Intelligent collaborative control system for high slope anchor frame beam winch platform

Abstract

The present application relates to the field of data analysis and control technology, and discloses an intelligent collaborative control system for a high-slope anchor frame beam winch platform. The system includes: a processing module that collects multi-sensor data and initializes the system, obtaining winch parameters and platform posture; an optimization module that plans drilling positions and paths, generates platform trajectories and winch sequences; an identification module that analyzes winch characteristics and compensates for differences; a control module that implements multi-point synchronous control through a neural network; a leveling module that monitors and adjusts the platform posture; and a compensation module that dynamically compensates for the drilling process and acquires geological data. The present application implements high-precision synchronous control of multiple winches on a slope anchor drilling platform, eliminating the problem of platform tilting and shaking caused by insufficient synchronization between winches, thereby improving drilling accuracy and construction safety.

Description

Technical Field

[0001] The present application relates to the field of data analysis and control technology, and in particular to an intelligent collaborative management and control system for a high-slope anchor frame beam winch platform. Background Art

[0002] During slope construction, anchor drilling operations are often required on steep slope surfaces to install anchor bolts and enhance slope stability. Traditional slope anchor drilling platforms typically use manually operated winches to adjust the platform's position, relying on the operator's experience and judgment to control the operating status of each winch to achieve platform movement and positioning. This operating method has been used in slope engineering for many years. The basic process includes initial platform installation, manually controlling the winch to move the platform to the target location, manually adjusting the platform's posture and stability, drilling, and then moving to the next location after completion. In flat terrain or simple working conditions, this manual operation method can meet basic construction needs.

[0003] However, traditional slope anchor drilling platforms have significant shortcomings in practical application, particularly in complex terrain and demanding working conditions. First, manually operating the winch to adjust the platform's position suffers from low positioning accuracy, making it difficult to meet high-precision drilling requirements. Second, operational efficiency is poor, requiring the coordinated efforts of multiple workers and a time-consuming adjustment process. Third, safety risks are high, as workers must operate the winch from elevated locations, posing a risk of falls. Most critically, in multi-point winch systems, insufficient synchronization between the winches often causes the platform to tilt and wobble, severely impacting drilling accuracy and construction safety. During the drilling process, vibration and impact forces generated by the drill rig can easily lead to platform instability. Furthermore, traditional systems cannot automatically adjust drilling parameters based on geological conditions, requiring workers to rely on empirical judgment, which is inefficient and prone to damage to drilling tools. Currently, there is a lack of effective multi-point winch synchronization control technology on the market, which cannot meet the control requirements of high-precision, intelligent slope anchor drilling platforms. Summary of the Invention

[0004] The present application provides an intelligent collaborative management and control system for a high-slope anchor frame beam winch platform, which is used to achieve high-precision synchronous control of multiple winches on the slope anchor drilling platform, eliminating the problem of platform tilting and shaking caused by insufficient synchronization between the winches, thereby improving drilling accuracy and construction safety.

[0005] The present application provides an intelligent collaborative management and control system for a high-slope anchor frame beam hoist platform, which includes:

[0006] The processing module is used to collect and initialize the multi-sensor data of the slope anchor drilling platform to obtain the mechanical characteristic parameters of the winch, coordinate system calibration data and the initial value of the drilling platform posture;

[0007] An optimization module is used to perform drilling position planning and path optimization processing on the three-dimensional digital model of the slope based on the coordinate system calibration data, so as to obtain the movement trajectory of the drilling platform and the control sequence of each winch;

[0008] an identification module for performing dynamic characteristic identification and compensation processing on each winch control sequence to obtain compensation controller parameters;

[0009] A control module, configured to input the compensation controller parameters into a neural network controller to perform multi-point synchronization control strategy calculations to obtain control instructions for position synchronization, speed synchronization, and torque synchronization;

[0010] A leveling module is used to monitor the drilling platform posture in real time and perform adaptive leveling control based on the control instructions to obtain a stable posture of the platform;

[0011] The compensation module is used to dynamically compensate and accurately control the drilling process under the stable posture of the platform to obtain drilling and geological profile data.

[0012] In the technical solution provided by the present application, the mechanical characteristic parameters of the winch, coordinate system calibration data and initial values ​​of the drilling platform posture are obtained through multi-sensor data acquisition and system initialization processing, which provides accurate basic data for subsequent control and improves the positioning accuracy to ±10mm, which is significantly better than the ±50mm accuracy level of the traditional manual control method. The drilling position planning and path optimization processing of the three-dimensional digital model of the slope based on the coordinate system calibration data, the obtained drilling platform movement trajectory and each winch control sequence optimize the platform movement path, avoid obstacles, reduce energy consumption, and improve the path planning efficiency by more than 40%. By performing dynamic characteristic identification and compensation processing on the control sequence of each winch, the winch transfer function model and personalized compensation controller parameters are obtained, which effectively overcomes the consistency differences between the winches and solves the synchronization control problem caused by equipment differences in traditional control. The response consistency deviation between winches was reduced from 15% to less than 3%. In particular, the transfer function model and compensation controller parameters were input into the neural network controller to calculate the multi-point synchronous control strategy, and control instructions for position synchronization, speed synchronization and torque synchronization were obtained, realizing precise synchronous control between winches, effectively solving the problem of insufficient synchronization in traditional systems, and reducing the platform shaking amplitude by more than 95%. At the same time, the drilling platform posture was monitored in real time and adaptively leveled based on the control instructions. The obtained stable platform posture provided stable working conditions for drilling operations, and the posture control accuracy reached ±0.1°, which was significantly better than the ±0.5° accuracy of traditional manual leveling. The drilling process under the stable platform posture was dynamically compensated and precisely controlled. The high-quality drilling and geological profile data obtained improved the drilling quality and construction efficiency, with drilling accuracy increased by 30% and drilling efficiency increased by more than 40%. When applying artificial intelligence algorithms in specific functions and application fields, the present invention fully considers the contribution of algorithm features to the solution: the neural network controller realizes high-precision control of nonlinear, time-varying, multi-variable complex winch systems through a multi-layer perceptron structure; the fuzzy inference algorithm effectively handles the uncertainty and ambiguity in geological condition identification, enabling the system to intelligently judge the formation characteristics according to the drilling parameters and automatically adjust the parameters; the Kalman filter algorithm solves the problem of multi-sensor data fusion and improves the accuracy of attitude measurement; and the improved A The path planning method that combines the algorithm with the artificial potential field method overcomes the limitations of traditional planning algorithms and realizes the optimal path generation in complex slope environments. The application of these artificial intelligence algorithms enables the present invention to have adaptability, robustness and intelligent decision-making capabilities in complex and changing slope environments, significantly improving the safety, accuracy and efficiency of slope anchor construction and reducing construction risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0014] Figure 1 This is a schematic diagram of an embodiment of the intelligent collaborative management and control system of the high slope anchor frame beam winch platform in the embodiment of this application. DETAILED DESCRIPTION

[0015] An embodiment of the present application provides an intelligent collaborative management and control system for a high-slope anchor frame beam winch platform. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or that are inherent to these processes, methods, products or devices.

[0016] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 In the embodiment of the present application, an embodiment of the intelligent collaborative management and control system of the high slope anchor frame beam winch platform includes:

[0017] Processing module 101 is used to collect and initialize the multi-sensor data of the slope anchor drilling platform to obtain the mechanical characteristic parameters of the winch, coordinate system calibration data and the initial value of the drilling platform posture;

[0018] An optimization module 102 is configured to perform drilling position planning and path optimization processing on the three-dimensional digital model of the slope based on the coordinate system calibration data, thereby obtaining a movement trajectory of the drilling platform and a control sequence for each winch;

[0019] Identification module 103, used to perform dynamic characteristic identification and compensation processing on each winch control sequence to obtain compensation controller parameters;

[0020] The control module 104 is used to input the compensation controller parameters into the neural network controller to calculate the multi-point synchronization control strategy and obtain control instructions for position synchronization, speed synchronization and torque synchronization;

[0021] A leveling module 105 is configured to perform real-time monitoring and adaptive leveling control of the drilling platform posture based on the control instructions to obtain a stable posture of the platform;

[0022] The compensation module 106 is used to dynamically compensate and precisely control the drilling process in the stable posture of the platform to obtain drilling and geological profile data.

[0023] Specifically, processing module 101 first collects multi-sensor data from the slope anchor drilling platform using load sensors, angle sensors, and speed sensors installed on each winch, as well as a six-degree-of-freedom inertial measurement unit (IMU), a laser position measurement system, and a visual recognition system on the drilling platform. The load sensor monitors changes in wire rope tension with an accuracy of ±0.1 kN; the angle sensor measures the rotation angle of the winch drum with a resolution of 0.01°; and the speed sensor monitors the drum speed with a measurement range of 0-120 rpm. The six-degree-of-freedom IMU integrates a three-axis accelerometer and a three-axis gyroscope to measure the platform's pitch, roll, and yaw angles with an accuracy better than 0.01°. During system initialization, the control unit executes a self-test routine to check the hardware status, build a three-dimensional digital model of the slope, perform coordinate system calibration, and measure the winch's mechanical characteristics, including rated power, maximum lifting force, and speed characteristics. Ultimately, the winch's mechanical characteristics, coordinate system calibration data, and the initial drilling platform attitude are obtained.

[0024] Optimization module 102 performs drilling location planning and path optimization on the three-dimensional digital slope model based on the coordinate system calibration data. The digital slope model is first gridded, with a grid size of 0.5m×0.5m in general areas and refined to 0.1m×0.1m in critical areas. Then, based on the slope's geological conditions, stability analysis results, and anchor design parameters, the optimal anchor distribution scheme is calculated, including anchor density, depth, inclination, and azimuth. Accurate three-dimensional coordinates and drilling direction vectors are generated. Next, the drilling platform's movement path is optimized, a spatial obstacle model of the slope is constructed, and the optimal path is calculated using an improved A algorithm combined with an artificial potential field method. The A algorithm evaluates path points using the heuristic function f(n)=g(n)+h(n), where g(n) represents the actual cost from the starting point to the current point, and h(n) represents the estimated cost from the current point to the target point. The system discretizes the calculated path into a series of path points, with 0.5m spacing between adjacent path points. Ultimately, the drilling platform's movement trajectory and the control sequence for each winch are obtained.

[0025] The identification module 103 performs dynamic characteristic identification and compensation processing on the control sequence of each winch. The system measures the dynamic response of the winch drive system under different load conditions by applying standard excitation signals, including step signals, sinusoidal sweep signals and random signals, and records the speed response, position response, current change and temperature change data. Based on these data, the least squares method is combined with the recursive neural network algorithm to identify the transfer function model of each winch. The model parameters include time constant, damping ratio, gain coefficient and dead zone width. In view of the differences in dynamic characteristics between winches, a personalized compensation controller is designed, using a feedforward compensation strategy, feedback compensator, friction compensation algorithm and tooth gap compensation algorithm to finally obtain the compensation controller parameters.

[0026] The control module 104 inputs the compensation controller parameters into the neural network controller to calculate the multi-point synchronous control strategy. The controller adopts a master-slave hierarchical control structure, including a coordination controller, a synchronization controller and a single-machine controller. The neural network model adopts a multi-layer perceptron structure, contains 3 hidden layers, and the number of neurons is 64, 128 and 64 respectively. The activation function adopts the ReLU function. The neural network input includes the target trajectory parameters, the current state of each winch, the platform posture parameters and the environmental interference parameters, and the output is the control instructions for each winch. The system implements a multi-target synchronous control strategy, while considering the three levels of position synchronization, speed synchronization and torque synchronization, and designs an anti-disturbance mechanism to deal with external interference and model uncertainty, and finally obtains the control instructions for position synchronization, speed synchronization and torque synchronization.

[0027] The leveling module 105 monitors the drilling platform's posture in real time and performs adaptive leveling control based on control instructions. The system establishes a mathematical model of the platform's posture, uses Euler angle representation to describe the platform's pitch, roll, and yaw angles, and establishes a mapping relationship between the tension of each winch and the platform's posture. Attitude measurement uses a redundant sensor fusion strategy and applies a Kalman filter algorithm for data fusion, with attitude measurement accuracy better than 0.05°. When the deviation exceeds the preset threshold, the system starts the leveling control program, uses an adaptive control strategy to calculate the tension adjustment of each winch, designs a smooth trajectory to ensure a smooth and impact-free leveling process, and executes an interactive leveling strategy until the posture deviation meets the accuracy requirements, ultimately achieving a stable platform posture.

[0028] Compensation module 106 dynamically compensates and precisely controls the drilling process while the platform maintains a stable posture. The system first performs drill rig positioning control, using visual guidance technology to assist with positioning. After setting drilling parameters, the drill rig is started. During the drilling process, the system implements comprehensive dynamic monitoring and uses multi-sensor data fusion technology to establish a real-time state assessment model, enabling active vibration suppression and control, reducing vibration amplitude by over 85%. The system also implements dynamic platform posture compensation and geologically adaptive drilling control during the drilling process. By monitoring drilling parameter trends, it identifies changes in formation characteristics and automatically adjusts drilling parameters, ultimately producing high-quality drilling and geological profile data.

[0029] In a specific embodiment, the processing module 101 is configured to:

[0030] Collect data from the load sensors, angle sensors, and speed sensors installed on each winch, and the six-degree-of-freedom inertial measurement unit, laser position measurement system, and visual recognition system installed on the drilling platform to obtain real-time status data of the drilling platform;

[0031] Input the real-time status data into the self-test program to detect the working status of sensors, actuators and communication modules, and obtain a system hardware operating status report;

[0032] Import the slope point cloud data obtained through 3D laser scanning or drone photogrammetry to build a 3D digital model of the slope and obtain the slope model data including the planned anchor drilling locations;

[0033] The local coordinate system of the drilling platform and the global coordinate system of the slope are input into the calibration algorithm for precise alignment calculation to obtain the mechanical characteristic parameters of the winch, the coordinate system calibration data and the initial value of the drilling platform posture.

[0034] Specifically, processing module 101 first acquires real-time status data from the slope anchor drilling platform through a multi-sensor data acquisition system. The load sensors on each winch utilize a strain gauge to convert changes in wire rope tension into electrical signals, with a monitoring accuracy of ±0.1 kN and a sampling frequency of 200 Hz. The angle sensor utilizes photoelectric encoder technology with a resolution of 0.01° and a sampling frequency of 500 Hz to record the rotation angle of the winch drum. The speed sensor utilizes the Hall effect principle, with a measurement range of 0-120 rpm and an accuracy of ±0.01 rpm to monitor the winch drum speed. The drilling platform's six-degree-of-freedom inertial measurement unit integrates a three-axis accelerometer and a three-axis gyroscope, with a sampling frequency of 1000Hz and measurement accuracy better than 0.01°. It records the platform's three-axis acceleration and angular velocity in real time and converts them into attitude data expressed as Euler angles using a quaternion algorithm. The laser position measurement system consists of multiple laser rangefinders with a measurement accuracy of ±1mm, using triangulation to determine the platform's spatial position. The visual recognition system, comprised of multiple 4K high-definition industrial cameras with a frame rate of 60fps, assists in positioning through feature point extraction and matching algorithms. These sensor data are transmitted to the central processing unit via fieldbus technology, generating structured real-time status data containing parameters such as position, attitude, velocity, and force.

[0035] The processing module 101 inputs the real-time status data into the self-test program and performs system initialization detection. The self-test program first detects whether the data of each sensor is within the normal range: the load sensor data is between 0-50kN, the angle sensor data changes continuously without jumps, and the speed sensor reading and the angle sensor derivative value have a consistency error of less than 1%. Then the actuator responsiveness is detected: a standard test signal is sent to each winch, the response time and movement accuracy are recorded, and the working status of the actuator is determined. Then the transmission stability of the communication module is verified: the packet loss rate, transmission delay and jitter are detected to ensure the communication quality. The self-test program summarizes the test results into a system hardware operation status report, which includes the status mark of each module, abnormal diagnosis and reliability rating, providing a basis for subsequent operations.

[0036] Processing module 101 imports three-dimensional point cloud data of the slope and builds a digital model. Point cloud data is acquired using a 3D laser scanner or drone photogrammetry, with a point density of 500-1000 points per square meter. Data preprocessing begins during the import process: outlier removal, noise reduction filtering, and point cloud registration. A triangulated meshing algorithm then converts the discrete point cloud into a continuous surface model, with mesh density adaptively adjusted based on surface complexity. Feature extraction then occurs, identifying key features such as slope gradient, aspect, steps, and concave and convex areas. Finally, the planned anchor drilling locations are marked on the model, creating a complete three-dimensional digital model of the slope that incorporates both topographic and engineering information. Processing module 101 performs coordinate system calibration and initialization. First, a global slope coordinate system and a local drilling platform coordinate system are established: the global coordinate system origin is located at the center of the slope base, with the X-axis running along the slope, the Y-axis perpendicular to the slope surface, and the Z-axis pointing vertically upward. The local coordinate system origin is located at the geometric center of the drilling platform, with the coordinate axes corresponding to the platform structure. Precise alignment is then achieved through a calibration algorithm. At least four feature points are selected and their positions are simultaneously calibrated in the global and local coordinate systems. A least-squares optimization problem is then constructed to solve the coordinate transformation matrix, including the rotation matrix and translation vector. Simultaneously, the mechanical characteristic parameters of the winches are measured: each winch's rated power, maximum lifting force, no-load speed, full-load speed, and other parameters are determined to establish a parameter database. Ultimately, coordinate system calibration data, winch mechanical characteristic parameters, and initial drilling platform posture values ​​are generated, laying the foundation for subsequent operations.

[0037] In a specific embodiment, the optimization module 102 is configured to:

[0038] Performing meshing processing on the three-dimensional digital model of the slope, refining the mesh size of the key area to 0.1m×0.1m, and obtaining slope mesh data of different densities;

[0039] Calculating the slope geological conditions, stability analysis results, and anchor design parameters based on the slope grid data to obtain an optimal anchor distribution plan including anchor arrangement density, depth, inclination, and azimuth;

[0040] The optimal anchor distribution scheme is input into the path optimization algorithm, combined with the improved A The algorithm and artificial potential field method are used to calculate the movement path of the drilling platform to obtain a sequence of path points containing the platform position coordinates and attitude angles;

[0041] A multi-objective optimization calculation is performed on the path point sequence to obtain the movement trajectory of the drilling platform and the control sequence of each winch.

[0042] Specifically, the optimization module 102 first performs adaptive meshing on the three-dimensional digital model of the slope. This process uses a quadtree or octree meshing algorithm to dynamically adjust the mesh density according to the complexity of the slope surface. Specifically, the entire slope area is first divided into a coarse grid with an initial grid size of 0.5m×0.5m, and then the areas with large slope curvature, steep slope, and complex structure are meshed, and the grid size of the key area is gradually refined to 0.1m×0.1m. The mesh subdivision rule is based on three key indicators: surface curvature value, elevation change gradient, and geological complexity score. When the surface curvature exceeds the set threshold of 0.5m -1 Grid subdivision is triggered when the elevation gradient is greater than 30° / m or the geological complexity score is higher than 7 (out of 10). After meshing is completed, a multi-resolution slope mesh data structure is formed, containing node coordinates, connection relationships, surface normal vectors, and slope attribute information, providing a geometric foundation for subsequent calculations.

[0043] Optimization module 102 calculates the optimal anchor distribution plan based on the slope grid data. This process first integrates multiple sources of geological data, including lithology distribution, structural surface occurrence, weathering degree, and groundwater distribution, to generate a comprehensive geological model. Slope stability analysis is then performed, using the limit equilibrium method and finite element method to calculate the slope safety factor and potential sliding surface distribution. Based on the analysis results, areas requiring anchor reinforcement are determined, with a focus on sections with a safety factor less than 1.5 and areas near potential sliding surfaces. Anchor design parameter optimization considers support effectiveness and construction economics, using a genetic algorithm to solve a multi-objective optimization problem. Decision variables include anchor density (1-3 anchors per square meter), depth (3-12 meters), inclination (5°-20°), and azimuth (relative to the principal stress direction). The optimization goal is to balance maximizing reinforcement effectiveness with minimizing engineering effort. Constraints include ensuring the slope safety factor is no less than the design requirement, ensuring that the anchor length crosses the potential sliding surface, and ensuring that spacing meets design specifications. The algorithm iterates until convergence and outputs the optimal anchor distribution plan, including the precise three-dimensional coordinates and drilling parameters of each anchor.

[0044] The optimization module 102 inputs the optimal anchor distribution plan into the path optimization algorithm to calculate the moving path of the drilling platform. The path planning adopts a hybrid strategy that combines the improved A algorithm with the artificial potential field method. The improved A algorithm is used for global path planning, in which the heuristic function comprehensively considers three factors: distance cost, elevation change cost and safety cost, in the form of f(n)=g(n)+h(n), where g(n) represents the actual cost from the starting point to the current node n, and h(n) represents the estimated cost from the current node to the target point. In a slope environment, h(n) not only considers the Euclidean distance, but also combines elevation changes and difficulty of passage for weighted calculation. The artificial potential field method is used for local path optimization, setting the target point as an attractive potential field and the obstacle as a repulsive potential field, and guiding the platform movement through the synthetic potential field. The combination of the two algorithms avoids the problem of simple A. The algorithm has the disadvantages of high computational complexity and the artificial potential field method's tendency to fall into local optimality. After the path calculation is completed, the continuous path is discretized into a series of path points, with a spacing of 0.5m between adjacent path points. Each path point contains the platform's position coordinates and attitude angle, forming a path point sequence.

[0045] The optimization module 102 performs multi-objective optimization on the pathpoint sequence to generate the drilling platform trajectory and control sequences for each hoist. This multi-objective optimization considers three core objectives: maximizing platform positioning accuracy, maximizing attitude stability, and minimizing energy consumption. The optimization process first maps the platform kinematic model to the motion relationships of each hoist, calculating the wire rope length and tension distribution for each pathpoint. A weighted sum method is then used to construct a comprehensive objective function, and the optimal control sequence is solved through nonlinear programming. The optimization also considers multiple constraints: maximum hoist lifting force, safe wire rope tension range, platform stability, and motion smoothness. For each hoist, a complete motion curve is generated, including detailed parameters for the acceleration, constant speed, and deceleration phases. Acceleration is controlled within 0.2 m / s², and the maximum operating speed does not exceed 0.5 m / s, ensuring smooth and controllable platform movement. The drilling platform trajectory and control sequences for each hoist are ultimately output, providing the command basis for the execution control module.

[0046] In a specific embodiment, the identification module 103 is configured to:

[0047] Standard excitation signals such as step signals, sine sweep signals, and random signals were applied to each winch, and dynamic response data under different load conditions were recorded to obtain the speed response, position response, current change, and temperature change data of each winch;

[0048] Inputting the dynamic response data into the least square method combined with the recursive neural network algorithm to perform parameter identification calculation to obtain a transfer function model including a time constant, a damping ratio, a gain coefficient and a dead zone width;

[0049] Calculating the difference of the dynamic characteristics of each winch in the transfer function model to obtain quantitative indicators of response speed difference, steady-state gain difference and nonlinear characteristic difference;

[0050] Based on the quantitative indicators, a feedforward compensation strategy and a feedback compensator are designed, and the compensation controller parameters of each winch are obtained by combining the friction compensation algorithm and the backlash compensation algorithm.

[0051] Specifically, the identification module 103 uses system identification technology to identify and compensate for the dynamic characteristics of each winch drive system. First, a standard excitation signal is applied to each winch for dynamic testing. Three typical signals are included: a step signal, used to measure system response speed and overshoot, with an amplitude set to 10% of the rated speed; a sine frequency sweep signal, used to obtain the system's frequency response characteristics, with a frequency range of 0.01Hz to 10Hz and a sweep time of 180 seconds; and a random signal, used for full-band excitation, with a bandwidth of 0-5Hz. Testing is conducted under five load conditions: no load, 25% rated load, 50% rated load, 75% rated load, and rated load. Precision sensors record each winch's speed response (speed change curve), position response (angle accumulation), current change (motor drive current), and temperature change (motor and drive temperature). Data acquisition is performed at a sampling frequency of 1kHz to ensure that system transient characteristics are captured.

[0052] Identification module 103 inputs the dynamic response data into a parameter identification algorithm for processing. Parameter identification utilizes a hybrid approach combining the least squares method with a recursive neural network algorithm. The least squares method is used for linear identification, determining model parameters by minimizing the mean square error between the actual output and the model's predicted output. For each winch, a second-order transfer function model is established: G(s) = K / (Ts^2 + 2ζTs + 1), where K is the gain coefficient, T is the time constant, s is the complex frequency variable, ζ is the damping ratio, and G(s) is the transfer function. The recursive neural network algorithm is used to capture nonlinear characteristics, particularly friction, backlash, and motor saturation. The recursive neural network contains one hidden layer with 20 neurons, uses a tanh activation function, and is trained via backpropagation with a learning rate of 0.01 and 500 training iterations. After the identification calculations are complete, a complete transfer function model is obtained, including the time constant (reflecting the response speed), the damping ratio (reflecting the oscillation characteristics), the gain coefficient (reflecting the input-output proportional relationship), and the deadband width (reflecting the effects of backlash and static friction).

[0053] The identification module 103 performs a variance analysis on the identified transfer function model. First, key dynamic characteristic parameters are extracted: the time constant reflects the system response speed, the damping ratio reflects overshoot and oscillation characteristics, the gain coefficient reflects the static amplification factor, and the dead zone width reflects the nonlinear characteristics. The dynamic characteristic differences between the winches are then calculated. Quantitative indicators include: response speed difference (maximum time constant difference rate), steady-state gain difference (maximum gain coefficient difference rate), and nonlinear characteristic differences (dead zone width difference and friction characteristic difference). The variance calculation uses the relative difference rate method, using the best-performing winch as a benchmark to calculate the percentage deviation of the corresponding parameters of the other winches. These quantitative indicators intuitively reflect the degree of consistency between the winches and provide a quantitative basis for the design of the compensation controller.

[0054] The identification module 103 designs a personalized compensation controller based on the quantitative indicators of characteristic differences. The compensation strategy includes four key components: the feedforward compensation strategy utilizes the identified transfer function model and calculates the control input required to achieve the desired output through inverse calculation of the model; the feedback compensator adopts a PID control structure, with PID parameters optimized individually according to the characteristics of each winch. The proportional gain Kp ranges from 5 to 15, the integral gain Ki ranges from 0.1 to 1, and the differential gain Kd ranges from 0.5 to 3; the friction compensation algorithm includes static friction compensation and dynamic friction compensation. Static friction compensation uses a pulse breakthrough strategy, applying a short, high-amplitude pulse when the control signal falls below the static friction threshold, while dynamic friction compensation uses a speed-dependent nonlinear function; the backlash compensation algorithm reduces the effect of backlash through a bidirectional approach strategy, that is, when reversing direction, it first overdrives and then adjusts to the target position. These compensation mechanisms work together to generate personalized compensation controller parameters for each winch, solving the technical problem of insufficient synchronization between winches in traditional multi-point winch systems.

[0055] In a specific embodiment, the control module 104 is configured to:

[0056] Construct a master-slave hierarchical control structure for the neural network controller, design a coordination controller, a synchronization controller and a single-machine controller, and obtain a multi-level control system architecture;

[0057] The compensation controller parameters, target trajectory parameters, current status of each winch, platform attitude parameters and environmental interference parameters are input into the multi-layer perceptron neural network to obtain the initial control instructions of each winch;

[0058] The initial control instruction is processed by a multi-objective synchronization control strategy, taking into account the three levels of position synchronization, speed synchronization and torque synchronization, to obtain a comprehensive control instruction that balances each synchronization objective;

[0059] The comprehensive control instruction is input into the disturbance observer and the sliding mode controller for anti-disturbance processing, and the control instruction is obtained by combining the model adaptive mechanism and the predictive control strategy.

[0060] Specifically, the control module 104 first constructs a master-slave hierarchical control structure for the neural network controller to form a multi-level control system architecture. The master-slave hierarchical control structure includes three layers of controllers: the top-level coordination controller is responsible for generating the global motion trajectory and coordinating the actions of each winch, handling the global target decomposition and task allocation; the middle-level synchronization controller ensures the coordination between the winches and handles the synchronous control between the winches; the bottom-level single-machine controller performs precise control for each winch and handles the closed-loop control of the single machine. This hierarchical structure effectively solves the control complexity and decomposes the complex multi-point synchronous control problem into several sub-problems. The transmission of the control signal adopts time-synchronized bus technology with a bus cycle of 1ms and jitter controlled within 10μs to ensure strict synchronous transmission of the control signal.

[0061] The control module 104 inputs the transfer function model, compensation controller parameters, target trajectory parameters, the current state of each hoist, platform attitude parameters, and environmental disturbance parameters into a multilayer perceptron neural network to generate initial control commands for each hoist. The multilayer perceptron neural network model utilizes three hidden layers, with 64, 128, and 64 neurons in each layer, respectively. The activation function uses the Reinforced Luminance (ReLU) function, and the output layer uses a linear activation function. The input layer receives various parameter data: transfer function model parameters include dynamic characteristic parameters such as the time constant, damping ratio, and gain coefficient of each hoist; compensation controller parameters include PID parameters, feedforward compensation parameters, and nonlinear compensation parameters; target trajectory parameters include target position, velocity, and acceleration; the current state of each hoist includes its current position, velocity, acceleration, and torque; platform attitude parameters include pitch, roll, and yaw angles, and their rates of change; and environmental disturbance parameters include wind speed and temperature. The neural network is optimized through a combination of offline training and online learning. Offline training utilizes 10,000 sets of data from different operating conditions generated based on a physical model, while online learning continuously collects data and updates network parameters during the actual control process. The output layer of the neural network generates the initial control instructions for each winch, including speed instructions, position instructions and torque instructions.

[0062] Control module 104 processes the initial control instructions using a multi-objective synchronization control strategy, simultaneously considering position synchronization, velocity synchronization, and torque synchronization, and generating comprehensive control instructions that balance each synchronization objective. Position synchronization control ensures that the displacement of each winch changes strictly according to a predetermined proportional relationship. The position synchronization error is defined as the difference between the actual position and the theoretical position, and the control objective is to control the position synchronization error within a range of ±5mm. Velocity synchronization control ensures that the speed of each winch remains consistent or changes according to a predetermined proportional relationship. The velocity synchronization error is defined as the difference between the actual speed and the theoretical speed, and the control objective is to control the velocity synchronization error within a range of ±0.01m / s. Torque synchronization control ensures balanced wire rope tension distribution. The torque synchronization error is defined as the difference between the actual torque and the theoretical torque, and the control objective is to control the torque synchronization error within a range of ±0.5kN. The multi-objective synchronization control strategy utilizes adaptive weighting technology to dynamically adjust the weights of the three synchronization objectives based on different operating conditions. For example, during the platform positioning phase, the emphasis is on ensuring position synchronization (weight 0.6), while during the platform hovering phase, the emphasis is on ensuring torque synchronization (weight 0.7). Multi-objective optimization constructs a comprehensive objective function through the weighted sum method, solves the optimal control instructions, and outputs comprehensive control instructions that balance all synchronous objectives.

[0063] Control module 104 inputs the integrated control instructions into the anti-disturbance control system for processing. The system generates final control instructions by combining a model-adaptive mechanism and a predictive control strategy. The anti-disturbance control system first uses a disturbance observer to estimate the external disturbance torque in real time. The observer bandwidth is set to 20 Hz. The disturbance torque is calculated based on the deviation between the hoist's motor current and output torque, and the corresponding compensation control variable is generated. A sliding mode controller provides robust control capabilities. The sliding surface is designed as a weighted sum of position error and velocity error, and an exponential reaching law is used to reduce the effects of chattering. The model-adaptive mechanism uses a recursive least squares method to update system model parameters online, adapting to load changes and mechanical property drift, and calculating parameter corrections. The predictive control strategy uses the system model to predict system behavior within the next two seconds. A rolling optimization calculation is performed every 100 ms to optimize the control sequence to minimize future state errors. The anti-disturbance control system integrates these technologies to generate final control instructions with high robustness and adaptability, addressing the lack of stability and accuracy of traditional hoist control systems in complex environments.

[0064] In a specific embodiment, the leveling module 105 is used to:

[0065] The Euler angle representation method in the platform attitude mathematical model is established, and the mapping relationship between the pulling force of each winch and the platform attitude is calculated to obtain the basic model of platform attitude control;

[0066] The data from the inertial measurement unit, visual recognition system, and laser ranging system are input into the Kalman filter algorithm for sensor data fusion processing to obtain high-precision attitude measurement results;

[0067] Calculating the deviation and rate of change between the posture measurement result and the target posture, and obtaining a leveling control instruction when the deviation exceeds a preset threshold;

[0068] The leveling control command is input into the adaptive control strategy for processing, and the pseudo-inverse method is used to solve the overdetermined equations, S-shaped velocity curve design and interactive leveling strategy to obtain the stable posture of the platform.

[0069] Specifically, the leveling module 105 establishes a mathematical model of the platform attitude and uses the Euler angle representation to describe the spatial attitude of the drilling platform. The Euler angle representation uses three angles - pitch, roll and yaw to describe the orientation of a rigid body in three-dimensional space. The pitch angle describes the rotation of the platform around the transverse axis, the roll angle describes the rotation of the platform around the longitudinal axis, and the yaw angle describes the rotation of the platform around the vertical axis. The leveling module establishes a mapping relationship between the tension of each winch and the attitude of the platform by analyzing the geometric structure of the platform and the spatial arrangement of each winch. The specific method is to establish a group of force and torque balance equations to describe how the torque generated by the tension of each winch on the platform affects the attitude of the platform. For a four-point supported drilling platform, the established group of equations contains four winch tensions as input variables and three Euler angles as output variables. The sensitivity of the input to the output is described by the Jacobian matrix, forming a tension-attitude mapping matrix, and finally obtaining the basic model of platform attitude control.

[0070] The leveling module 105 feeds the multi-sensor data into a Kalman filter algorithm for fusion processing. The Kalman filter is a recursive optimal estimation algorithm particularly well-suited for processing dynamic measurement data containing random noise. The leveling module collects data from three types of sensors: a six-degree-of-freedom inertial measurement unit provides angular velocity and acceleration data, which has a high sampling frequency but is subject to integration drift; a visual recognition system provides attitude information by identifying markers on the platform, which is highly accurate but may be affected by lighting and viewing angle; and a laser ranging system provides precise distance measurement, calculating the platform's position relative to a reference point. The Kalman filter process consists of two phases: prediction and update. The prediction phase estimates the current state based on the system state equation and the previous state; the update phase corrects the prediction based on the measured values. The filter state vector contains the platform's three Euler angles and their angular velocity. The system noise covariance matrix and the measurement noise covariance matrix are adaptively adjusted based on real-time data. The Kalman filter sampling period is set to 10ms, ensuring that the fused attitude measurement accuracy is better than 0.05°, resulting in a highly accurate attitude measurement result. The leveling module 105 calculates the deviation between the attitude measurement result and the target attitude. The leveling module first obtains the target attitude parameters. Ideally, the drilling platform should remain horizontal or aligned with the designed drilling direction. It then calculates the deviation between the actual and target attitudes, including pitch, roll, and yaw deviations. The rate of change of these deviations is also calculated, reflecting the dynamic trend of the platform's attitude. This deviation calculation utilizes a small-angle approximation method, directly performing algebraic differences on the angle values. The leveling module compares the calculated deviations with preset thresholds, which are dynamically adjusted based on the different operation phases. During the drilling preparation phase, the pitch and roll thresholds are set to 0.5°, and the yaw threshold is set to 1.0°. During the drilling process, each angle threshold is reduced by 50% to enhance stability. When any deviation exceeds the corresponding threshold, the leveling control program is triggered, generating a leveling control instruction containing the direction and magnitude of the adjustment.

[0071] The leveling module 105 inputs the leveling control instructions into the adaptive control strategy for processing. The adaptive control strategy first uses the pseudo-inverse method to solve the overdetermined system of equations and calculate the tension adjustment amount of each winch required to achieve the desired posture. Since the four winches control three posture angles, an overdetermined system of equations is formed. The pseudo-inverse method provides the optimal solution in the least squares sense, that is, minimizing the sum of the squares of the winch adjustments while meeting the posture adjustment requirements. After the tension adjustment amount is calculated, it is converted into the winch displacement adjustment amount based on the winch characteristics and wire rope stiffness. Next, an S-shaped velocity curve is designed to make the leveling process smooth and shock-free. The S-shaped curve gradually changes the acceleration to avoid mechanical shock caused by sudden changes. Finally, an interactive leveling strategy is adopted to decompose the overall leveling process into multiple small steps. The adjustment amount of each small step is controlled within a small range: the displacement adjustment amount is less than 50mm, and the posture adjustment amount is less than 0.2°. After each small step is completed, the platform attitude is remeasured, the remaining deviation is calculated, and the next adjustment amount is planned until the attitude deviation meets the accuracy requirements: the pitch angle is less than 0.1°, the roll angle is less than 0.1°, and the yaw angle is less than 0.2°, and the platform finally obtains a stable attitude.

[0072] In a specific embodiment, the compensation module 106 is configured to:

[0073] Positioning control is performed on the drilling rig, driving the drilling rig to the target drilling position, and using visual guidance technology to assist the drill head in aligning with the predetermined drilling direction to obtain the initial state of the drilling rig with precise positioning;

[0074] Perform all-round dynamic monitoring of drilling rig parameters, platform posture, and environmental parameters during the drilling process, and input the monitoring data into multi-sensor data fusion technology to obtain real-time status assessment results of the drilling process;

[0075] Active vibration suppression control is performed on the vibration and impact force generated during the drilling process based on the real-time state evaluation result, and the tension of each winch is adjusted to generate a compensation force to obtain a platform state after vibration reduction;

[0076] Monitor and analyze the changing trends of drilling parameters, identify changes in formation characteristics through fuzzy inference algorithms, automatically adjust drilling parameters, and obtain drilling and geological profile data.

[0077] Specifically, the compensation module 106 first performs precise positioning control on the drilling rig. The positioning process is divided into two stages: coarse positioning and fine positioning. The coarse positioning stage moves the drilling rig roughly to the target drilling location based on the optimal anchor distribution plan provided by the optimization module. The fine positioning stage uses visual guidance technology for precise alignment. Visual guidance technology uses a high-definition industrial camera mounted on a platform to capture the position and direction of the drill bit, and performs real-time image processing based on feature points pre-marked on the slope surface. The image processing process includes feature extraction, target recognition, and position calculation. Feature extraction uses the SIFT algorithm to extract stable feature points on the slope surface. Target recognition identifies the drill bit position through template matching. Position calculation calculates relative coordinates based on the principle of triangulation. Positioning control uses an iterative approach strategy. After each adjustment of the drilling rig position, visual inspection is performed again to calculate the deviation between the current position and the target position. Adjustments are continued until the deviation is less than the set threshold (position deviation <10mm, direction deviation <0.2°), thus achieving the initial state of the precisely positioned drilling rig.

[0078] Compensation module 106 provides comprehensive dynamic monitoring of the drilling process. The monitoring system includes three types of parameters: drilling rig parameter monitoring, including drilling speed (m / min), bit pressure (kN), torque (N·m), impact pressure (MPa), and flushing pressure (MPa), collected by dedicated sensors installed on the drilling rig; platform attitude monitoring, including pitch angle, roll angle, yaw angle, and their rate of change, collected by a six-degree-of-freedom inertial measurement unit; and environmental parameter monitoring, including wind speed (m / s), vibration (g), and temperature (°C), collected by corresponding sensors. This multi-source heterogeneous data is integrated and processed using multi-sensor data fusion technology, employing the DS evidence theory framework. DS evidence theory is particularly well-suited for processing uncertain information. It calculates the basic probability distribution of each sensor data and then uses the Dempster synthesis rule for information fusion. The specific steps include data preprocessing (normalization and denoising), basic probability distribution calculation, evidence fusion, and decision reasoning. The system updates the status assessment results every 100ms, generating a real-time status assessment report that includes drilling status rating (normal / abnormal), platform stability rating (stable / unstable), and risk rating (low / medium / high).

[0079] The compensation module 106 performs active vibration suppression control based on the real-time status assessment results. The vibration during the drilling process mainly comes from the impact between the drill bit and the rock. The vibration is transmitted to the platform through the drill pipe, affecting the drilling accuracy and equipment life. The vibration suppression control first identifies the vibration characteristics through frequency domain analysis, and uses fast Fourier transform (FFT) to convert the time domain vibration signal into a frequency domain representation to identify the main vibration frequency components and their amplitudes. Different suppression strategies are adopted for vibrations of different frequencies: low-frequency vibrations (<5Hz) are mainly suppressed by active winch control. The control algorithm calculates the winch tension adjustment required to generate a compensation force opposite to the vibration phase; medium and high frequency vibrations (5-50Hz) are mainly absorbed by the platform damping structure. The compensation force calculation is based on the vibration amplitude and frequency characteristics. The tension of each winch is adjusted in real time through closed-loop feedback control to generate accurate compensation force, offset the impact of vibration, and keep the platform stable.

[0080] Compensation module 106 monitors and analyzes drilling parameter trends to achieve geo-adaptive drilling control. The core of geo-adaptive drilling control is to identify formation characteristics through drilling parameter changes and optimize drilling parameters accordingly. Monitoring and analysis first establishes a mapping between parameter changes and geological characteristics: a decrease in drilling rate and an increase in torque typically indicates encountering hard rock formations; a sudden increase in drilling rate and torque fluctuations typically indicate encountering fractured zones; and an increase in flushing fluid pressure and a decrease in return water typically indicate encountering fractures. Fuzzy inference algorithms are used to identify geological characteristics and are suitable for handling complex systems with fuzzy boundaries and nonlinear relationships. The fuzzy inference process involves four steps: fuzzification converts precise input values ​​into fuzzy sets; a rule base stores expert knowledge in the form of "IF drilling rate decreases AND torque increases THEN the rock formation is hard"; inference calculations determine the activated rules and their activation levels based on the inputs; and defuzzification converts the fuzzy outputs into precise control values. Based on the identification results, the system automatically adjusts drilling parameters: increasing impact energy and reducing weight on bit when encountering hard rock formations; and reducing drilling rate and increasing flushing flow when encountering fractured zones. At the same time, the system records the stratum change information during the drilling process and generates geological profile data containing information such as rock hardness, integrity and degree of fragmentation.

[0081] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. An intelligent collaborative management and control system for a high-slope anchor frame beam winch platform, characterized in that: The system includes: The processing module is used to collect and initialize the multi-sensor data of the slope anchor drilling platform to obtain the mechanical characteristic parameters of the winch, coordinate system calibration data and the initial value of the drilling platform posture; An optimization module is used to perform drilling position planning and path optimization processing on the three-dimensional digital model of the slope based on the coordinate system calibration data, so as to obtain the movement trajectory of the drilling platform and the control sequence of each winch; an identification module for performing dynamic characteristic identification and compensation processing on each winch control sequence to obtain compensation controller parameters; A control module, configured to input the compensation controller parameters into a neural network controller to perform multi-point synchronization control strategy calculations to obtain control instructions for position synchronization, speed synchronization, and torque synchronization; A leveling module is used to monitor the drilling platform posture in real time and perform adaptive leveling control based on the control instructions to obtain a stable posture of the platform; The compensation module is used to dynamically compensate and precisely control the drilling process in the stable posture of the platform to obtain drilling and geological profile data; the compensation module is specifically used to: The drilling rig is positioned and controlled, driven to the target drilling location. Vision guidance technology is used to assist the drill head in aligning with the predetermined drilling direction. Positioning control adopts an iterative approach strategy. After each adjustment of the drilling rig position, visual inspection is performed again to calculate the deviation between the current position and the target position. Adjustments are continued until the deviation is less than the set threshold, forming the initial state of the drilling rig with precise positioning. Perform all-round dynamic monitoring of drilling rig parameters, platform posture, and environmental parameters during the drilling process, and input the monitoring data into multi-sensor data fusion technology to obtain real-time status assessment results of the drilling process; Based on the real-time status assessment results, active vibration suppression control is performed on the vibration and impact force generated during the drilling process, and the tension of each winch is adjusted to generate a compensation force. Different suppression strategies are adopted for vibrations of different frequencies: low-frequency vibrations less than 5 Hz are mainly suppressed by active winch control, and the control algorithm calculates the winch tension adjustment required to generate a compensation force opposite to the vibration phase; medium and high-frequency vibrations in the range of 5 Hz to 50 Hz are mainly absorbed by the platform damping structure. The compensation force calculation is based on the vibration amplitude and frequency characteristics. The tension of each winch is adjusted in real time through closed-loop feedback control to generate accurate compensation force, offset the impact of vibration, and keep the platform stable. The changing trends of drilling parameters are monitored and analyzed, and changes in formation characteristics are identified through fuzzy inference algorithms. The drilling parameters are automatically adjusted to obtain borehole and geological profile data. Geologically adaptive drilling control identifies formation characteristics through changes in drilling parameters. Based on the identification results, the system automatically adjusts drilling parameters: when encountering hard rock layers, the impact energy is increased and the drilling pressure is reduced; when encountering broken zones, the drilling speed is reduced and the flushing flow rate is increased.

2. The intelligent collaborative management and control system for the high-slope anchor frame beam winch platform according to claim 1 is characterized in that: The processing module is used to: Collect data from the load sensors, angle sensors, and speed sensors installed on each winch, and the six-degree-of-freedom inertial measurement unit, laser position measurement system, and visual recognition system installed on the drilling platform to obtain real-time status data of the drilling platform; Input the real-time status data into the self-test program to detect the working status of sensors, actuators and communication modules, and obtain a system hardware operating status report; Import the slope point cloud data obtained through 3D laser scanning or drone photogrammetry to build a 3D digital model of the slope and obtain the slope model data including the planned anchor drilling locations; The local coordinate system of the drilling platform and the global coordinate system of the slope are input into the calibration algorithm for precise alignment calculation to obtain the mechanical characteristic parameters of the winch, the coordinate system calibration data and the initial value of the drilling platform posture.

3. The intelligent collaborative management and control system for the high-slope anchor frame beam winch platform according to claim 1 is characterized in that: The optimization module is used to: Performing meshing processing on the three-dimensional digital model of the slope, refining the mesh size of the key area to 0.1m×0.1m, and obtaining slope mesh data of different densities; Calculating the slope geological conditions, stability analysis results, and anchor design parameters based on the slope grid data to obtain an optimal anchor distribution plan including anchor arrangement density, depth, inclination, and azimuth; The optimal anchor distribution scheme is input into the path optimization algorithm, combined with the improved A The algorithm and artificial potential field method are used to calculate the movement path of the drilling platform to obtain a sequence of path points containing the platform position coordinates and attitude angles; A multi-objective optimization calculation is performed on the path point sequence to obtain the movement trajectory of the drilling platform and the control sequence of each winch.

4. The intelligent collaborative management and control system for the high-slope anchor frame beam winch platform according to claim 1 is characterized in that: The identification module is used to: Apply standard excitation signals to each winch, including step signals, sine sweep signals, and random signals, and record dynamic response data under different load conditions to obtain the speed response, position response, current change, and temperature change data of each winch; Inputting the dynamic response data into the least square method combined with the recursive neural network algorithm to perform parameter identification calculation to obtain a transfer function model including a time constant, a damping ratio, a gain coefficient and a dead zone width; Calculating the difference of the dynamic characteristics of each winch in the transfer function model to obtain quantitative indicators of response speed difference, steady-state gain difference and nonlinear characteristic difference; Based on the quantitative indicators, a feedforward compensation strategy and a feedback compensator are designed, and the compensation controller parameters of each winch are obtained by combining the friction compensation algorithm and the backlash compensation algorithm.

5. The intelligent collaborative management and control system for the high-slope anchor frame beam winch platform according to claim 1 is characterized in that: The control module is used to: Construct a master-slave hierarchical control structure for the neural network controller, design a coordination controller, a synchronization controller and a single-machine controller, and obtain a multi-level control system architecture; The compensation controller parameters, target trajectory parameters, current status of each winch, platform attitude parameters and environmental interference parameters are input into the multi-layer perceptron neural network to obtain the initial control instructions of each winch; The initial control instruction is processed by a multi-objective synchronization control strategy, taking into account the three levels of position synchronization, speed synchronization and torque synchronization, to obtain a comprehensive control instruction that balances each synchronization objective; The comprehensive control instruction is input into the disturbance observer and the sliding mode controller for anti-disturbance processing, and the control instruction is obtained by combining the model adaptive mechanism and the predictive control strategy.

6. The intelligent collaborative management and control system for the high-slope anchor frame beam winch platform according to claim 1 is characterized in that: The leveling module is used to: The Euler angle representation method in the platform attitude mathematical model is established, and the mapping relationship between the pulling force of each winch and the platform attitude is calculated to obtain the basic model of platform attitude control; The data from the inertial measurement unit, visual recognition system, and laser ranging system are input into the Kalman filter algorithm for sensor data fusion processing to obtain high-precision attitude measurement results; Calculating the deviation and rate of change between the posture measurement result and the target posture, and obtaining a leveling control instruction when the deviation exceeds a preset threshold; The leveling control command is input into the adaptive control strategy for processing, and the pseudo-inverse method is used to solve the overdetermined equations, S-shaped velocity curve design and interactive leveling strategy to obtain the stable posture of the platform.

Images