, a hole position deviation model, and a drilling depth model.

[0088] S241, a verticality deviation model; the verticality deviation refers to the verticality of the drilling hole, i.e. the deviation degree of the actual drilling hole axis from the designed axis. It is assumed that the verticality deviation is affected by the rotating speed, the torque, the axial pressure, the drilling speed, the drill pipe inclination angle and the rock-soil layer property, and satisfies the following relationship:

[0089] Linear model:

[0090]

[0091] Nonlinear model (e.g. quadratic model):

[0092]

[0093] wherein, is the verticality deviation, is the regression coefficient to be fitted, is the rotating speed, is the torque, is the pressure, is the speed, is the drill pipe inclination angle, is the rock-soil strength, is the intercept term.

[0094] S242, a hole position deviation model; the hole position deviation refers to the deviation between the actual drilling hole position and the designed position, which usually affects the hole forming quality and the structure safety, and satisfies the following relationship:

[0095] Linear model:

[0096]

[0097] Nonlinear model (e.g. quadratic model):

[0098]

[0099] wherein, is the hole position deviation, is the regression coefficient to be fitted, is the rotating speed, is the torque, is the pressure, is the speed, is the drill pipe inclination angle, is the rock-soil strength, is the intercept term.

[0100] S243, a drilling depth model satisfying the following relationship:

[0101] Linear model:

[0102]

[0103] Nonlinear model (e.g. quadratic model):

[0104]

[0105] wherein, is the error of the drilling depth, is the regression coefficient to be fitted, is the rotation speed, is the torque, is the pressure, is the inclination angle of the drill pipe, is the speed, is the rock-soil strength, is the intercept term.

[0106] S25, determination of optimal model parameters.

[0107] S251, linear model parameters are determined by the least square method, satisfying the following relationship:

[0108]

[0109] wherein, is the residual sum of squares, is the total number of data samples, is the index variable of the data sample, is the actually measured pile foundation hole forming quality index parameter, is the regression coefficient to be fitted, is the data sample, is the intercept term.

[0110] Further, the pile foundation hole forming quality index parameter includes: a linear model is established for each quality index, each represents a single hole forming quality index, i.e. three independent equations are constructed, satisfying the following relationship:

[0111]

[0112] wherein, represents the verticality model, is the regression coefficient to be fitted of the verticality model, is the data sample, represents the hole position deviation model, is the regression coefficient to be fitted of the hole position deviation model, represents the drilling depth model, is the regression coefficient to be fitted of the drilling depth model, is the intercept term.

[0113] Let the above function take partial derivative for each regression coefficient to be fitted and let it be 0, establish equation group, meet the following relationship:

[0114]

[0115] Wherein, is the residual sum of squares, is the regression coefficient to be fitted.

[0116] Solving the above linear equation group, the parameter value of the regression coefficient can be obtained.

[0117] S252, the LM algorithm determines the nonlinear model parameters.

[0118] S2521, take the verticality deviation of pile foundation as an example, first give the initial parameter value, the initial parameter value is generally determined according to the empirical rule (such as obtained according to linear fitting approximation), or small range random setting, meet the following relationship:

[0119]

[0120] Wherein, is the initial parameter vector, is the initial model parameter.

[0121] S2522, construct error function and objective function, define error term as the difference between actual measurement value and predicted value, meet the following relationship:

[0122]

[0123] Objective function, meet the following relationship:

[0124]

[0125] Wherein, is the error term of the th data group, is the model parameter, is the actual measured pile hole forming quality index parameter, is the prediction function of nonlinear model, is the sample, is the objective function, is the number of data groups, is the index variable of data group.

[0126] S2523, the algorithm updates the parameters at each iteration, the specific steps are as follows:

[0127] Calculate the Jacobian matrix, which represents the partial derivative matrix of the error function with respect to each parameter, meet the following relationship:

[0128]

[0129] in, For Jacobian matrices, For model parameters, For the first Error terms for each data set The regression coefficients to be fitted are . Number of data sets; The number of parameters.

[0130] The algorithm iteratively updates the parameters, satisfying the following relationship:

[0131]

[0132]

[0133] in, For the first Model parameters for the next iteration For the first Model parameters for the next iteration For Jacobian matrices, Indicates transpose. The damping coefficient is... It is the identity matrix. For the error vector, For the first Error terms for each data set.

[0134] When the error function decreases significantly, the damping coefficient is reduced to approximate the Gauss-Newton method; when the error function does not decrease significantly or even increases, the damping coefficient is increased to approximate the gradient descent method.

[0135] S2524, Iteration termination condition judgment.

[0136] The model is considered convergent when both the parameter convergence condition and the objective function convergence condition are met simultaneously, and the iterative update process stops. If only one of them is met, iteration continues until both are met simultaneously or the preset maximum number of iterations is reached.

[0137] Parameter convergence condition: The difference between two iterations of the parameter is less than a set threshold, satisfying the following relationship:

[0138]

[0139] in, For the first Model parameters for the next iteration For the first Model parameters for the next iteration The threshold value is used.

[0140] Objective function convergence condition: the change of objective function twice is less than the set threshold, which satisfies the following relationship:

[0141]

[0142] Wherein, is the objective function of the first iteration, is the objective function of the first iteration, is the threshold value.

[0143] As the construction progresses, new real-time construction data is continuously collected and stored. When new data enters the system, incremental learning is performed based on these data to continuously update and optimize the model. The core step of incremental learning is the update of the existing model. When new real-time data is input, the model will gradually adjust its parameters through incremental update algorithm without the need to retrain the entire model.

[0144] The relationship between the drilling rig operating parameters and the hole quality is analyzed through correlation analysis and statistical modeling methods, and the model parameters are fitted and optimized through optimization methods such as least squares method and LM algorithm. This process actually involves the determination of model parameters, that is, how to determine the optimal parameters of the model through training data. Incremental learning will affect the model parameters through real-time calculation data without the need to retrain the entire model, which enables the model to continuously optimize the model parameters and adjust them according to new data after each data collection.

[0145] S26, after the model parameters are fitted, the model cannot be completely effective and needs to be verified for effectiveness through model validation to further adjust and optimize the model to improve its prediction accuracy. Model validation includes the following steps:

[0146] S261, model fitting effect evaluation; statistical indicators are used to evaluate the fitting effect, and the coefficient of determination satisfies the following relationship:

[0147]

[0148] Wherein, is the coefficient of determination, is the actual value, is the model prediction value, is the mean value of the actual value.

[0149] S262, residual analysis, which satisfies the following relationship:

[0150]

[0151] Wherein, is the error term of the first data set, is the actual value, is the model predicted value.

[0152] In an optional embodiment, a residual scatter plot is drawn to observe the distribution, the residuals should be randomly distributed around 0, and there should be no obvious trend or regularity; a residual histogram or quantile-quantile plot (Q-Q plot) is drawn to test whether it is approximately normally distributed, normal distribution indicates that the model is effective, and non-normal distribution needs to be re-considered.

[0153] S263, cross-validation; the model prediction performance is evaluated by using an independent data set (test set), the data is randomly divided into training set and test set; the training set is used to fit the parameters, and the test set is used to test the prediction accuracy.

[0154] The root mean square error satisfies the following relationship:

[0155]

[0156] The mean absolute percentage error satisfies the following relationship:

[0157]

[0158] wherein, is the root mean square error, is the total amount of data, is the index variable, is the actual value, is the model predicted value, is the mean absolute percentage error.

[0159] S27, based on the trained dynamic correlation model, a judgment standard of pile foundation hole forming quality is established, which is used to judge whether the real-time predicted quality in the construction process meets the requirements.

[0160] Specifically, the output of the model is compared with the preset quality tolerance range. For example, the tolerance range of verticality, the tolerance range of hole position deviation and the tolerance range of drilling depth are set; if the model prediction result exceeds these tolerance ranges, it indicates that there is a quality deviation, which needs to trigger an alarm and enter the subsequent feedback control stage.

[0161] The model output satisfies the following relationship:

[0162]

[0163] wherein, is the model output, is the hole position deviation predicted value, is the verticality prediction error, is a drilling depth error prediction value.

[0164] The allowable error range of the hole position deviation satisfies the following relationship:

[0165]

[0166] wherein, is a hole position deviation prediction value, is a minimum allowable amount of hole position deviation, is a maximum allowable amount of hole position deviation.

[0167] The allowable error range of the perpendicularity satisfies the following relationship:

[0168]

[0169] wherein, is a perpendicularity prediction error, is a minimum allowable deviation amount of perpendicularity, is a maximum allowable deviation amount of perpendicularity.

[0170] The allowable error range of the drilling depth satisfies the following relationship:

[0171]

[0172] wherein, is a drilling depth error prediction value, is a minimum allowable amount of drilling depth error, is a maximum allowable amount of drilling depth error.

[0173] The updated dynamic correlation model will better adapt to the changes in the construction environment and improve the accuracy of real-time prediction. The above historical construction data collection and real-time construction data collection are correlated through data standards. Specifically, the real-time data collected during construction are compared and analyzed in real time with the initial dynamic correlation model and quality determination standard established based on the historical construction data, so that the quality of the pile foundation hole is accurately judged in real time.

[0174] S3, dynamically predicting the pile foundation hole according to the dynamic correlation model to obtain the quality state of the pile foundation hole, and combining the quality determination standard to obtain the quality deviation result of the pile foundation hole.

[0175] In this embodiment, the real-time monitored drilling rig working parameters are input into the dynamic correlation model, the current pile foundation hole quality state is dynamically predicted in real time by using the dynamic correlation model, and possible quality deviations are identified. The real-time prediction result is compared and analyzed with the established quality standard, if it is out of limit, a warning is triggered, and the prediction accuracy is ensured through data review or correction mechanism, which provides a basis for dynamic regulation.

[0176] Specifically, the real-time collected drilling rig working parameters (such as rotation speed, torque, pressure, speed (displacement), and drill pipe inclination angle) are taken as input data and input into the previously trained dynamic correlation model. The input of the model is the real-time data of the drilling rig working parameters, and the output is the predicted pile foundation hole forming quality state, including hole position deviation, perpendicularity, and drilling depth quality indicators. The input data satisfies the following relationship:

[0177]

[0178] wherein, is a real-time key working parameter, is time, is rotation speed, is torque, is pressure, is speed, is drill pipe inclination angle.

[0179] Specifically, for each real-time collected sample, real-time prediction is performed through the model to obtain the quality indicator prediction value of the pile foundation hole forming at the current time, and the prediction value represents the quality state of the pile foundation hole forming under the current construction state. The prediction frequency is synchronized with the data collection frequency (not less than once per second), ensuring real-time performance.

[0180] The quality indicator prediction value satisfies the following relationship:

[0181]

[0182] wherein, is a quality indicator prediction value, is time, is a hole position deviation prediction value, is a perpendicularity prediction error, is a drilling depth error prediction value.

[0183] The prediction process satisfies the following relationship:

[0184]

[0185] wherein, is a pile foundation hole forming quality prediction indicator, is time, is a mapping function of the dynamic correlation model, is a real-time key working parameter, is a model parameter of the dynamic correlation model.

[0186] In an optional embodiment, a sliding window mechanism is employed. Based on the dynamic characteristics of the construction process (such as the frequency of drilling rig parameter changes), the time span of the sliding window is defined and combined with recent historical data for short-term trend analysis, reducing instantaneous noise interference. In conventional soil layers, the window length is set to 10 seconds (sampling once per second, with the window containing 10 sets of data). Real-time data is stored in a time-series database with timestamps.

[0187] Extraction time each time new data arrives At that time The 10 sets of data form a window dataset, and the window slides forward as new data arrives (removing the oldest data). ,join in ), dynamically update window content.

[0188] The window dataset satisfies the following relationship:

[0189]

[0190]

[0191] in, For window datasets, for Drilling rig parameters at any given time For rotational speed, For torque, For pressure, For speed, The angle of inclination of the drill pipe.

[0192] Specifically, trend feature extraction involves fitting a straight line to the parameter sequence within the window, calculating the slope as the trend strength, and satisfying the following relationship:

[0193]

[0194] in, For trend strength, Relative time within the window, For parameter values.

[0195] Furthermore, the difference between the first and last data points within the window is calculated to measure the magnitude of parameter change, satisfying the following relationship:

[0196]

[0197] in, For the range of parameter change, The first data value of the window. This is the last data value in the window.

[0198] Furthermore, the moving average is calculated, satisfying the following relationship:

[0199]

[0200] wherein, is the sliding mean, is the data value within the window.

[0201] The original input is extended to meet the following relationship:

[0202]

[0203] wherein, is the trend prediction value within the window, is the drilling rig parameter at the moment, is the rotation speed, is the torque, is the pressure, is the speed, is the angle of inclination of the drill pipe, is the window average of the rotation speed, is the dynamic change rate of the rotation speed, is the parameter change amplitude, is the standard deviation of the pressure, is the window average of the angle of inclination of the drill pipe, is the dynamic change rate of the angle of inclination of the drill pipe.

[0204] The abnormal point determination condition meets the following relationship:

[0205]

[0206] wherein, is the drilling rig parameter at the moment, is the trend prediction value within the window, is the standard deviation within the window.

[0207] If the above abnormal point determination condition is met, it is determined to be an abnormal point, and data review or model prediction result correction is triggered.

[0208] In the embodiment, the real-time predicted quality index is compared with the pre-set quality determination standard to determine whether there is a quality deviation, and the set quality determination standard includes the above-mentioned allowable error range of perpendicularity, the allowable error range of hole position deviation and the allowable error range of drilling depth.

[0209] The predicted quality index is compared with the pre-established standard in real time, and once the prediction result shows that there is a risk of quality deviation, for example, the pile hole may be deflected or the hole position may be obviously deviated, the system immediately triggers an automatic early warning prompt and prepares to enter the feedback control stage.

[0210] S4, constructing a regulation strategy of the drilling rig based on the quality deviation result, and adjusting the key working parameters according to the regulation strategy to realize intelligent quality control of the pile hole forming.

[0211] In the embodiment, for the predicted hole forming quality problem or deviation, a dynamic feedback regulation strategy and regulation instruction are automatically generated through a closed-loop feedback control system, an adjustment scheme of the drilling rig operation parameters is determined, and real-time automatic implementation of the drilling rig parameter adjustment is performed to quickly correct the hole forming quality deviation and correct the quality problem. The closed-loop control architecture forms an adaptive regulation closed loop through linkage of sensors, model prediction and execution devices.

[0212] Specifically, the intelligent control system automatically generates a dynamic feedback regulation strategy of the drilling rig operation parameters, such as needing to reduce the drilling speed or adjust the drill bit speed to correct the hole deviation, or adjust the axial pressure to correct the hole position deviation, etc. Through the closed-loop feedback control system, the quality deviation in the construction process is automatically corrected in time. The system automatically delivers the generated regulation instruction to the automatic closed-loop control execution device of the drilling rig to immediately implement real-time adjustment of the parameters, so as to quickly and efficiently solve the quality deviation problem in the construction process and ensure the hole forming quality.

[0213] According to the type and degree of the predicted hole forming quality deviation, a regulation instruction is generated through a rule engine and an optimization algorithm:

[0214] S41, verticality out-of-limit regulation.

[0215] If the predicted value of the verticality is greater than the maximum allowable deviation amount of the verticality, the rotation speed is preferentially reduced to reduce the lateral swing, and the hydraulic deviation correction system is used to finely adjust the inclination angle of the drill pipe.

[0216] Based on a proportional-integral-derivative control algorithm (PID), the parameter is adjusted according to the deviation ratio, and the following relationship is satisfied:

[0217]

[0218] wherein, is the rotation speed adjustment amount of the drilling rig, is a proportional coefficient, is a verticality deviation amount, is a maximum allowable deviation amount of the verticality, is an integral coefficient, is a differential coefficient.

[0219] It should be noted that the proportional coefficient, integral coefficient and differential coefficient are empirical tuning coefficients, which are determined by historical data optimization.

[0220] S42, hole position deviation regulation.

[0221] If the hole position deviation prediction value is greater than the maximum allowable amount of hole position deviation, reduce the axial pressure to reduce the bit lateral force, and adjust the drilling direction.

[0222] S43, depth error regulation.

[0223] If the drilling depth error prediction value is greater than the maximum allowable amount of drilling depth error, increase the displacement sensor calibration frequency, and dynamically correct the drilling speed.

[0224] Please refer to Figure 2 In an optional embodiment, the present application provides a pile foundation hole forming quality intelligent control system based on drilling rig working parameters, which comprises input devices, output devices, processors and memories, and is connected between the hardware facilities, wherein the memory is used to store computer programs, the computer programs comprise program instructions, and the processor is configured to call the program instructions to execute the specific steps of the embodiments of the pile foundation hole forming quality intelligent control method based on drilling rig working parameters provided by the present application. The pile foundation hole forming quality intelligent control system based on drilling rig working parameters provided by the present application has complete structure, objective stability, improves the overall applicability and practical application ability of the present application.

[0225] Please refer to Figure 3 In an optional embodiment, the rotation speed, torque, pressure, speed and inclination angle obtained by the drilling rig sensor are taken as the original signal, the real-time data acquisition system acquires the original signal and takes it as the data input, which is input to the dynamic model for hole forming quality prediction, and then the intelligent control algorithm generates a control command according to the prediction result, transmits the control command to the drilling rig actuator to adjust the rotation speed, torque, axial pressure, speed and inclination angle, realizes intelligent adjustment of drilling parameters, and continuously monitors in real time, forming a closed loop. The intelligent control system is further based on real-time monitoring, dynamically adjusts the drilling parameters through closed-loop control, ensures that the hole forming quality is in a controlled state, collects the drilling rig parameters through the sensor, predicts the quality by using the intelligent control algorithm, and feeds back the control drilling device to realize closed-loop adaptive control.

[0226] In summary, the method provided by the pile foundation hole forming quality intelligent management and control method and system based on the drilling rig working parameters comprehensively analyzes the dynamic correlation between the drilling rig key parameters and the hole forming quality in real time, realizes real-time accurate prediction of the hole forming quality by using the dynamic correlation model, and performs closed-loop real-time feedback adjustment, thereby significantly improving the timeliness, accuracy and economy of the pile foundation hole forming quality control, the method is easy to understand, simple to calculate, has small workload and is convenient for engineering application, and provides a theoretical basis and technical support for further development of pile foundation construction technology.

[0227] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application, and they should be covered in the scope of the claims and description of the present application.

Claims

1. A pile foundation hole-forming quality intelligent management and control method based on rig working parameters, characterized in that, The method comprises the following steps: constructing a data acquisition array, acquiring key working parameters of a drilling machine and quality index data of pile foundation hole forming based on the data acquisition array, and establishing a standard time sequence database; constructing a dynamic correlation model based on the standard time sequence database, and establishing a quality determination standard of the pile foundation hole forming based on the dynamic correlation model; performing dynamic prediction on the pile foundation hole forming according to the dynamic correlation model to obtain a pile foundation hole forming quality state, and combining the quality determination standard to obtain a quality deviation result of the pile foundation hole forming; constructing a regulation and control strategy of the drilling machine based on the quality deviation result, adjusting the key working parameters according to the regulation and control strategy, and realizing intelligent management and control of the quality of the pile foundation hole forming; the construction of the data acquisition array, the acquisition of the key working parameters of the drilling machine and the quality index data of the pile foundation hole forming based on the data acquisition array, and the establishment of the standard time sequence database comprise: constructing the data acquisition array based on high-frequency sensors and intelligent sensing devices; drilling a test hole in a region to be constructed, acquiring initial real-time operating parameters of the drilling machine as historical key working parameters through the data acquisition array, and acquiring historical quality index data of the pile foundation hole forming by using the data acquisition array; establishing a basic database based on the historical key working parameters and the historical quality index data; in the construction process, constructing a real-time construction database based on real-time key working parameters of the drilling machine acquired by the data acquisition array, and establishing a quality index database by collecting real-time quality index data of the pile foundation hole forming through the data acquisition array; standardizing the basic database, the real-time construction database and the quality index database to establish the standard time sequence database; the construction of the dynamic correlation model based on the standard time sequence database, and the establishment of the quality determination standard of the pile foundation hole forming based on the dynamic correlation model comprise: establishing a model data set based on the basic database, and constructing the dynamic correlation model between the key working parameters and the quality index data according to the model data set, wherein the dynamic correlation model comprises a verticality deviation model, a hole position deviation model and a drilling depth model; the key working parameters are used as input parameters of the dynamic correlation model, and the quality index data are used as output indicators of the dynamic correlation model, and correlation analysis is performed on the input parameters and the output indicators by a statistical method to obtain a correlation analysis result; determining a model form of the dynamic correlation model according to the correlation analysis result, wherein the model form comprises a linear model or a nonlinear model; performing model optimization and model verification on the dynamic correlation model, and establishing the quality determination standard by using model output of the dynamic correlation model; the model optimization and model verification on the dynamic correlation model comprise: performing parameter fitting optimization on the linear model and the nonlinear model by an optimization algorithm to obtain optimal model parameters; performing effectiveness test on the dynamic correlation model, wherein the effectiveness test comprises model fitting effect evaluation, residual analysis and cross-validation; determining parameters of the linear model by a least square method, and satisfying the following relationship: wherein, is the residual sum of squares, is the total number of data samples, is the index variable of data samples, is the actually measured pile foundation hole-forming quality index parameter, is the regression coefficient to be fitted, is the data sample, is the intercept term; An equation is established for each quality indicator to meet the following relationship: wherein, represents a perpendicularity model, is a regression coefficient to be fitted for the perpendicularity model, is a data sample, represents a hole position deviation model, is a regression coefficient to be fitted for the hole position deviation model, represents a drilling depth model, is a regression coefficient to be fitted for the drilling depth model, is an intercept term; The partial derivatives of the above functions with respect to each regression coefficient to be fitted are taken and set to zero to establish an equation group that meets the following relationship: wherein, is the sum of the squared residuals, is the regression coefficient to be fitted; Solving the equation group obtains the parameter values of the regression coefficients as the parameters of the linear model; The parameters of the nonlinear model are determined by the LM algorithm, and given initial parameter values meet the following relationship: wherein, is an initial parameter vector, is an initial model parameter; An error function and an objective function are constructed, the error function meets the following relationship: The objective function meets the following relationship: wherein is the error term for the i-th data set, is a model parameter, is a model parameter, is a measured pile quality parameter, is a prediction function of the nonlinear model, is a sample, is an objective function, is the number of data sets, is an index variable for the data sets. The LM algorithm iteratively updates the parameters to meet the following relationship: wherein is the model parameter of the first iteration, is the model parameter of the first iteration, is the model parameter of the first iteration, is the model parameter of the first iteration, is the Jacobian matrix, denotes the transpose, is the damping coefficient, is the identity matrix, is the error vector, is the error term of the first data set, is the error term of the first data set. A parameter convergence condition and an objective function convergence condition are constructed as an iteration termination condition, and when the iteration of the LM algorithm meets the iteration termination condition, the updated parameters are obtained as the parameters of the nonlinear model; The model fitting effect evaluation is realized by using the coefficient of determination; The residual analysis meets the following relationship: wherein, is the error term for the th data set, is the actual value, is the model predicted value; The cross-validation is realized by the root mean square error and the mean absolute percentage error; The parameters of the dynamic correlation model include the rock and soil strength, and when the borehole passes through multiple layers of soil, the equivalent rock and soil strength can be calculated according to the weighted formula to meet the following relationship: wherein, is the rock strength, is the total amount of rock, is the index variable of rock, is currently entered into the is the thickness of the layer of rock, is the current drilling depth, is the first is the characteristic value of the bearing capacity of the ground The correlation analysis includes Pearson correlation coefficient analysis and Spearman rank correlation coefficient analysis; The quality determination standard includes the allowable error range of hole position deviation, the allowable error range of verticality, and the allowable error range of drilling depth; The quality deviation result includes the quality deviation type and the quality deviation degree of the pile foundation hole forming; The quality deviation result includes the quality deviation type and the quality deviation degree of the pile foundation hole forming; The quality deviation result includes the quality deviation type and the quality deviation degree of the pile foundation hole forming; The verticality over-limit regulation includes: wherein, is a rotational speed adjustment amount of the rig, is a proportional coefficient, is a deviation amount of the perpendicularity, is a maximum allowable deviation amount of the perpendicularity, is an integral coefficient, is a differential coefficient; The hole position deviation regulation includes: if the hole position deviation predicted value is greater than the maximum allowable amount of hole position deviation, the axial pressure is reduced to reduce the lateral force of the drill bit, and the drilling direction is adjusted; The depth error regulation includes: if the drilling depth error predicted value is greater than the maximum allowable amount of drilling depth error, the displacement sensor calibration frequency is increased, and the drilling speed is dynamically corrected.

2. The pile foundation hole-forming quality intelligent management and control method based on rig operating parameters according to claim 1, characterized in that, The model form of the dynamic correlation model is determined according to the correlation analysis result, and the model form includes a linear model or a nonlinear model, including: The linear model meets the following relationship: wherein, is the dependent variable, is the regression coefficient to be fitted, is the independent variable, is the intercept term; The nonlinear model includes a quadratic polynomial or a cubic polynomial and a nonlinear exponential model: The quadratic polynomial meets the following relationship: The cubic polynomial meets the following relationship: wherein, is the dependent variable, is the regression coefficient to be fitted, is the independent variable, is the intercept term; The nonlinear exponential model meets the following relationship: wherein, is the dependent variable, is the baseline parameter, is the exponential growth rate, is the independent variable, is the offset.

3. The pile foundation hole-forming quality intelligent management and control method based on rig operating parameters according to claim 1, characterized in that, According to the dynamic correlation model, the quality state of the pile foundation hole forming is dynamically predicted, and the quality deviation result of the pile foundation hole forming is obtained in combination with the quality determination standard, including: The real-time construction database is used as the model input of the dynamic correlation model, and the model output is obtained by real-time prediction based on the model input through the dynamic correlation model, which is the predicted quality state of the pile foundation hole forming; The real-time construction database is used as the model input of the dynamic correlation model, and the model output is obtained by real-time prediction based on the model input through the dynamic correlation model, which is the predicted quality state of the pile foundation hole forming; The pile foundation hole-forming quality state is compared with the quality determination standard in real time to obtain the quality deviation result.

4. The pile foundation hole-forming quality intelligent management and control method based on rig operating parameters according to claim 3, characterized in that, The real-time construction database is taken as model input of the dynamic correlation model, and model output is obtained through real-time prediction of the dynamic correlation model based on the model input, the model output being a predicted pile foundation hole-forming quality state, including: wherein, is a pile foundation hole-forming quality prediction index, is time, is a mapping function of the dynamic correlation model, is a real-time key work parameter, is a model parameter of the dynamic correlation model.

5. The pile foundation hole-forming quality intelligent management and control system based on the drilling rig working parameters, characterized in that, The system comprises an input device, an output device, a processor and a memory, which are connected with each other, wherein the memory is used for storing a computer program, the computer program comprises program instructions, and the processor is configured to invoke the program instructions to execute the pile foundation hole-forming quality intelligent management and control method based on drilling rig working parameters according to any one of claims 1-4. The pile foundation hole-forming quality state is compared with the quality determination standard in real time to obtain the quality deviation result. The real-time construction database is taken as model input of the dynamic correlation model, and model output is obtained through real-time prediction of the dynamic correlation model based on the model input, the model output being a predicted pile foundation hole-forming quality state, including: The system comprises an input device, an output device, a processor and a memory, which are connected with each other, wherein the memory is used for storing a computer program, the computer program comprises program instructions, and the processor is configured to invoke the program instructions to execute the pile foundation hole-forming quality intelligent management and control method based on drilling rig working parameters according to any one of claims 1-4.

Images