Method, system and device for improving efficiency of rotary drilling system of foundation pit excavation equipment

Abstract

This invention belongs to the field of foundation pit construction and relates to a method, system, and equipment for improving the efficiency of rotary drilling systems for foundation pit excavation. It aims to solve the problem of low construction efficiency caused by the fragmented control of drilling and cuttings removal processes and the lack of formation adaptive capability in rotary drilling systems. This invention aligns energy injection rate, crushing efficiency, and cuttings removal smoothness; constructs energy conservation equations and mass conservation equations based on the aligned data, and extracts continuous homology features to output formation categories; constructs a set of differential equations including the state variable of cuttings accumulation at the bottom of the borehole, and sets physical matching inequality constraints between cuttings removal capacity and cuttings accumulation; constructs drilling control agents and cuttings removal control agents, using state variables as shared state inputs, and outputting drilling parameters and cuttings removal lifting parameters. This invention achieves integrated adaptive control of drilling and cuttings removal in rotary drilling systems, significantly improving construction efficiency under complex formation conditions.

Description

Technical Field

[0001] This invention belongs to the field of foundation pit engineering construction technology, and specifically relates to a method, system and equipment for improving the efficiency of a rotary drilling system for foundation pit excavation. Background Technology

[0002] In foundation pit engineering, rotary drilling systems (such as rotary drilling rigs and continuous wall grab buckets) are the core hole-forming equipment, and their drilling and slag removal efficiency directly affects the construction cycle, energy consumption costs, and project quality. As urban underground space development moves towards deep and large foundation pits, rotary drilling systems face prominent problems such as complex and variable geological conditions, strong coupling of drilling parameters, and repeated breakage at the bottom of the hole due to poor slag removal, which places higher demands on the refined control of the rotary drilling process.

[0003] Currently, parameter setting and process control of rotary drilling systems mainly rely on the experience and judgment of operators. Operators manually adjust control parameters based on changes in apparent parameters such as power head pressure, rotational speed, and hoisting speed during drilling, combined with their understanding of the geological formation. This manual, experience-based control method has significant limitations: firstly, differences in experience among operators lead to large fluctuations in construction efficiency, making it difficult to achieve stable and controllable construction quality; secondly, manual control cannot respond in real time to sudden changes in the geological formation, often resulting in sudden drops in drilling efficiency, drill cuttings accumulation, and stuck drill bits, which can even lead to drill bit burial accidents in severe cases.

[0004] To improve the automation level of rotary drilling systems, some existing technologies have introduced sensor monitoring methods. These involve installing pressure sensors, speed sensors, displacement sensors, and flow sensors on the rotary drilling equipment to collect various operational data in real time during the drilling process. However, existing monitoring systems generally use the data from each sensor independently, only for threshold alarms or post-analysis of single parameters, failing to achieve deep fusion and collaborative optimization of multi-source data. Regarding parameter control, existing technologies mostly employ fixed threshold control or simple proportional-integral-derivative control strategies. Once control parameters are pre-set, they are difficult to adaptively adjust according to changes in the formation, resulting in insufficient adaptability to complex formations.

[0005] Furthermore, existing rotary drilling control methods generally treat the drilling process and the cuttings removal process separately. Drilling parameter adjustments do not consider the transport capacity constraints of the cuttings removal channel, and cuttings removal and drill string lifting parameters do not consider the impact of the cuttings accumulation at the bottom of the hole on drilling efficiency. They lack a joint optimization mechanism for the time and energy consumption of the entire "drilling-cutting-drill string lifting" process. At the same time, existing technologies have failed to establish a multi-source data fusion decision model based on geological feature identification and online equipment status assessment, making it difficult to achieve adaptive optimal control of rotary drilling systems under varying geological conditions.

[0006] Therefore, there is an urgent need for a method to improve the efficiency of rotary drilling systems that can integrate geological features, equipment status, and process data, and achieve coordinated optimization control of drilling parameters and slag removal and drilling parameters. Summary of the Invention

[0007] To address the aforementioned problems in the prior art, namely the low construction efficiency of existing rotary drilling systems due to the disconnected control of drilling and slag removal processes and the lack of adaptability to geological formations, this invention provides a method, system, and equipment for improving the efficiency of rotary drilling systems for foundation pit excavation.

[0008] In a first aspect, the present invention proposes a method for improving the efficiency of a rotary drilling system for foundation pit excavation, the method comprising: During the drilling process of the rotary drilling system, the output data of multiple source sensors are collected, and the output data is converted into energy injection rate, crushing work efficiency and slag discharge smoothness, and then aligned under the time reference to obtain aligned data. Based on the alignment data, energy conservation equations and mass conservation equations are constructed. A two-dimensional phase space is constructed based on the crushing efficiency and slag discharge smoothness. The continuous coherence characteristics of the phase trajectories in the two-dimensional phase space are extracted. The formation category is output based on the continuous coherence characteristics. Based on the energy conservation equation, the mass conservation equation, and the formation type, a set of differential equations is constructed, which includes the state variable of the amount of rock cuttings accumulated at the bottom of the borehole. In the set of differential equations, a physical matching inequality constraint between the slag removal capacity and the amount of rock cuttings accumulated is set. Construct drilling control agent and slag removal control agent, using the state variables in the differential equation system as shared state inputs. Set the utility function of drilling control agent as effective breaking power divided by cuttings accumulation, and set the utility function of slag removal control agent as slag removal mass flow rate minus the penalty term for exceeding cuttings accumulation limit. The two agents iteratively update their respective policy networks to maximize their own utility functions, and output drilling parameters and slag removal drilling parameters.

[0009] Furthermore, the output data is converted into energy injection rate, crushing efficiency, and slag discharge smoothness, and aligned under a time reference to obtain aligned data. The method is as follows: By performing time-series synchronization calibration on the raw output data collected by the multi-source sensors of the rotary drilling system, the phase shift introduced by the sampling frequency difference and transmission path delay is eliminated, and a synchronized data sequence is obtained. Based on the synchronous data sequence, the following were extracted: the energy injection rate, which characterizes the rate of change of mechanical energy input to the rotary drilling head over time; the crushing work efficiency, which characterizes the crushing volume per unit energy consumption when the drill teeth crush rock and soil; and the slag discharge smoothness, which characterizes the smoothness of the movement of drill cuttings along the slag discharge channel. The energy injection rate, crushing efficiency, and slag discharge smoothness are embedded into the same time reference according to the unified time axis of the drilling process to form aligned data with time consistency.

[0010] Furthermore, based on the aligned data, energy conservation equations and matter conservation equations are constructed using the following method: The energy injection rate in the alignment data is used as the total mechanical energy input to the rotary drilling system. The effective energy consumed by rock and soil breaking is inverted by combining the breaking work efficiency in the alignment data with the drilling parameters. The difference between the energy injection rate and the effective energy is allocated to the energy dissipated by frictional heat and the energy dissipated by elastic waves. Based on the energy conservation relationship, an energy conservation equation is established that the energy injection rate is equal to the sum of the energy consumption of rock and soil breaking, the energy consumption of frictional heat dissipation, and the energy consumption of elastic waves. The amount of cuttings generated per unit time is inverted by the crushing efficiency in the aligned data, and the amount of cuttings transported and discharged per unit time is inverted by the slag discharge smoothness in the aligned data combined with the slag discharge parameters. Based on the material conservation relationship, a material conservation equation is established in which the amount of cuttings generated is equal to the sum of the rate of change of the amount of cuttings transported and discharged and the amount of cuttings accumulated at the bottom of the borehole. The energy consumption of rock and soil breaking in the energy conservation equation and the amount of rock cuttings generated in the material conservation equation are correlated and coupled through the rock and soil breaking specific energy parameter to form a set of simultaneous equations describing the interaction between energy distribution and material migration during the drilling process of the rotary drilling system.

[0011] Furthermore, a two-dimensional phase space is constructed based on the crushing efficiency and slag discharge smoothness. The continuous coherence characteristics of the phase trajectories in the two-dimensional phase space are extracted, and the formation category is output based on these continuous coherence characteristics. The method is as follows: Using crushing efficiency as the first dimension parameter of the two-dimensional phase space and slag discharge unobstructedness as the second dimension parameter, the crushing efficiency value and slag discharge unobstructedness value corresponding to each moment during the drilling process are mapped as phase points in the two-dimensional phase space, and the phase points are connected in time sequence to form a phase trajectory. The phase trajectory is subjected to continuous cohomology analysis to extract the continuous cohomology features of the connected components and the generation and disappearance of the void structure at different scales. The extracted continuous coherence features are input into a pre-trained formation classifier, which outputs the formation category corresponding to the current drilling process.

[0012] Furthermore, based on the energy conservation equation, the mass conservation equation, and the formation type, a system of differential equations is constructed that includes the state variable of the amount of rock debris deposited at the bottom of the borehole. The method is as follows: Extract the differential equation form with the amount of rock debris accumulation at the bottom of the borehole as the state variable from the energy conservation equation and the matter conservation equation; The rate of change of rock cuttings accumulation at the bottom of the borehole is determined by the rock cuttings generation rate minus the rock cuttings discharge rate. The rock cuttings generation rate is obtained by inverting the rock and soil crushing energy consumption in the energy conservation equation combined with the rock and soil crushing specific energy. The rock cuttings discharge rate is obtained by inverting the slag discharge smoothness in the material conservation equation combined with the slag discharge capacity. Based on the geological formation, the corresponding rock and soil breaking energy and rock and soil friction coefficient are matched from the pre-established rock and soil parameter library. The matched parameters are substituted into the differential equation form to form a closed differential equation system with the amount of rock cuttings accumulated at the bottom of the borehole as the state variable and the drilling parameters and the cuttings removal and drilling parameters as control variables.

[0013] Furthermore, a physical matching inequality constraint is set in the system of differential equations between the slag removal capacity and the amount of rock cuttings accumulation. The method is as follows: Based on the inner diameter of the slag discharge channel, the pitch and helix angle of the spiral blades, the friction coefficient between the drill cuttings and the wall of the slag discharge channel, and the rotational power parameters and drilling speed of the spiral blades during the drilling process, a functional relationship is established between the upper limit of the slag discharge capacity and the drilling speed, the rotational speed of the spiral blades, and the density of the drill cuttings. The maximum slag discharge mass flow rate that can be output under the current slag discharge and drilling parameters is calculated using the aforementioned functional relationship; The actual required flow rate for slag removal is determined based on the amount of rock cuttings accumulated at the bottom of the borehole, the density of drill cuttings, and the cross-sectional area of ​​the slag removal channel inlet. A physical matching inequality constraint is set to limit the actual slag discharge demand flow rate to no more than the product of the maximum slag discharge mass flow rate and the slag discharge smoothness, and this inequality constraint is used as the boundary condition of the feasible region of the differential equation system.

[0014] Furthermore, the drilling control agent and the slag removal control agent are constructed using the following method: A drilling control agent is constructed. The state space of the drilling control agent includes the state variables in the differential equation system. The action space of the drilling control agent includes the drilling parameters. The utility function of the drilling control agent is set as the effective breaking power divided by the amount of rock cuttings accumulated at the bottom of the hole. A slag removal control intelligent agent is constructed. The state space of the slag removal control intelligent agent shares the state variables in the differential equation system with the drilling control intelligent agent. The action space of the slag removal control intelligent agent includes slag removal and drilling parameters. The utility function of the slag removal control intelligent agent is set as the slag removal mass flow rate minus the penalty term for excessive rock cuttings accumulation at the bottom of the hole. Using the system of differential equations as an environmental model, the drilling control agent and the slag removal control agent select actions according to their respective policy networks. After executing the actions in the environmental model, they update the state variables, calculate the utility function value, and iteratively update the policy network parameters to maximize their own utility functions, and output the optimized drilling parameters and slag removal drilling parameters.

[0015] Furthermore, after each drilling cycle, the energy conservation equation and the mass conservation equation are substituted into the equations to calculate the energy conservation residual and the mass conservation residual. When any residual exceeds the inherent threshold of the equipment, the inherent parameters of the equipment in the differential equation set are adjusted, and the policy network of the two agents is iteratively updated again. The method is as follows: The energy injection rate, crushing efficiency, and slag discharge smoothness in the aligned data at the end of each drilling cycle, together with the drilling parameters and slag discharge lifting parameters output by the drilling control agent and the slag discharge control agent in that cycle, are substituted into the energy conservation equation and the mass conservation equation. The difference between the two ends of the equation is then calculated to obtain the energy conservation residual and the mass conservation residual. The energy conservation residual and the material conservation residual are compared with the preset equipment inherent thresholds respectively. When any residual exceeds the corresponding threshold, the equipment inherent parameters in the differential equation system are corrected in reverse according to the direction and magnitude of the residual, so that the corrected differential equation system matches the actual observation data of the current drilling cycle. The strategy network parameters of the drilling control agent and the slag removal control agent are iteratively updated based on the modified differential equation system until the residuals converge to within the inherent threshold of the equipment.

[0016] In a second aspect, the present invention proposes an efficiency improvement system for a rotary drilling system of a foundation pit excavation equipment, which is used to realize a method for improving the efficiency of a rotary drilling system of a foundation pit excavation equipment, comprising: The data acquisition and processing module is used to acquire the output data of multi-source sensors during the drilling process of the rotary drilling system, convert the output data into energy injection rate, crushing work efficiency and slag discharge smoothness, and align them under the time reference to obtain aligned data. The equation construction and formation identification module is used to construct energy conservation equations and mass conservation equations based on the alignment data, construct a two-dimensional phase space based on crushing work efficiency and slag discharge smoothness, extract the continuous coherence features of phase trajectories in the two-dimensional phase space, and output the formation category based on the continuous coherence features. The differential equation construction and constraint module is used to construct a set of differential equations containing the state variable of the amount of rock cuttings at the bottom of the borehole based on the energy conservation equation, the matter conservation equation and the formation type, and to set physical matching inequality constraints between the slag removal capacity and the amount of rock cuttings in the set of differential equations. The intelligent agent control module includes a drilling control intelligent agent and a cuttings removal control intelligent agent. The intelligent agent control module uses the state variables in the differential equation system as shared state inputs to the drilling control intelligent agent and the cuttings removal control intelligent agent. The utility function of the drilling control intelligent agent is set as the effective breaking power divided by the amount of cuttings accumulation, and the utility function of the cuttings removal control intelligent agent is set as the cuttings removal mass flow rate minus the penalty term for exceeding the cuttings accumulation limit. The two intelligent agents iteratively update their respective policy networks to maximize their own utility functions and output drilling parameters and cuttings removal and drilling lifting parameters.

[0017] A third aspect of the present invention provides an apparatus comprising: At least one processor; and a memory communicatively connected to at least one of the processors; The memory stores instructions that can be executed by the processor to implement a method for improving the efficiency of a rotary drilling system for foundation pit excavation.

[0018] The beneficial effects of this invention are: This invention eliminates phase shifts caused by sampling frequency differences and transmission delays by performing time-series synchronization calibration on the output data of multi-source sensors in a rotary drilling system. The synchronized data sequence is then converted into three types of characteristic parameters with clear physical meaning: energy injection rate, crushing efficiency, and slag discharge smoothness. These parameters are embedded under a unified time reference to form aligned data, solving the problem of difficulty in integrating and utilizing multi-source heterogeneous data. This provides a high-fidelity data foundation for subsequent energy and material analysis.

[0019] This invention utilizes the energy injection rate and crushing efficiency in aligned data to allocate the mechanical energy input to the rotary drilling system to the energy consumption of rock and soil crushing, frictional heat dissipation, and elastic wave dissipation, thus constructing an energy conservation equation. Using crushing efficiency and slag discharge unobstructed flow, a material conservation equation is established between the rate of change of rock cuttings generation, rock cuttings transport and discharge, and rock cuttings accumulation at the bottom of the borehole. By using the rock and soil crushing specific energy, the energy consumption of rock and soil crushing in the energy conservation equation is correlated with the amount of rock cuttings generated in the material conservation equation, achieving coupled analysis of energy allocation and material migration. Simultaneously, this invention constructs a two-dimensional phase space using crushing efficiency and slag discharge unobstructed flow, extracting the continuous coherence characteristics of the connected components of the phase trajectory at different scales and the generation and disappearance of borehole structures. Through a formation classifier, the formation category corresponding to the current drilling process is output, achieving online identification of formation conditions and providing geological prior information for adaptive control under varying formation conditions.

[0020] This invention extracts a differential equation form with the amount of rock cuttings accumulated at the bottom of the borehole as the state variable from the energy conservation equation and the mass conservation equation. Based on the identified stratum type, it matches the corresponding rock and soil fragmentation energy and rock and soil friction coefficient and substitutes them into the differential equation to form a closed differential equation system with the amount of rock cuttings accumulated at the bottom of the borehole as the state variable and the drilling parameters and slag removal and drilling parameters as control variables, thus realizing a quantitative description of the dynamics of rock cuttings accumulation at the bottom of the borehole. Based on this, the present invention constructs a functional relationship between the upper limit of the slag discharge capacity and the drilling speed, the drilling speed, the rotational speed of the spiral blades, and the drilling speed, according to the inner diameter of the slag discharge channel, the pitch and helix angle of the spiral blades, the friction coefficient between the drill cuttings and the wall of the slag discharge channel, and the rotational power parameters and drilling speed of the spiral blades during the drilling process. It calculates the maximum slag discharge mass flow rate that can be output under the current slag discharge and drilling parameters, and establishes an inequality constraint with the actual slag discharge demand flow rate determined by the amount of rock cuttings accumulated at the bottom of the hole. It limits the actual slag discharge demand flow rate to not greater than the product of the maximum slag discharge mass flow rate and the slag discharge smoothness. This inequality constraint is used as the boundary condition of the feasible region of the differential equation system, which effectively avoids the problem of repeated crushing or stuck drill due to the accumulation of rock cuttings at the bottom of the hole caused by insufficient slag discharge capacity.

[0021] This invention constructs a drilling control agent and a cuttings removal control agent. The state spaces of the two agents share state variables in a set of differential equations. The action space of the drilling control agent includes drilling parameters, and the utility function is the effective breaking power divided by the amount of cuttings accumulated at the bottom of the hole. The action space of the cuttings removal control agent includes cuttings removal and drilling parameters, and the utility function is the cuttings removal mass flow rate minus the penalty term for exceeding the amount of cuttings accumulated at the bottom of the hole. Using the set of differential equations as the environmental model, the two agents select actions according to their respective policy networks. After executing the actions in the environmental model, they update the state variables, calculate the utility function values, and iteratively update the policy network parameters to maximize their own utility functions. This achieves the collaborative optimization control of the two coupled sub-processes of drilling and cuttings removal, overcoming the limitation of existing technologies that treat drilling and cuttings removal separately.

[0022] After each drilling cycle, this invention aligns the energy injection rate, crushing efficiency, and slag discharge smoothness in the data, along with the drilling parameters and slag discharge and drilling parameters output by the two agents within that cycle. These parameters are then substituted into the energy conservation equation and the mass conservation equation to calculate the energy conservation residual and the mass conservation residual. When any residual exceeds the inherent threshold of the equipment, the inherent parameters of the equipment in the differential equation set are corrected in reverse according to the direction and magnitude of the residual. Based on the corrected differential equation set, the strategy network parameters of the two agents are iteratively updated again until the residual converges to within the inherent threshold of the equipment. This forms a self-correction mechanism for equipment parameters based on closed-loop feedback of actual drilling data, further improving the matching accuracy between the control model and the actual working conditions. Attached Figure Description

[0023] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a flowchart of a method for improving the efficiency of a rotary drilling system for foundation pit excavation equipment according to the present invention; Figure 2 This is a structural diagram of an efficiency improvement system for a rotary drilling system for foundation pit excavation equipment according to the present invention; Figure 3 This is a schematic diagram of the structure of a computer system used to implement the methods, systems, and electronic devices of this application. Detailed Implementation

[0024] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0025] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0026] The first embodiment of the present invention proposes a method for improving the efficiency of a rotary drilling system for foundation pit excavation, the method comprising: Step S10: Collect the output data of multi-source sensors during the drilling process of the rotary drilling system, convert the output data into energy injection rate, crushing work efficiency and slag discharge smoothness, and align them under the time reference to obtain aligned data; Step S20: Construct energy conservation equations and mass conservation equations based on the alignment data; construct a two-dimensional phase space based on crushing work efficiency and slag discharge smoothness; extract the continuous coherence characteristics of phase trajectories in the two-dimensional phase space; and output the formation category based on the continuous coherence characteristics. Step S30: Construct a set of differential equations containing the state variable of the amount of rock cuttings accumulated at the bottom of the borehole based on the energy conservation equation, the matter conservation equation, and the formation type, and set physical matching inequality constraints between the slag removal capacity and the amount of rock cuttings accumulated in the set of differential equations. Step S40: Construct a drilling control agent and a cuttings removal control agent. Use the state variables in the differential equation system as shared state inputs. Set the utility function of the drilling control agent to the effective breaking power divided by the amount of cuttings accumulated. Set the utility function of the cuttings removal control agent to the cuttings mass flow rate minus the penalty term for exceeding the cuttings accumulation limit. Iteratively update the policy networks of the two agents to maximize their own utility functions and output drilling parameters and cuttings removal drilling parameters.

[0027] To more clearly explain the efficiency improvement method of the rotary drilling system for foundation pit excavation equipment according to the present invention, the following is in conjunction with... Figure 1 The steps in the embodiments of the present invention are described in detail below: Step S10: Collect the output data of multi-source sensors during the drilling process of the rotary drilling system, convert the output data into energy injection rate, crushing work efficiency and slag discharge smoothness, and align them under the time reference to obtain aligned data; In this embodiment, the output data is converted into energy injection rate, crushing work efficiency, and slag discharge smoothness, and aligned under a time reference to obtain aligned data. The method is as follows: Step S11: By performing time-series synchronization calibration on the raw output data collected by the multi-source sensors of the rotary drilling system, the phase shift introduced by the sampling frequency difference and transmission path delay is eliminated, and a synchronized data sequence is obtained. Step S12: Based on the synchronous data sequence, extract the energy injection rate, which represents the rate of change of the mechanical energy input of the rotary drilling head with time; the crushing work efficiency, which represents the crushing volume corresponding to the unit energy consumption when the drill bit crushes the rock and soil; and the slag discharge smoothness, which represents the smoothness of the movement of drill cuttings along the slag discharge channel. Step S13: The energy injection rate, crushing efficiency and slag discharge smoothness are embedded into the same time reference according to the unified time axis of the drilling process to form aligned data with time consistency.

[0028] In step S10, the output data of multiple sensors during the drilling process of the rotary drilling system are first collected, and these output data are converted into energy injection rate, crushing efficiency, and slag discharge smoothness with clear physical meaning. Then, they are aligned under a unified time reference to obtain aligned data. In specific implementation, the rotary drilling system performs drilling operations in the foundation pit project. Various types of sensors are pre-installed on its power head, drill rod, drill bit, and slag discharge mechanism. For example, torque and speed sensors are installed on the output shaft of the power head to measure the torque and speed output by the power head in real time; pressure sensors are installed on the top of the drill rod or at the connection between the power head and the drill rod to collect drilling pressure signals; displacement sensors or encoders are installed at the stroke positions of the drill rod or power head to obtain drilling speed and hoisting speed in real time; solid flow meters or punch flow meters are installed at the outlet of the slag discharge channel to detect the mass flow rate of the discharged rock cuttings; and vibration sensors are installed on the sidewalls of the slag discharge channel to monitor the movement status during the slag discharge process. All sensors are connected to the industrial control computer via a data acquisition card. Due to the different sampling frequencies of each sensor (e.g., the torque sensor has a sampling frequency of 100Hz, the displacement sensor has a sampling frequency of 50Hz, and the solid flow meter has a sampling frequency of 10Hz), and factors such as signal conditioning circuits and transmission cable lengths will introduce different time delays between the signals of each sensor, the data collected at the same physical moment will not be aligned on the time axis. Therefore, timing synchronization calibration is required.

[0029] In step S11, the raw output data collected by the multi-source sensors of the rotary drilling system are time-series synchronized and calibrated to eliminate phase offsets introduced by sampling frequency differences and transmission path delays, thereby obtaining a synchronized data sequence. Specifically, a unified high-precision time reference is set in the industrial control computer, for example, using a 1000Hz reference frequency to generate system timestamps. Each sensor's data packet, upon arrival at the data acquisition card, is triggered by a hardware interrupt to record the current timestamp, which is then stored in the buffer along with the sensor data. For sensors with lower sampling frequencies, the time axis of the highest sampling frequency is used as the reference, and cubic spline interpolation is employed for encryption, ensuring that all sensor data have the same output frequency on the time axis. For example, the original sampling frequency of the solid flow meter is 10Hz, meaning one data point is collected every 0.1 seconds, while the torque sensor's sampling frequency is 100Hz, meaning one data point is collected every 0.01 seconds. Through cubic spline interpolation, the interpolated estimate of the solid flow meter data is calculated at each 0.01-second time point, ensuring that all sensors have corresponding data values ​​at 0.01-second time intervals. Simultaneously, using pre-determined delay calibration values ​​for each sensor channel—for example, a torque sensor signal transmission delay of 5 milliseconds and a displacement sensor signal transmission delay of 8 milliseconds—delay compensation is applied to each data point by shifting the timestamp of the data point forward by the corresponding delay amount, eliminating phase shifts caused by different signal transmission path lengths. After this processing, a synchronized data sequence of all sensors is obtained under the same time reference. Each data point in this sequence fully contains all information, including torque, rotational speed, drilling pressure, drilling speed, hoisting speed, slag discharge mass flow rate, and vibration amplitude.

[0030] In step S12, based on the synchronous data sequence, the following are extracted: the energy injection rate, which characterizes the rate of change of the mechanical energy input of the rotary drilling head with time; the crushing work efficiency, which characterizes the crushing volume corresponding to the unit energy consumption when the drill teeth crush the rock and soil; and the slag discharge smoothness, which characterizes the smoothness of the movement of drill cuttings along the slag discharge channel.

[0031] The method for extracting the energy injection rate is as follows: multiply the torque value and rotational speed value at the same moment in the synchronous data sequence to obtain the mechanical power input of the power head at that moment, i.e., the energy injection rate, measured in watts. It reflects the total mechanical energy injected by the rotary drilling system into the drilling interface per unit time. For example, if the torque value collected at a certain moment is 35 kN·m and the rotational speed is 25 r / min, then first convert the rotational speed to angular velocity. 25 r / min is converted to 2.618 rad / s. The energy injection rate is 35 × 1000 × 2.618 = 91630 W, approximately 91.6 kW.

[0032] The method for extracting the crushing efficiency is as follows: First, calculate the theoretical crushing volume per unit time based on the drilling speed and borehole cross-sectional area in the synchronous data sequence. Assuming the borehole diameter is 1.2m, the borehole cross-sectional area is π×(0.6)²≈1.131m². 2 If the current drilling speed is 0.005 m / s, then the theoretical breaking volume per unit time is 1.131 × 0.005 = 0.005655 m³. 3 / s. Combined with the pre-input formation loosening coefficient, for example, if the current drilling formation is clay, and the loosening coefficient is taken as 1.25 according to the previous geological survey report, then the actual rock cuttings volume is 0.005655 × 1.25 = 0.007069 m³. 3 / s. Assuming the average fragmentation ratio of the soil and rock can be dynamically updated based on subsequent stratigraphic identification results, an empirical value can be used in the initial stage, for example, 6 × 10 for clay layers. 6 J / m 3 Therefore, the effective crushing energy consumption is 0.007069 × 6 × 10. 6 = 42414W. Dividing the effective crushing energy consumption by the energy injection rate, we get the crushing efficiency as 42414 ÷ 91630 ≈ 0.463, or 46.3%. This parameter reflects the ability of the rotary drilling system to convert mechanical energy into effective work for rock and soil crushing under the current geological conditions. To avoid the influence of transient fluctuations, a sliding window averaging method can be used to smooth the calculation results. For example, the arithmetic mean of the calculated values ​​at the most recent 5 moments can be taken as the crushing efficiency output at the current moment.

[0033] The method for extracting the slag discharge smoothness is as follows: The smoothness is calculated based on the measured slag discharge mass flow rate from the synchronous data sequence, combined with the theoretical maximum transport capacity of the current slag discharge channel. The theoretical maximum transport capacity is determined by the inner diameter of the slag discharge channel, the geometric parameters of the spiral blades, and the current drilling speed and spiral blade rotation speed. For example, the slag discharge channel is a spiral blade conveying structure with an inner diameter of 0.3m, an outer diameter of 0.28m, a spiral blade pitch of 0.25m, a spiral blade helix angle of 30°, a current drilling speed of 0.01m / s, a spiral blade rotation speed of 40r / min (i.e., 0.667r / s), and a slag bulk density of 1.8×10⁻⁶. 3 kg / m 3 According to the theory of screw conveyors, the volume of rock cuttings conveyed per revolution of the screw blade is the annular area between the outer and inner diameters of the screw blade multiplied by the pitch. The annular cross-sectional area is π × (0.14). 2 -0.15 2 The calculation here is incorrect and should be corrected. The actual calculation is the area of ​​the circle corresponding to the outer diameter of the propeller blade minus the area of ​​the circle corresponding to the inner diameter, i.e., π × (0.14). 2 )-π×(0.075 2)≈0.0616-0.0177=0.0439m 2 However, a more accurate formula for calculating the screw conveyor capacity should consider factors such as blade helix angle and filling rate. For ease of explanation, a simplified model is used here: Theoretical maximum mass flow rate = Annular cross-sectional area of ​​screw blade × Pitch × Screw blade rotation speed × Cuttings bulk density × Filling coefficient. Taking the filling coefficient as 0.6, the theoretical maximum mass flow rate = 0.0439 × 0.25 × 0.667 × 1800 × 0.6 ≈ 7.92 kg / s. If the current measured cuttings discharge mass flow rate is 5.2 kg / s, then the cuttings discharge smoothness is 5.2 ÷ 7.92 ≈ 0.657, or 65.7%. This value is between 0 and 1. The closer the value is to 1, the smoother the cuttings discharge. When the cuttings discharge smoothness is consistently low, it indicates that there is a risk of blockage in the cuttings discharge channel or insufficient conveying capacity.

[0034] In step S13, the calculated energy injection rate, crushing efficiency, and slag discharge smoothness are embedded into the same time reference according to the unified time axis of the drilling process, forming aligned data with time consistency. Specifically, three parallel time series arrays are established with the absolute time of the drilling process as the horizontal axis. The array length is the same as the synchronous data sequence. The energy injection rate, crushing efficiency, and slag discharge smoothness values ​​calculated at each moment in step S12 are stored in the corresponding arrays. The three arrays have identical time indices. For example, at timestamp t = 10.00 seconds, the energy injection rate is 91.6 kW, the crushing efficiency is 46.3%, and the slag discharge smoothness is 65.7%. At t = 10.01 seconds, the corresponding three values ​​are recalculated based on the synchronous data at that moment and stored sequentially in the arrays. All three values ​​corresponding to any index come from the same drilling moment and the same drilling depth, thus forming aligned data with time consistency. After the above processing, the sampling frequency differences and transmission delays between multiple sensors are eliminated, and the original physical signals are transformed into characteristic parameters with clear engineering physical meaning, providing a high-fidelity and highly consistent data foundation for the subsequent construction of energy conservation equations, material conservation equations, and stratigraphic category identification.

[0035] Step S20: Construct energy conservation equations and mass conservation equations based on the alignment data; construct a two-dimensional phase space based on crushing work efficiency and slag discharge smoothness; extract the continuous coherence characteristics of phase trajectories in the two-dimensional phase space; and output the formation category based on the continuous coherence characteristics. In this embodiment, the energy conservation equation and the matter conservation equation are constructed based on the alignment data, and the method is as follows: Step S21: The energy injection rate in the alignment data is used as the total mechanical energy input to the rotary drilling system. The effective energy consumed by rock and soil breaking is inverted by combining the breaking work efficiency in the alignment data with the drilling parameters. The difference between the energy injection rate and the effective energy is allocated to the frictional heat dissipation energy and the elastic wave dissipation energy. Based on the energy conservation relationship, an energy conservation equation is established that the energy injection rate is equal to the sum of the rock and soil breaking energy consumption, the frictional heat dissipation energy consumption and the elastic wave dissipation energy consumption. Step S22: Invert the amount of rock cuttings generated per unit time by using the crushing efficiency in the alignment data, and invert the amount of rock cuttings transported and discharged per unit time by combining the slag discharge smoothness in the alignment data with the slag discharge parameters. Based on the material conservation relationship, establish a material conservation equation in which the amount of rock cuttings generated is equal to the sum of the rate of change of the amount of rock cuttings transported and discharged and the amount of rock cuttings accumulated at the bottom of the borehole. Step S23: The energy consumption of rock and soil breaking in the energy conservation equation and the amount of rock cuttings generated in the material conservation equation are correlated and coupled through the rock and soil breaking specific energy parameter to form a set of simultaneous equations describing the interaction between energy distribution and material migration during the drilling process of the rotary drilling system.

[0036] A two-dimensional phase space is constructed based on crushing efficiency and slag discharge smoothness. The continuous coherence characteristics of the phase trajectories in the two-dimensional phase space are extracted. The formation category is then output based on these continuous coherence characteristics. The method is as follows: Step S24: Using crushing efficiency as the first dimension parameter of the two-dimensional phase space and slag discharge unobstructedness as the second dimension parameter of the two-dimensional phase space, the crushing efficiency value and slag discharge unobstructedness value corresponding to each moment during the drilling process are mapped as phase points in the two-dimensional phase space, and the phase points are connected in time sequence to form a phase trajectory. Step S25: Perform continuous cohomology analysis on the phase trajectory to extract the continuous cohomology features of the connected components and the generation and disappearance of the hole structure at different scales of the phase trajectory. Step S26: Input the extracted continuous coherence features into a pre-trained formation classifier, and output the formation category corresponding to the current drilling process through the classifier.

[0037] In step S20, energy conservation equations and mass conservation equations are first constructed based on the alignment data. Then, a two-dimensional phase space is constructed based on the crushing efficiency and slag discharge smoothness. The continuous coherence characteristics of the phase trajectory are extracted, and the formation category is output accordingly. In specific implementation, the alignment data obtained in step S13 is used as the basis. This alignment data contains three characteristic parameters with clear physical meaning at each drilling moment: energy injection rate, crushing efficiency, and slag discharge smoothness.

[0038] In step S21, the energy injection rate in the alignment data is used as the total mechanical energy input to the rotary drilling system. Let the energy injection rate in the alignment data at a certain moment be... The unit is watts. This is to align the breaking efficiency in the data. The effective energy consumed in rock and soil fracturing is inverted using drilling parameters. Drilling parameters include drill pressure, rotational speed, and drilling rate, among which drilling rate... It can be obtained directly from the synchronous data sequence. Effective breaking energy. The calculation method is as follows: First, calculate the theoretical crushing volume per unit time based on the drilling speed and borehole cross-sectional area, and then multiply it by the rock and soil crushing specific energy. To obtain effective crushing power, i.e. Among them, the crushing efficiency It is defined as the ratio of effective breaking power to total input power, therefore the effective breaking energy can be directly obtained using this relationship. The difference between the energy injection rate and the effective energy is allocated to frictional heat dissipation energy and elastic wave dissipation energy. Frictional heat dissipation energy The energy is mainly generated by the friction between the drill bit and the rock and soil, and between the drill rod and the borehole wall, resulting in elastic wave dissipation. This refers to the portion of energy lost due to drill bit impact or drill rod vibration radiating into the surrounding strata. Based on the law of conservation of energy, an energy conservation equation is established whereby the energy injection rate equals the sum of the energy consumed by rock and soil fracturing, the energy dissipated by frictional heat, and the energy dissipated by elastic waves: ; In practice, and Instead of being directly measured, these values ​​are allocated as residuals or through empirical models. For example, under a certain drilling condition, aligned data shows the energy injection rate. Crushing efficiency Then the effective crushing power Remaining energy This portion of energy is allocated as frictional heat dissipation and elastic wave dissipation. Frictional heat dissipation can be estimated based on drilling pressure and friction coefficient, while elastic wave dissipation is treated as a residual term.

[0039] In step S22, the amount of cuttings generated per unit time is inverted using the breaking work efficiency in the aligned data. Specifically, the cuttings generation rate... It can be obtained by the ratio of effective crushing power to the specific energy of rock and soil crushing, that is... Among them, the rock and soil fragmentation ratio energy This represents the energy required to break a unit volume or unit mass of rock and soil, expressed in J / kg or J / m³. This parameter is closely related to the formation type and can be dynamically matched using subsequent formation identification results. (This is used to align the slag discharge flow rate in the data.) The amount of cuttings transported and discharged per unit time is calculated by combining the cuttings removal parameters. These parameters include the drill string hoisting speed. Rotation speed of the helical blades Geometric parameters of the slag discharge channel, etc., and theoretical maximum slag discharge capacity. These parameters determine the actual slag discharge rate. This is the product of the theoretical maximum slag discharge capacity and the slag discharge smoothness, i.e. Based on the law of conservation of mass, a mass conservation equation is established whereby the amount of cuttings generated equals the sum of the rates of change of the amount of cuttings transported and discharged and the amount of cuttings accumulated at the bottom of the borehole: ; in The amount of rock cuttings deposited at the bottom of the borehole, in kilograms, and its rate of change. This reflects the net growth rate of rock cuttings accumulation at the bottom of the borehole. For example, at a certain moment, based on an effective breaking power of 42.4 kW and a rock-soil breaking energy of 6 × 10⁻⁶... 6 J / kg (clay layer) Calculate the rock cuttings formation rate Based on the slag discharge smoothness of 0.657 and the theoretical maximum slag discharge capacity of 7.92 kg / s, the actual slag discharge rate was calculated. The rate of change of the amount of rock cuttings accumulated at the bottom of the borehole. A negative value indicates that the amount of rock cuttings accumulating at the bottom of the borehole is decreasing, meaning that the cuttings removal capacity is greater than the rock cuttings generation rate, and the bottom of the borehole is in a cleared state.

[0040] In step S23, the energy consumption for rock and soil fracturing in the energy conservation equation and the amount of rock cuttings generated in the material conservation equation are coupled together using the rock and soil fracturing specific energy parameter to form a set of simultaneous equations describing the interaction between energy distribution and material migration during the rotary drilling system's drilling process. Specifically, the rock and soil fracturing specific energy... It is a key bridge connecting the energy domain and the material domain. On the one hand, it links effective fragmentation power with the rock cuttings generation rate, that is... On the other hand, effective crushing power is itself an important component of the energy conservation equation. Through this connection, the energy conservation equation and the matter conservation equation are no longer isolated, but constitute a coupled set of equations describing the drilling process of a rotary drilling system. In practical applications, this simultaneous equation set provides a physical basis for subsequent differential equation construction and intelligent agent control, enabling the optimization of drilling parameters and slag removal parameters to be carried out within a unified energy-matter framework.

[0041] In step S24, the crushing efficiency is used as the first dimension parameter of the two-dimensional phase space, and the slag discharge unobstructedness is used as the second dimension parameter. The crushing efficiency value and slag discharge unobstructedness value at each moment during the drilling process are mapped as phase points in the two-dimensional phase space, and the phase points are connected in chronological order to form a phase trajectory. In specific operation, the horizontal axis is set as the crushing efficiency. The vertical axis represents the slag discharge smoothness. Extracted from the aligned data at each drilling moment Points are plotted in a two-dimensional coordinate system, and adjacent points are connected by line segments in chronological order to form a phase trajectory that evolves over time. For example, at t=10.00 seconds, , This is marked as point A in phase space; at t = 10.01 seconds, if the calculated value is... , Mark point B, and connect A and B with a line segment; continue in this manner to form a continuous phase trajectory. The morphology of this phase trajectory reflects the synergistic evolution of the crushing efficiency and the smoothness of slag removal during the drilling process of the rotary drilling system. The phase trajectory exhibits different topological characteristics under different formation conditions.

[0042] In step S25, persistent cohomology analysis is performed on the phase trajectory to extract persistent cohomology features of the generation and disappearance of connected components and hole structures at different scales. Persistent cohomology is a computational topology method used to analyze the topological features of point cloud data or trajectory data at different scales. In practice, all phase points on the phase trajectory are considered as a point set, and a series of spherical neighborhoods with gradually increasing radii are constructed. The method observes how these spherical neighborhoods connect to form connected components as the radius increases, and whether hole (i.e., one-dimensional loop) structures appear. It also records the scale at which these connected components and hole structures are generated and disappear. Each topological feature corresponds to a generation scale and a disappearance scale; the difference between the two is the persistence of the feature. This persistence information is encoded into a persistent cohomology feature vector. For example, a barcode or persistence map for each dimension can be extracted and converted into a feature vector of fixed dimensions. In practice, existing persistent cohomology calculation libraries (such as GUDHI, Dionysus, etc.) can be used to calculate the phase trajectory point set to obtain the persistence information of each topological feature, which can then be used as input features for subsequent stratigraphic classification.

[0043] In step S26, the extracted continuous homology features are input into a pre-trained formation classifier, which outputs the formation category corresponding to the current drilling process. This formation classifier is a machine learning model, such as a support vector machine, random forest, or lightweight neural network classifier, pre-trained using a large amount of historical drilling data or experimental data before or in the early stages of construction. During training, continuous homology features collected under different formation conditions are used as input, and the corresponding formation category label is used as output, training the classifier to accurately identify formations based on the input features. During actual drilling, the continuous homology feature vector calculated from the current phase trajectory is input into the pre-trained classifier, which outputs the formation category, such as "clay layer," "sand layer," "gravel layer," or "rock layer." This formation category information is used for dynamic matching of soil and rock parameters (such as soil and rock fracture energy, friction coefficient, etc.) in subsequent differential equation systems, and also provides operators with formation change prompts, enabling the rotary drilling system to adaptively adjust its control strategy according to formation changes. By implementing the above step S20, online formation identification based on the energy-matter coupling model and topological feature extraction is realized, providing key geological prior information for subsequent closed-loop adaptive control.

[0044] Step S30: Construct a set of differential equations containing the state variable of the amount of rock cuttings accumulated at the bottom of the borehole based on the energy conservation equation, the matter conservation equation, and the formation type, and set physical matching inequality constraints between the slag removal capacity and the amount of rock cuttings accumulated in the set of differential equations. In this embodiment, the energy conservation equation and the matter conservation equation are constructed based on the alignment data, and the method is as follows: Step S31: Construct a set of differential equations including the state variable of the amount of rock debris deposited at the bottom of the borehole, based on the energy conservation equation, the mass conservation equation, and the formation type. The method is as follows: Step S32: Extract the differential equation form with the amount of rock debris accumulation at the bottom of the borehole as the state variable from the energy conservation equation and the matter conservation equation; The rate of change of rock cuttings accumulation at the bottom of the borehole is determined by the rock cuttings generation rate minus the rock cuttings discharge rate. The rock cuttings generation rate is obtained by inverting the rock and soil crushing energy consumption in the energy conservation equation combined with the rock and soil crushing specific energy. The rock cuttings discharge rate is obtained by inverting the slag discharge smoothness in the material conservation equation combined with the slag discharge capacity. Step S33: Match the corresponding rock and soil breaking energy and rock and soil friction coefficient from the pre-established rock and soil parameter library according to the stratum type, and substitute the matched parameters into the differential equation form to form a closed differential equation system with the amount of rock cuttings accumulated at the bottom of the hole as the state variable and the drilling parameters and the cuttings removal and drilling parameters as control variables.

[0045] The method for setting physical matching inequality constraints between slag removal capacity and rock cuttings accumulation in the system of differential equations is as follows: Step S34: Based on the inner diameter of the slag discharge channel, the pitch and helix angle of the spiral blades, the friction coefficient between the drill cuttings and the wall of the slag discharge channel, and the rotational power parameters and drilling speed of the spiral blades during the drilling process, construct a functional relationship between the upper limit of the slag discharge capacity and the drilling speed, the rotational speed of the spiral blades and the density of the drill cuttings. Step S35: Calculate the maximum slag discharge mass flow rate that can be output under the current slag discharge and drilling parameters using the aforementioned functional relationship; Step S36: Determine the actual required flow rate for slag discharge based on the amount of rock cuttings accumulated at the bottom of the borehole, the density of drill cuttings, and the cross-sectional area of ​​the slag discharge channel inlet. Step S37: Set a physical matching inequality constraint, limiting the actual slag discharge demand flow rate to no greater than the product of the maximum slag discharge mass flow rate and the slag discharge smoothness, and use this inequality constraint as the boundary condition of the feasible region of the differential equation system.

[0046] In step S30, firstly, based on the energy conservation equation, mass conservation equation, and identified formation type constructed in step S20, a set of differential equations is constructed, including the state variable of the amount of cuttings accumulated at the bottom of the borehole. A physical matching inequality constraint between the cuttings removal capacity and the amount of cuttings accumulated is then set in this set of differential equations. Specifically, based on the simultaneous equations obtained in step S23 and the formation type output in step S26, a mathematical model that can be used for dynamic control is further established.

[0047] In step S31, a set of differential equations including the state variable of the amount of rock cuttings deposited at the bottom of the borehole is constructed based on the energy conservation equation, the mass conservation equation, and the formation type. Specifically, the state variable of the amount of rock cuttings deposited at the bottom of the borehole is extracted from the energy conservation equation of step S21 and the mass conservation equation of step S22. This is a differential equation in the form of state variables. It is derived from the law of conservation of mass. It can be seen that the rate of change of the amount of cuttings accumulated at the bottom of the borehole is the cuttings generation rate minus the cuttings discharge rate, i.e. Among them, the rock cuttings formation rate The energy consumption of rock and soil fracturing in the energy conservation equation is combined with the specific energy of rock and soil fracturing, which is obtained through inversion. Specifically, the effective fracturing power is obtained from the energy conservation equation. The relationship between the rock cuttings generation rate and the effective breaking power is as follows: ,in Specific energy for rock and soil fragmentation. Rock cuttings ejection rate. It is obtained by inverting the slag discharge smoothness and slag discharge capacity in the mass conservation equation, that is... ,in To ensure smooth slag discharge, Let be the theoretical maximum cuttings mass flow rate under the current cuttings removal and drilling parameters. Therefore, the differential equation for the amount of cuttings accumulated at the bottom of the borehole can be written as:

[0048] In practice, the energy injection rate Crushing efficiency Slag discharge smoothness All are known time series derived from aligned data, while This depends on the slag removal and drilling parameters (such as drilling speed, auger blade speed, etc.). It depends on the stratigraphic type.

[0049] In step S32, based on the stratigraphic category output in step S26, the corresponding rock and soil fracturing energy and rock and soil friction coefficient are matched from a pre-established rock and soil parameter library. The rock and soil parameter library is a database established through preliminary geological surveys, laboratory geotechnical tests, or historical drilling data, recording typical parameter ranges for different stratigraphic categories. For example, for "clay layer," the matched rock and soil fracturing energy... 6×10 can be taken 6 J / m 3 coefficient of friction Take 0.35; for the "sand layer", Take 3×10 6 J / m 3 , Take 0.45; for the "gravel layer", Take 1.2 × 10 7 J / m 3 , Take 0.55; for "rock strata", Take 2.5 × 10 7 J / m 3 , We set the value to 0.65. Substituting the matched parameters into the above differential equation form, and simultaneously using drilling parameters (such as drilling pressure, rotation speed, and drilling rate) and cuttings removal / lifting parameters (such as lifting speed and auger blade rotation speed) as control variables, we form a closed system of differential equations. This system of differential equations uses the amount of cuttings accumulated at the bottom of the borehole... As state variables, with drilling parameters And slag removal and drilling parameters To control the input, it is possible to dynamically describe the variation of the amount of rock cuttings accumulated at the bottom of the borehole with time and control parameters.

[0050] In step S33, a physical matching inequality constraint is set in the differential equation system between the cuttings removal capacity and the amount of cuttings accumulation. This constraint ensures that the cuttings removal system can promptly remove the cuttings generated at the bottom of the borehole, avoiding excessive accumulation of cuttings that could lead to repeated breakage or stuck drill bit. Specifically, based on the inner diameter of the cuttings removal channel, the pitch and helix angle of the helical blades, the friction coefficient between the drill cuttings and the wall of the cuttings removal channel, and the rotational power parameters and drilling speed of the helical blades during drilling, a functional relationship is established between the upper limit of the cuttings removal capacity and the drilling speed, the rotational speed of the helical blades, and the density of the drill cuttings accumulation. Taking a common helical blade cuttings removal mechanism as an example, the theoretical maximum cuttings mass flow rate... It can be represented as:

[0051] in, Density of drill cuttings (unit: kg / m³) 3 ), The inner diameter of the slag discharge channel is (m). The diameter of the helical blade shaft (m). The pitch of the helical blade is (m). The rotational speed of the helical blades is (r / s). This is the filling coefficient, typically ranging from 0.3 to 0.7, and can be corrected based on the friction coefficient between drill cuttings and the wall surface, as well as the helix angle. Additionally, the drilling speed... It will also affect the cuttings removal capacity because the speed of the auger blades relative to the cuttings changes during the drilling process. In practical applications, the above formula can be multiplied by a drilling speed influence factor, for example... ,in It represents the annular flow area.

[0052] In step S34, the maximum output slag mass flow rate under the current slag removal and drilling parameters is calculated using the functional relationship established in step S33. For example, let the slag bulk density be... Inner diameter of slag discharge channel Spiral blade shaft diameter Then the annular flow area Pitch Rotation speed of the helical blade Drilling speed Fill factor The theoretical maximum slag discharge mass flow rate is: ; The calculation here uses a more complete formula. In practical applications, this value can be dynamically updated based on real-time measurements of the drilling speed and the rotational speed of the auger blades.

[0053] In step S35, the actual required flow rate for cuttings removal is determined based on the amount of cuttings accumulated at the bottom of the borehole, the density of drill cuttings, and the cross-sectional area of ​​the cuttings removal channel inlet. The actual required flow rate for cuttings removal refers to the minimum mass flow rate that the cuttings removal system needs to discharge to ensure that the amount of cuttings accumulated at the bottom of the borehole does not continue to increase. This required flow rate is equal to the cuttings generation rate, i.e. However, a certain amount of rock debris had already accumulated at the bottom of the borehole. If the accumulation is too high, a larger slag discharge flow rate may be needed to quickly reduce the accumulation. Therefore, the actual required flow rate can be expressed as: ； in The target stockpile size is set (usually a small safety value, such as 0.5 kg). The adjustment coefficient is positive. When the actual accumulation exceeds the target value, the demand flow rate is increased by an additional term proportional to the excess to accelerate the emptying of the bottom of the hole.

[0054] In step S36, a physical matching inequality constraint is set to limit the actual slag discharge demand flow rate to no more than the product of the maximum slag discharge mass flow rate and the slag discharge smoothness, that is: ; in The slag discharge smoothness reflects the actual conveying efficiency of the current slag discharge channel (which may be lower than the theoretical maximum due to blockages, adhesion, etc.). This inequality constraint serves as the boundary condition of the feasible region of the differential equation system. That is, the system is physically feasible only when the control parameters (drilling speed, auger blade speed, etc.) satisfy this constraint; otherwise, the amount of cuttings accumulated at the bottom of the hole cannot be effectively discharged, leading to a continuous increase in the accumulation, ultimately causing repeated breakage, increased drill bit wear, or even stuck drill accidents. In actual control, the actions of the drilling control agent and the slag discharge control agent must ensure that this inequality holds, or when the inequality is about to be violated, priority should be given to adjusting the slag discharge and drilling parameters (such as increasing the drilling speed or auger blade speed) to increase the efficiency. Alternatively, reduce drilling parameters to decrease the cuttings generation rate. This maintains the physical matching of the system. Through the implementation of step S30 above, a closed differential equation system with the amount of rock debris accumulation at the bottom of the borehole as the state variable is constructed, and inequality constraints with clear physical meaning are set, providing an accurate dynamic model and feasible domain boundary for subsequent collaborative optimization control of the two agents.

[0055] Step S40: Construct a drilling control agent and a cuttings removal control agent. Use the state variables in the differential equation system as shared state inputs. Set the utility function of the drilling control agent to the effective breaking power divided by the amount of cuttings accumulated. Set the utility function of the cuttings removal control agent to the cuttings mass flow rate minus the penalty term for exceeding the cuttings accumulation limit. Iteratively update the policy networks of the two agents to maximize their own utility functions and output drilling parameters and cuttings removal drilling parameters.

[0056] In this embodiment, the drilling control agent and the slag removal control agent are constructed as follows: Step S41: Construct a drilling control agent. The state space of the drilling control agent includes the state variables in the differential equation system. The action space of the drilling control agent includes drilling parameters. The utility function of the drilling control agent is set as the effective breaking power divided by the amount of rock cuttings accumulated at the bottom of the hole. Step S42: Construct a slag removal control agent. The state space of the slag removal control agent shares the state variables in the differential equation system with the drilling control agent. The action space of the slag removal control agent includes slag removal and drilling parameters. The utility function of the slag removal control agent is set as the slag removal mass flow rate minus the penalty term for excessive rock cuttings accumulation at the bottom of the hole. Step S43: Using the set of differential equations as an environment model, the drilling control agent and the slag removal control agent select actions according to their respective policy networks, update the state variables after executing the actions in the environment model, calculate the utility function value and iteratively update the policy network parameters to maximize their own utility functions, and output the optimized drilling parameters and slag removal drilling parameters.

[0057] In step S40, a drilling control agent and a cuttings removal control agent are constructed. The state variables in the differential equation system obtained in step S30 are used as shared state inputs, and their respective utility functions are set. The optimized drilling parameters and cuttings removal and drilling parameters are output through iterative updates of the policy network. In specific implementation, two deep reinforcement learning agents are first deployed on an industrial control computer or embedded industrial control computer, namely a drilling control agent and a cuttings removal control agent. These two agents share the same environment model, namely the model established in step S33 based on the amount of cuttings accumulated at the bottom of the borehole. It is a closed system of differential equations for state variables, while sharing state observations.

[0058] In step S41, a drilling control agent is constructed. The state space of this agent contains state variables from a system of differential equations, primarily the amount of cuttings accumulated at the bottom of the borehole. In addition, auxiliary state variables such as energy injection rate can be included. Crushing efficiency The current stratum category is encoded, etc., to enhance the agent's ability to perceive working conditions. The state vector can be represented as... The agent's action space includes drilling parameters, such as drilling pressure. (Unit: kN) Power head speed (Unit: r / min) Drilling speed setpoint (Unit: m / s), etc. To adapt to actual control requirements, the action space is usually discretized into several levels or continuous intervals. For example, the drilling pressure is continuously adjustable from 20kN to 80kN, and the rotation speed is continuously adjustable from 10r / min to 50r / min. The utility function (i.e., reward function) of the drilling control agent is set as the effective breaking power divided by the amount of cuttings accumulated at the bottom of the hole, specifically in the form of... ,in , The value is a small positive number (e.g., 0.001) to prevent division by zero errors. The purpose of this utility function is to ensure that the higher the effective breaking power and the smaller the amount of cuttings accumulated at the bottom of the borehole, the higher the drilling efficiency and the greater the reward for the agent. Conversely, if the accumulation increases, even with high breaking power, the reward will decrease significantly, thus incentivizing the drilling control agent to pursue high breaking efficiency while avoiding excessive cuttings accumulation at the bottom of the borehole.

[0059] In step S42, the construction of the slag discharge control intelligent agent specifically includes: Step S421: Set the state space of the slag discharge control agent. The state space is completely shared with the state space of the drilling control agent. Both include the state variable of the amount of rock cuttings accumulated at the bottom of the hole in the differential equation set, as well as the energy injection rate, crushing work efficiency, slag discharge smoothness and formation type obtained from the alignment data. Step S422: Set the action space of the slag discharge control agent. The action space includes slag discharge drilling parameters. The slag discharge drilling parameters include drilling speed and spiral blade rotation speed. The drilling speed and spiral blade rotation speed are adjustable within their respective continuous or discrete value ranges. Step S423, set the utility function of the slag discharge control agent as follows: ,in The actual slag discharge mass flow rate is determined by the product of the slag discharge smoothness and the theoretical maximum slag discharge mass flow rate. The preset penalty coefficient, This represents the amount of rock debris accumulated at the bottom of the borehole. The preset safety threshold for the amount of rock cuttings accumulation at the bottom of the borehole is when... No more than When the penalty is zero, Exceed The penalty is proportional to the amount exceeded. Step S424: Construct a policy network for the slag removal control agent. The policy network takes the state vector in the state space as input and the slag removal and drilling parameters in the action space as output. A deep reinforcement learning algorithm is used to iteratively update the policy network parameters so that the slag removal control agent outputs the slag removal and drilling parameters according to the current state in each control step and maximizes its cumulative utility function value.

[0060] The state space of this agent is completely shared with that of the drilling control agent, meaning it uses the same state space. As input, this ensures that the two agents make collaborative decisions based on the same environmental perception. The action space of the slag removal control agent includes slag removal and drilling parameters, such as drilling speed. (Unit: m / s) Rotational speed of the helical blade (Unit: r / s), etc. These parameters directly affect the theoretical maximum slag discharge mass flow rate. This, in turn, affects the actual slag discharge rate. The utility function of the slag discharge control agent is set as the slag discharge mass flow rate minus the penalty for exceeding the limit of rock cuttings accumulation at the bottom of the borehole, specifically in the form of... ,in The actual slag discharge mass flow rate (unit: kg / s). This is the penalty coefficient (e.g., 5). Let 2 kg be the safe threshold for the amount of cuttings accumulated at the bottom of the borehole. When the accumulation does not exceed the threshold, the penalty term is zero, and the agent only aims to maximize the cuttings discharge mass flow rate. When the accumulation exceeds the threshold, the excess portion will incur a penalty, forcing the cuttings control agent to increase the cuttings discharge intensity (increase the drilling speed or the auger blade speed) to reduce the accumulation as quickly as possible. The utility functions of the two agents are coupled but have the same goal: the drilling control agent wants to efficiently break up the rock and soil but avoid generating too much cuttings, while the cuttings discharge control agent wants to quickly remove cuttings but consumes energy. The cooperation between the two agents achieves Pareto optimality through shared state and independent reward design.

[0061] In step S43, the closed differential equation system obtained in step S33 is used as the environment model. The drilling control agent and the slag removal control agent select actions according to their respective current policy networks. After executing the actions in the environment model, the state variables are updated, and each agent calculates its utility function value and iteratively updates the policy network parameters to maximize its own utility function. Specifically, this includes: Step S431: Within each control step, obtain the current state vector from the alignment data. The state vector includes the amount of cuttings accumulated at the bottom of the borehole, energy injection rate, crushing efficiency, slag discharge smoothness, and formation type output by step S26. Step S432: The state vector is simultaneously input into the strategy network of the drilling control agent and the strategy network of the slag removal control agent. The strategy network of the drilling control agent outputs drilling parameter actions, which include drilling pressure, power head rotation speed and drilling speed setpoints. The strategy network of the slag removal control agent outputs slag removal and drilling parameter actions, which include drilling speed and auger blade rotation speed. Step S433: Substitute the drilling parameter action and the cuttings removal and drilling parameter action into the differential equation system, solve the differential equation system using the numerical integration method, obtain the updated value of the cuttings accumulation at the bottom of the hole at the next moment, and update the state vector according to the updated value of the cuttings accumulation at the bottom of the hole at the next moment and the corresponding time value in the alignment data. Step S434: Based on the updated state vector, calculate the utility function value of the drilling control agent as the first reward value. The first reward value is the effective breaking power divided by the amount of rock cuttings accumulated at the bottom of the hole. Calculate the utility function value of the slag discharge control agent as the second reward value. The second reward value is the slag discharge mass flow rate minus the penalty term for exceeding the limit of rock cuttings accumulation at the bottom of the hole. Store the state transition quadruples into their respective experience replay buffers. The state transition quadruples include the current state vector, the action performed, the corresponding reward value, and the state vector at the next moment. Step S435: Every preset number of steps, a batch of data is randomly sampled from the experience replay buffer, and the commentator network and actor network of the drilling control agent and the commentator network and actor network of the slag removal control agent are updated respectively, so that the drilling control agent aims to maximize its cumulative first reward value and the slag removal control agent aims to maximize its cumulative second reward value. The updates are iteratively updated until the policy network converges, and the optimized drilling parameters and slag removal drilling parameters are output.

[0062] In practical implementation, algorithms such as Multi-Agent Deep Deterministic Policy Gradient (MADDPG) or Independent Proximal Policy Optimization (IPPO) can be used for training. Taking MADDPG as an example, each agent includes an actor network for outputting actions and a critic network for evaluating the value of the actions. At each control step... Within (e.g., 0.5 seconds), perform the following operations: First, obtain the current state from the aligned data. Including the amount of rock debris accumulated at the bottom of the borehole (Obtained recursively from the mass conservation equation or indirectly estimated via sensors), energy injection rate, crushing efficiency, slag discharge smoothness, and formation type. The actor network of the drilling control agent is based on... Output drilling action The actor network of the slag discharge control intelligent agent is based on the same Output slag removal action These two actions are input into the environment model, i.e., substituted into the system of differential equations. The amount of rock debris accumulated at the bottom of the borehole at the next time step can be obtained by integrating using the Euler method or the fourth-order Runge-Kutta method. Simultaneously, based on the environmental model and actions, updated values ​​for parameters such as energy injection rate, crushing efficiency, and slag discharge smoothness are calculated (these parameters can also be obtained from actual sensor feedback; a simulation model can be used during the training phase). Then, the reward values ​​for the two agents are calculated separately. and Transition to the state Each agent's experience replay buffer is stored. Every few steps, a batch of data is randomly sampled from the buffers to update the critic network and actor network of each agent. The critic network learns the state-action value function by minimizing the temporal difference error, while the actor network maximizes the expected cumulative reward through policy gradient ascent. As training progresses, the policy networks of both agents gradually converge to the optimal policy: the drilling control agent learns to adjust the drilling pressure, rotation speed, and feed rate according to the formation and stockpile volume to maximize the ratio of effective breaking power to stockpile volume; the cuttings removal control agent learns to rationally control the drilling speed and auger blade rotation speed to ensure smooth cuttings removal while avoiding excessive energy consumption. After training, the actor networks of both agents are deployed to the controller of the actual rotary drilling system.

[0063] During real-time control, every control cycle (e.g., 0.5 seconds), the controller reads the current sensor data and calculates the state vector. Recommended drilling parameters and cuttings removal / drilling parameters are then obtained through forward propagation via two actor networks. These parameters are sent as analog or digital commands to the hydraulic actuators and variable frequency motors of the rotary drilling system, achieving integrated adaptive optimal control of drilling and cuttings removal. For example, in clay layers, if the amount of cuttings at the bottom of the hole is low, the drilling control agent may output higher drilling pressure and rotation speed to improve breaking efficiency; while the cuttings removal control agent outputs a moderate drilling speed and auger blade rotation speed, ensuring that the cuttings removal capacity is slightly greater than the cuttings generation rate, maintaining the accumulation within a safe range. If the system suddenly enters a sandy layer, the change in breaking efficiency leads to an increased cuttings generation rate. After updating the state variables, the drilling control agent will appropriately reduce the drilling speed to avoid excessive accumulation, while the cuttings removal control agent will automatically increase the drilling speed and auger blade rotation speed to enhance cuttings removal capacity, thus achieving rapid adaptive adjustment. By implementing step S40 above, the coordinated optimization control of the two coupled sub-processes of drilling and slag removal is achieved, which significantly improves the construction efficiency under complex geological conditions.

[0064] In this embodiment, after step S40, step S50 is further included, which includes: After each drilling cycle, the energy conservation equation and the mass conservation equation are substituted into the equations to calculate the energy conservation residual and the mass conservation residual. When any residual exceeds the inherent threshold of the equipment, the inherent parameters of the equipment in the differential equation set are adjusted, and the policy network of the two agents is iteratively updated again. The method is as follows: Step S51: The energy injection rate, crushing efficiency and slag discharge smoothness in the alignment data at the end of each drilling cycle, together with the drilling parameters and slag discharge lifting parameters output by the drilling control agent and the slag discharge control agent in that cycle, are substituted into the energy conservation equation and the material conservation equation. The difference between the two ends of the equation is calculated to obtain the energy conservation residual and the material conservation residual. Step S52: Compare the energy conservation residual and the material conservation residual with the preset equipment inherent thresholds respectively. When any residual exceeds the corresponding threshold, correct the equipment inherent parameters in the differential equation set in reverse according to the direction and magnitude of the residual, so that the corrected differential equation set matches the actual observation data of the current drilling cycle. Step S53: Based on the corrected differential equation set, iteratively update the strategy network parameters of the drilling control agent and the slag removal control agent until the residual converges to within the inherent threshold of the equipment.

[0065] In step S50, after each drilling cycle, the energy conservation equation and the mass conservation equation are substituted into the equations to calculate the energy conservation residual and the mass conservation residual. When any residual exceeds the inherent threshold of the equipment, the inherent parameters of the equipment in the differential equation set are adjusted, and the policy network of the two agents is iteratively updated again. In specific implementation, a drilling cycle is usually defined as the process of the drill bit drilling from the bottom of the hole to a predetermined depth (e.g., the length of a single drill rod, usually 3 to 6 meters), then lifting the drill bit to remove the cuttings, and then drilling down again to start the next cycle. At the end of each drilling cycle, the industrial control computer collects all the alignment data, control parameters output by the agents, and sensor measurement data within the cycle, and performs residual analysis and parameter correction.

[0066] In step S51, the energy injection rate, crushing efficiency, and slag removal smoothness from the aligned data at the end of each drilling cycle, along with the drilling parameters and slag removal lifting parameters output by the drilling control agent and slag removal control agent within that cycle, are substituted into the energy conservation equation and the mass conservation equation. The difference between the two ends of the equations is then calculated to obtain the energy conservation residual and the mass conservation residual. In specific operation, the duration of one drilling cycle is set as follows: From time arrive First, the energy injection rate at all sampling moments within this loop. Effective crushing power Frictional heat dissipation power Elastic wave dissipation power Integrating, we obtain the total energy input and energy consumption for the entire cycle. The energy conservation equation can be written in integral form: ； In actual calculations, the total input energy on the left side... It can be obtained directly from the integral of the product of torque and speed; the first term on the right is the effective crushing energy consumption. It can be obtained by integrating the product of the crushing work efficiency and the energy injection rate; the sum of the last two terms on the right is the residual term, which is usually not measured separately, but is obtained by subtracting the effective crushing energy consumption from the total input. However, in order to verify the accuracy of the model, the frictional heat dissipation and elastic wave dissipation can be estimated separately using empirical formulas, and then the energy conservation residual can be calculated. Similarly, the integral form of the mass conservation equation is: ; in This represents the change in the amount of cuttings accumulated at the bottom of the borehole. (Total mass of cuttings generated on the left side) The effective crushing energy consumption can be divided by the specific energy of rock and soil crushing. The integral is obtained as follows: Total mass of rock cuttings discharged on the right side It can be obtained by integrating the product of slag discharge smoothness and theoretical maximum slag discharge capacity; change in accumulation volume. It can be obtained from the difference between the measured or estimated accumulation at the beginning and end of the drilling cycle. (Matter conservation residual) For example, in a drilling cycle through a clay layer, the total input energy is calculated. Effective crushing energy consumption Estimated frictional dissipation Elastic wave dissipation Then the residual of energy conservation In terms of materials, the total mass of rock fragments generated... Total discharge mass Changes in accumulation Then the residual of matter conservation .

[0067] In step S52, the energy conservation residual and the mass conservation residual are compared with preset equipment inherent thresholds. When either residual exceeds the corresponding threshold, the equipment inherent parameters in the differential equation system are corrected in reverse according to the direction and magnitude of the residual, so that the corrected differential equation system matches the actual observation data of the current drilling cycle. The equipment inherent parameters include, but are not limited to, the rock and soil fracturing specific energy. Initial reference value, soil friction coefficient slag discharge channel filling coefficient Elastic wave escape coefficient, etc. The preset threshold is set according to equipment accuracy and engineering requirements; for example, the energy residual threshold is set to 5% of the total input energy. The residual threshold for matter is set at 3% of the total generated mass, i.e. In the example above, the total input energy... 5% of Actual residual If the mass is less than the threshold, energy conservation is satisfied; 3% of the total generated mass of 2000kg is 60kg, and the actual material residual of 50kg is less than 60kg, which also meets the requirements and requires no correction. If any residual exceeds the threshold, for example, if the material residual reaches 100kg, exceeding 60kg, it indicates a deviation between the model parameters and the actual working conditions. In this case, the parameters need to be corrected in reverse: if The calculated value is too high, while the sum of the actual changes in discharge and accumulation is too low, which may be due to the specific energy of soil and rock fracturing. If the value is too small (leading to an overestimation of the generated quality), it should be increased appropriately. If the estimated slag discharge capacity is too high, it will lead to... If the calculated value is too high, it may be due to the fill factor. The value is too large and should be reduced appropriately. The correction method can employ gradient descent or simple proportional correction, for example... The corrected inherent parameters of the device are stored in the parameter library and used to update the system of differential equations.

[0068] In step S53, the policy network parameters of the drilling control agent and the slag removal control agent are iteratively updated again based on the corrected differential equation system until the residuals converge to within the inherent threshold of the equipment. In practice, when residuals exceed the limit and parameter correction is completed, it means that the environmental model (differential equation system) has changed, and the previously trained agent policies may no longer be optimal or even applicable. Therefore, it is necessary to use the corrected differential equation system as a new environmental model to retrain or fine-tune the policy networks of the two agents online. Typically, it is not necessary to train from scratch; instead, based on the original policy network parameters, several steps of gradient updates are performed under the new environmental model to adapt the policy networks to the corrected model.

[0069] For example, using online reinforcement learning, in the next drill-up loop, the agent continues to interact with the environment, but at this time the environment model... or The system has been updated. The agent collects new interaction data and updates the policy network parameters periodically, gradually improving the utility functions of both agents under the new parameter environment. This process is repeated until the energy and mass conservation residuals of multiple consecutive drilling cycles stabilize within the threshold, indicating that the differential equations match the actual physical process well and the agent's strategy has adapted to the actual working conditions. Through the implementation of step S50, a self-correcting mechanism for equipment parameters based on closed-loop feedback from actual drilling data is formed. This effectively solves the model mismatch problem caused by factors such as formation parameter uncertainty and equipment wear, further improving the matching accuracy between the control model and the actual working conditions, and ensuring the agent's adaptability and robustness in long-term operation.

[0070] like Figure 2 As shown, a second embodiment of the present invention provides an efficiency improvement system for a rotary drilling system for foundation pit excavation, used to implement a method for improving the efficiency of a rotary drilling system for foundation pit excavation, comprising: The data acquisition and processing module is used to acquire the output data of multi-source sensors during the drilling process of the rotary drilling system, convert the output data into energy injection rate, crushing work efficiency and slag discharge smoothness, and align them under the time reference to obtain aligned data. The equation construction and formation identification module is used to construct energy conservation equations and mass conservation equations based on the alignment data, construct a two-dimensional phase space based on crushing work efficiency and slag discharge smoothness, extract the continuous coherence features of phase trajectories in the two-dimensional phase space, and output the formation category based on the continuous coherence features. The differential equation construction and constraint module is used to construct a set of differential equations containing the state variable of the amount of rock cuttings at the bottom of the borehole based on the energy conservation equation, the matter conservation equation and the formation type, and to set physical matching inequality constraints between the slag removal capacity and the amount of rock cuttings in the set of differential equations. The intelligent agent control module includes a drilling control intelligent agent and a cuttings removal control intelligent agent. The intelligent agent control module uses the state variables in the differential equation system as shared state inputs to the drilling control intelligent agent and the cuttings removal control intelligent agent. The utility function of the drilling control intelligent agent is set as the effective breaking power divided by the amount of cuttings accumulation, and the utility function of the cuttings removal control intelligent agent is set as the cuttings removal mass flow rate minus the penalty term for exceeding the cuttings accumulation limit. The two intelligent agents iteratively update their respective policy networks to maximize their own utility functions and output drilling parameters and cuttings removal and drilling lifting parameters.

[0071] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process and related explanations of the methods described above can be found in the corresponding processes in the foregoing system embodiments, and will not be repeated here.

[0072] It should be noted that the methods and systems provided in the above embodiments are only illustrative examples of the division of functional modules. In practical applications, the functions can be assigned to different functional modules as needed, that is, the modules or steps in the embodiments of the present invention can be further decomposed or combined. For example, the modules in the above embodiments can be merged into one module, or further divided into multiple sub-modules to complete all or part of the functions described above. The names of the modules and steps involved in the embodiments of the present invention are only for distinguishing the various modules or steps and are not considered as an improper limitation of the present invention.

[0073] A device according to a third embodiment of the present invention includes: At least one processor; and a memory communicatively connected to at least one of the processors; The memory stores instructions that can be executed by the processor to implement the above-described method for improving the efficiency of a rotary drilling system for foundation pit excavation.

[0074] A fourth embodiment of the present invention provides a computer-readable storage medium storing computer instructions, which are executed by the computer to implement the above-described method for improving the efficiency of a rotary drilling system for foundation pit excavation equipment.

[0075] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process and related descriptions of the storage device and processing device described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0076] The following is for reference. Figure 3 It shows a schematic diagram of the structure of a computer system for implementing embodiments of the systems, methods, and electronic devices of this application. Figure 3 The server shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0077] like Figure 3As shown, the computer system includes a Central Processing Unit (CPU) 301, which can perform various appropriate actions and processes based on programs stored in Read Only Memory (ROM) 302 or programs loaded from storage section 308 into Random Access Memory (RAM) 303. The RAM 303 also stores various programs and data required for system operation. The CPU 301, ROM 302, and RAM 303 are interconnected via a bus 304. An Input / Output (I / O) interface 305 is also connected to the bus 304.

[0078] The following components are connected to I / O interface 305: an input section 306 including a keyboard, mouse, etc.; an output section 307 including a cathode ray tube (CRT), liquid crystal display (LCD), and speakers, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card and a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to I / O interface 305 as needed. Removable media 311, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 310 as needed so that computer programs read from them can be installed into storage section 308 as needed.

[0079] Specifically, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 309, and / or installed from removable medium 311. When the computer program is executed by central processing unit (CPU) 301, it performs the functions defined in the methods of this application. It should be noted that the computer-readable medium described above in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on a computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0080] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0081] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0082] The terms “first”, “second”, etc., are used to distinguish similar objects, not to describe or indicate a specific order or sequence.

[0083] The term "comprising" or any other similar term is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus / device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent in such process, method, article, or apparatus / device.

[0084] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A method for improving the efficiency of a rotary drilling system for foundation pit excavation, characterized in that, Includes the following steps: During the drilling process of the rotary drilling system, the output data of multiple source sensors are collected, and the output data is converted into energy injection rate, crushing work efficiency and slag discharge smoothness, and then aligned under the time reference to obtain aligned data. Based on the alignment data, energy conservation equations and mass conservation equations are constructed. A two-dimensional phase space is constructed based on the crushing efficiency and slag discharge smoothness. The continuous coherence characteristics of the phase trajectories in the two-dimensional phase space are extracted. The formation category is output based on the continuous coherence characteristics. Based on the energy conservation equation, the mass conservation equation, and the formation type, a set of differential equations is constructed, which includes the state variable of the amount of rock cuttings accumulated at the bottom of the borehole. In the set of differential equations, a physical matching inequality constraint between the slag removal capacity and the amount of rock cuttings accumulated is set. Construct drilling control agent and slag removal control agent, using the state variables in the differential equation system as shared state inputs. Set the utility function of drilling control agent as effective breaking power divided by cuttings accumulation, and set the utility function of slag removal control agent as slag removal mass flow rate minus the penalty term for exceeding cuttings accumulation limit. The two agents iteratively update their respective policy networks to maximize their own utility functions, and output drilling parameters and slag removal drilling parameters.

2. The method for improving the efficiency of a rotary drilling system for foundation pit excavation according to claim 1, characterized in that, The output data is converted into energy injection rate, crushing efficiency, and slag discharge smoothness, and aligned under a time reference to obtain aligned data. The method is as follows: By performing time-series synchronization calibration on the raw output data collected by the multi-source sensors of the rotary drilling system, the phase shift introduced by the sampling frequency difference and transmission path delay is eliminated, and a synchronized data sequence is obtained. Based on the synchronous data sequence, the following were extracted: the energy injection rate, which characterizes the rate of change of mechanical energy input to the rotary drilling head over time; the crushing work efficiency, which characterizes the crushing volume per unit energy consumption when the drill teeth crush rock and soil; and the slag discharge smoothness, which characterizes the smoothness of the movement of drill cuttings along the slag discharge channel. The energy injection rate, crushing efficiency, and slag discharge smoothness are embedded into the same time reference according to the unified time axis of the drilling process to form aligned data with time consistency.

3. The method for improving the efficiency of a rotary drilling system for foundation pit excavation according to claim 1, characterized in that, The method for constructing energy conservation equations and matter conservation equations based on the aligned data is as follows: The energy injection rate in the alignment data is used as the total mechanical energy input to the rotary drilling system. The effective energy consumed by rock and soil breaking is inverted by combining the breaking work efficiency in the alignment data with the drilling parameters. The difference between the energy injection rate and the effective energy is allocated to the energy dissipated by frictional heat and the energy dissipated by elastic waves. Based on the energy conservation relationship, an energy conservation equation is established that the energy injection rate is equal to the sum of the energy consumption of rock and soil breaking, the energy consumption of frictional heat dissipation, and the energy consumption of elastic waves. The amount of cuttings generated per unit time is inverted by the crushing efficiency in the aligned data, and the amount of cuttings transported and discharged per unit time is inverted by the slag discharge smoothness in the aligned data combined with the slag discharge parameters. Based on the material conservation relationship, a material conservation equation is established in which the amount of cuttings generated is equal to the sum of the rate of change of the amount of cuttings transported and discharged and the amount of cuttings accumulated at the bottom of the borehole. The energy consumption of rock and soil breaking in the energy conservation equation and the amount of rock cuttings generated in the material conservation equation are correlated and coupled through the rock and soil breaking specific energy parameter to form a set of simultaneous equations describing the interaction between energy distribution and material migration during the drilling process of the rotary drilling system.

4. The method for improving the efficiency of a rotary drilling system for foundation pit excavation according to claim 1, characterized in that, A two-dimensional phase space is constructed based on crushing efficiency and slag discharge smoothness. The continuous coherence characteristics of the phase trajectories in the two-dimensional phase space are extracted. The formation category is then output based on these continuous coherence characteristics. The method is as follows: Using crushing efficiency as the first dimension parameter of the two-dimensional phase space and slag discharge unobstructedness as the second dimension parameter, the crushing efficiency value and slag discharge unobstructedness value corresponding to each moment during the drilling process are mapped as phase points in the two-dimensional phase space, and the phase points are connected in time sequence to form a phase trajectory. The phase trajectory is subjected to continuous cohomology analysis to extract the continuous cohomology features of the connected components and the generation and disappearance of the void structure at different scales. The extracted continuous coherence features are input into a pre-trained formation classifier, which outputs the formation category corresponding to the current drilling process.

5. The method for improving the efficiency of a rotary drilling system for foundation pit excavation according to claim 1, characterized in that, Based on the energy conservation equation, the mass conservation equation, and the formation type, a system of differential equations is constructed, including the state variable of the amount of rock debris deposited at the bottom of the borehole. The method is as follows: Extract the differential equation form with the amount of rock debris accumulation at the bottom of the borehole as the state variable from the energy conservation equation and the matter conservation equation; The rate of change of rock cuttings accumulation at the bottom of the borehole is determined by the rock cuttings generation rate minus the rock cuttings discharge rate. The rock cuttings generation rate is obtained by inverting the rock and soil crushing energy consumption in the energy conservation equation combined with the rock and soil crushing specific energy. The rock cuttings discharge rate is obtained by inverting the slag discharge smoothness in the material conservation equation combined with the slag discharge capacity. Based on the geological formation, the corresponding rock and soil breaking energy and rock and soil friction coefficient are matched from the pre-established rock and soil parameter library. The matched parameters are substituted into the differential equation form to form a closed differential equation system with the amount of rock cuttings accumulated at the bottom of the borehole as the state variable and the drilling parameters and the cuttings removal and drilling parameters as control variables.

6. The method for improving the efficiency of a rotary drilling system for foundation pit excavation according to claim 1, characterized in that, The method for setting physical matching inequality constraints between slag removal capacity and rock cuttings accumulation in the system of differential equations is as follows: Based on the inner diameter of the slag discharge channel, the pitch and helix angle of the spiral blades, the friction coefficient between the drill cuttings and the wall of the slag discharge channel, and the rotational power parameters and drilling speed of the spiral blades during the drilling process, a functional relationship is established between the upper limit of the slag discharge capacity and the drilling speed, the rotational speed of the spiral blades, and the density of the drill cuttings. The maximum slag discharge mass flow rate that can be output under the current slag discharge and drilling parameters is calculated using the aforementioned functional relationship; The actual required flow rate for slag removal is determined based on the amount of rock cuttings accumulated at the bottom of the borehole, the density of drill cuttings, and the cross-sectional area of ​​the slag removal channel inlet. A physical matching inequality constraint is set to limit the actual slag discharge demand flow rate to no more than the product of the maximum slag discharge mass flow rate and the slag discharge smoothness, and this inequality constraint is used as the boundary condition of the feasible region of the differential equation system.

7. The method for improving the efficiency of a rotary drilling system for foundation pit excavation according to claim 1, characterized in that, The method for constructing drilling control agents and slag removal control agents is as follows: A drilling control agent is constructed. The state space of the drilling control agent includes the state variables in the differential equation system. The action space of the drilling control agent includes the drilling parameters. The utility function of the drilling control agent is set as the effective breaking power divided by the amount of rock cuttings accumulated at the bottom of the hole. A slag removal control intelligent agent is constructed. The state space of the slag removal control intelligent agent shares the state variables in the differential equation system with the drilling control intelligent agent. The action space of the slag removal control intelligent agent includes slag removal and drilling parameters. The utility function of the slag removal control intelligent agent is set as the slag removal mass flow rate minus the penalty term for excessive rock cuttings accumulation at the bottom of the hole. Using the system of differential equations as an environmental model, the drilling control agent and the slag removal control agent select actions according to their respective policy networks. After executing the actions in the environmental model, they update the state variables, calculate the utility function value, and iteratively update the policy network parameters to maximize their own utility functions, and output the optimized drilling parameters and slag removal drilling parameters.

8. The method for improving the efficiency of a rotary drilling system for foundation pit excavation according to claim 1, characterized in that, After each drilling cycle, the energy conservation equation and the mass conservation equation are substituted into the equations to calculate the energy conservation residual and the mass conservation residual. When any residual exceeds the inherent threshold of the equipment, the inherent parameters of the equipment in the differential equation set are adjusted, and the policy network of the two agents is iteratively updated again. The method is as follows: The energy injection rate, crushing efficiency, and slag discharge smoothness in the aligned data at the end of each drilling cycle, together with the drilling parameters and slag discharge lifting parameters output by the drilling control agent and the slag discharge control agent in that cycle, are substituted into the energy conservation equation and the mass conservation equation. The difference between the two ends of the equation is then calculated to obtain the energy conservation residual and the mass conservation residual. The energy conservation residual and the material conservation residual are compared with the preset equipment inherent thresholds respectively. When any residual exceeds the corresponding threshold, the equipment inherent parameters in the differential equation system are corrected in reverse according to the direction and magnitude of the residual, so that the corrected differential equation system matches the actual observation data of the current drilling cycle. The strategy network parameters of the drilling control agent and the slag removal control agent are iteratively updated based on the modified differential equation system until the residuals converge to within the inherent threshold of the equipment.

9. An efficiency improvement system for a rotary drilling rig for foundation pit excavation, used to implement the efficiency improvement method for a rotary drilling rig for foundation pit excavation as described in any one of claims 1-8, characterized in that, include: The data acquisition and processing module is used to acquire the output data of multi-source sensors during the drilling process of the rotary drilling system, convert the output data into energy injection rate, crushing work efficiency and slag discharge smoothness, and align them under the time reference to obtain aligned data. The equation construction and formation identification module is used to construct energy conservation equations and mass conservation equations based on the alignment data, construct a two-dimensional phase space based on crushing work efficiency and slag discharge smoothness, extract the continuous coherence features of phase trajectories in the two-dimensional phase space, and output the formation category based on the continuous coherence features. The differential equation construction and constraint module is used to construct a set of differential equations containing the state variable of the amount of rock cuttings at the bottom of the borehole based on the energy conservation equation, the matter conservation equation and the formation type, and to set physical matching inequality constraints between the slag removal capacity and the amount of rock cuttings in the set of differential equations. The intelligent agent control module includes a drilling control intelligent agent and a cuttings removal control intelligent agent. The intelligent agent control module uses the state variables in the differential equation system as shared state inputs to the drilling control intelligent agent and the cuttings removal control intelligent agent. The utility function of the drilling control intelligent agent is set as the effective breaking power divided by the amount of cuttings accumulation, and the utility function of the cuttings removal control intelligent agent is set as the cuttings removal mass flow rate minus the penalty term for exceeding the cuttings accumulation limit. The two intelligent agents iteratively update their respective policy networks to maximize their own utility functions and output drilling parameters and cuttings removal and drilling lifting parameters.

10. A device, characterized in that, include: At least one processor; and a memory communicatively connected to at least one of the processors; The memory stores instructions that can be executed by the processor to implement the efficiency improvement method of the rotary drilling system for foundation pit excavation equipment as described in any one of claims 1-8.

Images