Control system and method for land ladle overturning machine

Abstract

The invention relates to the field of engineering machinery control, and discloses a self-adaptive control system and method for a land ladle overturning machine. The system comprises a multi-mode sensing unit, a central control unit, a hydraulic execution unit and a man-machine interaction terminal, according to the method, dip angle / pressure / visual data are fused in real time, a ladle overturning track is dynamically corrected based on the material gravity center offset, when the overturning probability exceeds a threshold value, progressive hydraulic pressure reduction is started, and an energy consumption feedback model is constructed to optimize execution parameters. The problems that the overturning risk is high and the energy consumption difference is large due to the fact that a traditional ladle overturning machine lacks perception are solved, the abnormal interruption rate is reduced by 76% or above, and the comprehensive energy efficiency is improved to 82%.

Description

Technical Field

[0001] The present invention relates to the field of engineering machinery control, and more particularly, to a control system and method for a land bag turning machine. Background Art

[0002] In the field of bulk material handling, land-based bale tippers are widely used as key equipment in mines, ports, and other scenarios. They use hydraulically driven bale tipping arms to achieve material dumping operations. With the advancement of intelligent manufacturing, traditional manual operation modes can no longer meet the industrial needs of high safety and low energy consumption. Especially when handling heterogeneous materials such as ore and scrap steel, problems such as sudden changes in the center of gravity and adhesion of materials frequently occur. There is an urgent need to improve operational stability and efficiency through intelligent control. Existing technologies still have certain shortcomings: Limitations of perception Existing bag turning machines rely heavily on inclination sensors and basic weighing devices, which are unable to accurately measure the three-dimensional distribution of materials in real time. For example, when handling loose stacks of material, traditional systems are unable to detect boundary slippage. This results in approximately 23% of bag turning operations resulting in equipment shaking or even tipping due to center of gravity shift, forcing operators to frequently interrupt operations for manual adjustments.

[0003] 2. Rigid control strategy Mainstream hydraulic control systems use fixed pressure thresholds, and even when material distribution is abnormal, they still execute the flipping action according to a preset program. A 2023 industry accident report shows that 38% of bale flipper failures are due to the hydraulic system's failure to dynamically respond to changes in the center of gravity. In particular, the lack of a progressive pressure reduction mechanism after the deviation exceeds the safety threshold can cause overload damage to the hydraulic cylinder or material spillage.

[0004] 3. Lack of energy efficiency optimization Traditional energy consumption control focuses solely on motor power, ignoring the coordinated optimization of hydraulic flow and turnover trajectory. Field data shows that in continuous operation scenarios, due to the lack of an established energy consumption feedback model, the energy consumption of the same model of bag turning machine varies by up to 31% per operation. Furthermore, without a hydraulic energy recovery design, the overall system energy efficiency is less than 65%.

[0005] Therefore, in order to solve the above problems, a control system and method for a land bag turning machine are proposed. Summary of the Invention

[0006] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a control system and method for a land bag turning machine to solve the problems raised in the above-mentioned background technology.

[0007] To achieve the above-mentioned purpose, the present invention provides the following technical solution: a control system for a land bag turning machine, comprising: The multimodal sensing unit monitors the status of the bag turning operation in real time. It consists of a digital tilt sensor installed at the rotating joint of the bag turning arm, a multi-point pressure sensor integrated into the hydraulic branch, an array weighing sensor fixed to the bottom of the material loading platform, and a 3D vision sensor facing the material pile. The central control unit is connected to the multi-modal sensor unit data and has a built-in adaptive decision module to generate hydraulic control instructions by fusing multi-source sensor data; The hydraulic actuator receives instructions from the central control unit and includes an electro-hydraulic proportional valve group and a hydraulic cylinder with displacement feedback to dynamically adjust the lifting trajectory of the bag turning arm; The human-computer interaction terminal is equipped with an industrial-grade touch screen and a physical emergency stop switch, which can display the material distribution heat map, overturning risk level and hydraulic system status parameters in real time.

[0008] Preferably, the adaptive decision module includes a dynamic load analysis submodule and a hydraulic parameter optimization submodule. The dynamic load analysis submodule calculates the three-dimensional coordinates of the material center of gravity and the uniformity of mass distribution in real time based on the pressure distribution data of the weighing sensor and the point cloud data of the 3D vision sensor. The overturning risk prediction submodule inputs the material center of gravity coordinates, real-time flip angle and equipment structure parameters into a pre-trained machine learning model, and outputs a dynamic overturning probability value. The hydraulic parameter optimization submodule adjusts the working pressure of the hydraulic system in stages according to the overturning probability value, and starts a progressive pressure control strategy when the probability value exceeds a preset safety threshold.

[0009] Preferably, the progressive pressure control strategy is executed according to the following logic: first-level response, when the overturning probability value is in the range of 0.25 to 0.4, the target working pressure of the hydraulic cylinder is reduced to 90%-95% of the rated pressure; second-level response, when the overturning probability value is in the range of 0.4 to 0.6, the hydraulic pressure is reduced in steps every time the bag turning arm rotates 10 degrees, and the single pressure reduction amplitude is 3%-5% of the current pressure; third-level response, when the overturning probability value exceeds 0.6, the bag turning action is immediately suspended and the vibration compensation mechanism is activated to stabilize the material shape through high-frequency micro-vibration.

[0010] Preferably, the 3D vision sensor realizes material status analysis through the following process: a depth camera based on the time-of-flight principle is used to continuously collect three-dimensional point cloud data of the bag turning area, the material pile and the equipment background point cloud are separated by a spatial clustering algorithm, the material boundary slip characteristics are identified based on the point cloud normal vector change rate, and the boundary stability index and pile height uniformity parameters are output.

[0011] A control method for a land bag turning machine comprises the following steps: Step S1: Initialize the operating parameters and automatically match the preset flip speed curve according to the material type database; Step S2: Real-time monitoring of the center of gravity offset of the material. When the lateral offset exceeds 15% of the effective width of the load-bearing platform, online correction of the trajectory of the bag turning arm is triggered. Step S3: Analyze the hydraulic pressure fluctuation spectrum during the lifting process. When high-frequency fluctuations above 2 Hz are detected, it is determined to be material adhesion and a decoupling vibration signal is injected; Step S4: Compare the deviation between the actual energy consumption and the theoretical energy consumption model, and optimize the hydraulic valve control parameters of the subsequent operation cycle based on the deviation amplitude.

[0012] Preferably, the online correction of the motion trajectory of step S2 is achieved by first establishing a multi-rigid body dynamics model of the bag flipping arm, taking the center of gravity offset vector as the input parameter, using an iterative optimization algorithm to calculate the real-time compensation angle of each joint of the robotic arm, converting the compensation angle value into a proportional valve opening instruction, and dynamically adjusting the hydraulic cylinder displacement through a PID controller.

[0013] Preferably, the decoupling vibration signal generation in step S3 includes: generating a variable frequency sinusoidal wave pressure instruction with an amplitude not exceeding 8% of the rated pressure of the hydraulic system, setting the vibration frequency and the material's natural frequency to maintain a difference of more than 3 Hz to avoid resonance, and automatically determining the adhesion release state and stopping vibration based on the pressure fluctuation spectrum characteristics.

[0014] Preferably, the theoretical energy consumption model of step S4 is composed of the following elements: the integral term of the product of the real-time pressure and flow of the hydraulic system over time, the square term of the maximum flip angle of a single operation, and when the actual energy consumption continues to exceed the theoretical value by 15%, the opening rate of the hydraulic valve in the next working cycle is automatically reduced and the hydraulic energy recovery circuit is enabled.

[0015] Technical effects and advantages of the present invention: Full-dimensional perception and decision-making capabilities Through multimodal sensor fusion (inclination / pressure / weighing / 3D vision), a three-dimensional material distribution model is constructed in real time, accurately identifying boundary slip and center of gravity shift. During an actual iron ore bag flipping test, the system provided a 1.8-second advance warning of slip risks, reducing the interruption rate by 76% and significantly reducing the frequency of manual intervention.

[0016] Adaptive security protection mechanism Based on a graded response strategy for rollover probability, the hydraulic system pressure threshold is dynamically adjusted. When the center of gravity offset reaches a critical value, a progressive pressure reduction mechanism suppresses the peak hydraulic cylinder pressure to below 85% of the rated value, reducing the hydraulic system failure rate by 52% and completely preventing material spillage accidents.

[0017] Dynamic trajectory collaborative optimization Utilizing online trajectory correction technology, the optimal compensation angle is calculated in real time when the center of gravity deflects beyond a certain limit. Testing on scrap steel showed that after compensation, the vibration amplitude at the end of the bale-turning arm was reduced to ±2.3cm (compared to ±8.5cm in traditional systems), improving equipment stability by 64%.

[0018] Energy efficiency closed-loop optimization system A theoretical-to-actual energy consumption feedback model was established, combining adaptive hydraulic valve opening rate regulation with an energy recovery loop. Data from continuous port operations showed that the variance in energy consumption per cycle was reduced from 31% to less than 9%, improving overall energy efficiency to 82%, and saving over 150,000 kWh of electricity annually (based on a 300-day operation). BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a system framework diagram of the present invention. DETAILED DESCRIPTION

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0021] As attached Figure 1 As shown, (1) a control system for a land bag turning machine, comprising: The multimodal sensing unit monitors the status of the bag turning operation in real time. It consists of a digital tilt sensor installed at the rotating joint of the bag turning arm, a multi-point pressure sensor integrated into the hydraulic branch, an array weighing sensor fixed to the bottom of the material loading platform, and a 3D vision sensor facing the material pile. The central control unit is connected to the multi-modal sensor unit data and has a built-in adaptive decision module to generate hydraulic control instructions by fusing multi-source sensor data; The hydraulic actuator receives instructions from the central control unit and includes an electro-hydraulic proportional valve group and a hydraulic cylinder with displacement feedback to dynamically adjust the lifting trajectory of the bag turning arm; The human-machine interface (HMI) terminal, equipped with an industrial-grade touchscreen and a physical emergency stop switch, displays real-time material distribution heat maps, tipping risk levels, and hydraulic system status parameters. Specifically, the system hardware deployment includes: high-precision digital tilt sensors installed on the flange surfaces of the two rotating joints of the bale turning arm to monitor joint angle changes in real time, with data uploaded via the CAN bus at a 100ms cycle. High-pressure dynamic sensors are embedded in the rodless cavity piping of the hydraulic oil circuit to continuously monitor oil pressure fluctuations. A 4×4 array of piezoelectric weighing cells is embedded in the steel plate at the bottom of the material loading platform, with each cell independently measuring local pressure. An obliquely mounted 3D depth camera uses time-of-flight ranging to scan the material pile at 30 frames per second to generate a 3D point cloud. The central processor synchronously collects all sensor data via industrial Ethernet and runs a real-time operating system to process decision instructions. An electro-hydraulic proportional valve group receives control signals to drive the hydraulic cylinder, and an integrated displacement feedback sensor within the cylinder body forms a closed-loop regulation mechanism. The HMI interface, equipped with an industrial-grade touchscreen, dynamically displays a simulated heat map of the material distribution and indicates the risk level with red, yellow, and green warning lights.

[0022] (2) The adaptive decision module includes a dynamic load analysis submodule and a hydraulic parameter optimization submodule. The dynamic load analysis submodule calculates the three-dimensional coordinates of the material center of gravity and the uniformity of mass distribution in real time based on the pressure distribution data of the weighing sensor and the point cloud data of the 3D vision sensor. The overturning risk prediction submodule inputs the material center of gravity coordinates, real-time flip angle and equipment structure parameters into a pre-trained machine learning model and outputs a dynamic overturning probability value. The hydraulic parameter optimization submodule adjusts the hydraulic system working pressure in stages according to the overturning probability value, and starts a progressive pressure control strategy when the probability value exceeds a preset safety threshold. The dynamic load analysis submodule execution process is as follows: first, read the 16-point pressure values ​​of the weighing sensor array, and calculate the material center of gravity position based on the weighted average coordinates of the installation position of each sensor; at the same time, process the 3D camera point cloud data, separate adjacent material points through a spatial clustering algorithm, and statistically analyze the elevation data to obtain the stockpile uniformity index. The capsizing risk prediction submodule feeds real-time center of gravity coordinates, current bale turning angle, stockpile uniformity, and historical accident records into a pretrained machine learning model. The model, comprised of 20 decision trees, operates in parallel and ultimately outputs a capsizing probability value between 0 and 1. The hydraulic optimization submodule linearly reduces the hydraulic system's target pressure based on this probability, decreasing the set pressure by 3% for every 0.1 increase in probability.

[0023] (3) The progressive pressure control strategy is executed according to the following logic: first-level response, when the overturning probability value is in the range of 0.25 to 0.4, the target working pressure of the hydraulic cylinder is reduced to 90%-95% of the rated pressure; second-level response, when the overturning probability value is in the range of 0.4 to 0.6, the hydraulic pressure is reduced step by step every time the bag turning arm rotates 10 degrees, and the single pressure reduction amplitude is 3%-5% of the current pressure; third-level response, when the overturning probability value exceeds 0.6, the bag turning action is immediately suspended and the vibration compensation mechanism is activated to stabilize the material shape through high-frequency micro-vibration. Operation process: When the overturning probability value is in the range of 25%-40%, the control proportional valve will reduce the opening to 92% within 0.5 seconds, and the system working pressure will be adjusted to 93% of the rated pressure simultaneously; when the probability rises to 40%-60%, the step-by-step pressure reduction is triggered every time the bale turning arm rotates 10 degrees, and the pressure is reduced by 4% of the current value each time; when the probability exceeds 60%, the main oil circuit is immediately cut off, and a specific frequency vibration wave is injected into the hydraulic cylinder at the same time. The vibration amplitude is controlled within 6% of the rated pressure, and the frequency is set to the natural frequency of the material plus or minus 3.5 Hz to stagger the resonance point, and the vibration continues until the pressure fluctuation tends to be stable.

[0024] (4) The 3D vision sensor realizes material status analysis through the following process: a depth camera based on the time-of-flight principle is used to continuously collect three-dimensional point cloud data of the bag turning area, a spatial clustering algorithm is used to separate the material pile and the equipment background point cloud, and the material boundary slip characteristics are identified based on the point cloud normal vector change rate, and the boundary stability index and pile height uniformity parameters are output. Among them, the 3D vision sensor workflow is as follows: the depth camera collects 30 frames of point cloud data per second, pre-loads the equipment three-dimensional model to generate a background template, and automatically filters out interference points with a height of more than 1.2 meters; the material point cloud calculates the normal vector direction change rate of the boundary area, and when it is detected that the normal vector angle changes by more than 15 radians within a unit distance, it is determined that there is a slip risk; finally, the boundary stability index is output, which is inversely proportional to the normal vector change rate. The greater the change rate, the lower the index.

[0025] (5) A control method for a land bag turning machine, comprising the following steps: Step S1: Initialize the operating parameters and automatically match the preset flip speed curve according to the material type database; Step S2: Real-time monitoring of the center of gravity offset of the material. When the lateral offset exceeds 15% of the effective width of the load-bearing platform, online correction of the trajectory of the bag turning arm is triggered. Step S3: Analyze the hydraulic pressure fluctuation spectrum during the lifting process. When high-frequency fluctuations above 2 Hz are detected, it is determined to be material adhesion and a decoupling vibration signal is injected; Step S4: Compare the deviation values ​​of the actual energy consumption and the theoretical energy consumption model, and optimize the hydraulic valve control parameters of the subsequent operation cycle based on the deviation amplitude, wherein the control method execution steps: in the initialization stage, call the preset parameter library according to the material type code, for example, iron ore corresponds to the three-stage flipping speed curve of slow, fast and slow; continuously monitor the lateral offset rate of the center of gravity during operation, and activate the trajectory correction program when it exceeds 15% of the effective width of the platform; analyze the spectral characteristics of the hydraulic pressure fluctuation in real time, and determine that the material is sticking when the energy proportion of the 2-5 Hz frequency band exceeds 40% of the total energy, and trigger decoupling vibration; each time a single bag flipping action is completed, compare the deviation value of the actual energy consumption with the theoretical model, and optimize the subsequent operation parameters if the deviation continues to exceed 15%.

[0026] (6) The online correction of the motion trajectory of step S2 is achieved by the following method: first, a multi-rigid body dynamics model of the bag turning arm is established, the center of gravity offset vector is used as an input parameter, and an iterative optimization algorithm is used to calculate the real-time compensation angle of each joint of the robot arm. The compensation angle value is converted into a proportional valve opening instruction, and the hydraulic cylinder displacement is dynamically adjusted by a PID controller. The online correction of the trajectory is achieved by: establishing a mechanical balance equation of the bag turning arm, using the center of gravity offset as an input variable; using an iterative optimization algorithm to calculate the angle value that needs to be compensated for each joint of the robot arm, so that the line of action of the material gravity and the direction of the hydraulic support force tend to coincide; converting the calculated angle compensation amount into the target displacement of the hydraulic cylinder, and dynamically adjusting the proportional valve opening through a proportional-integral controller. The entire closed-loop control response time is controlled within 35 milliseconds.

[0027] (7) The decoupling vibration signal generation in step S3 includes: generating a variable frequency sinusoidal pressure instruction with an amplitude not exceeding 8% of the rated pressure of the hydraulic system, setting the vibration frequency to maintain a difference of more than 3 Hz with the natural frequency of the material to avoid resonance, automatically determining the adhesion release state and stopping vibration based on the pressure fluctuation spectrum characteristics, wherein the decoupling vibration signal generation: setting the upper limit of the vibration wave amplitude to 8% of the rated pressure of the system, and setting the basic frequency to the natural frequency ±3.5 Hz range according to the material characteristics; the signal generator outputs a variable frequency sinusoidal wave with a frequency fluctuation of 0.3 Hz per second to prevent resonance with the material; and automatically stopping the output when it is detected that the pressure fluctuation amplitude drops by 40% or the vibration continues for 5 seconds.

[0028] (8) The theoretical energy consumption model of step S4 is composed of the following elements: the integral term of the product of the real-time pressure and flow of the hydraulic system over time, the square term of the maximum flip angle of a single operation, and when the actual energy consumption continues to exceed 15% of the theoretical value, the opening rate of the hydraulic valve in the next working cycle is automatically reduced and the hydraulic energy recovery circuit is enabled, wherein the energy consumption closed-loop optimization is implemented: the theoretical energy consumption model includes a fixed part (proportional to the square of the maximum flip angle) and a dynamic part (the cumulative product of pressure and flow); when the actual energy consumption continues to exceed 15% of the theoretical value, the opening speed of the hydraulic valve in the next working cycle is reduced to 80% of the original rate, and the energy recovery device is enabled at the same time to convert hydraulic energy into electrical energy storage during the descent of the bag flipping arm. Example

[0029] Step 1: System initialization and parameter loading 1. Hardware self-test: After power is turned on, the central control unit (MCU) sequentially detects the zero drift of the tilt sensor, the range of the pressure sensor, and the focus status of the 3D vision camera. If an abnormality occurs, an alarm code will be triggered on the human-machine interface. 2. Material database matching: The operator enters the material code (such as iron ore F001), and the MCU calls the pre-stored parameters: Rated flip angular velocity curve (Example: 3° / s in the 0°-45° range, 1.5° / s in the 45°-90° range) Material natural frequency range (iron ore: 1.2-1.8Hz) Safe center of gravity offset threshold (15% of platform width) Step 2: Real-time data fusion and status analysis 1.Synchronous collection of multi-source data: The tilt sensor uploads the joint angle (α1, α2) every 100ms Array weighing sensor output pressure distribution matrix [P mn ] 3D vision sensor generates point cloud {Q(x,y,z)} 2. Dynamic load calculation: Based on the pressure matrix [P mn ]Calculate the coordinates of the center of gravity X_c = Σ(P mn ·x mn ) / ΣP mn Perform spatial clustering on the point cloud {Q}, remove outliers and calculate the standard deviation of the height σ_h (reflecting uniformity) 3. Capsizing probability prediction: Input vector = [X_c, α1, σ_h, historical overturning records] Random forest model processing flow: Feature normalization: Map the input vector to the [0,1] interval Decision tree voting: 20 pre-trained trees compute capsize risk labels in parallel Probability output: Calculate the proportion of high-risk labels as the probability value P_risk Step 3: Hydraulic Adaptive Control Execution 1. Trajectory correction decision: When |X_c| > 0.15W (W is the table width), start online correction: Construct a virtual moment balance equation: Material weight moment = hydraulic support moment + inertia compensation term The joint compensation angle Δα is iteratively solved by the golden section method to minimize the angle between the gravity vector and the support force vector 2. Progressive pressure regulation (associated with P_risk value): If 0.25 ≤ P_risk < 0.4: Target pressure = current pressure × 0.93 The hydraulic valve opening is lowered to K_p1 (preset opening level) If 0.4 ≤ P_risk < 0.6: Every 10° flip triggers a pressure drop: New pressure = previous pressure × (1 - 0.04 × P_risk) If P_risk ≥ 0.6: Immediately pause the bag turning arm Inject vibration signal: amplitude = 8% rated pressure, frequency = f_material ± 3.5Hz Step 4: Energy consumption closed-loop optimization 1. Calculation of energy consumption for a single operation: Actual energy consumption E_act = pressure sensor reading × flow meter reading × cumulative sum of time steps Theoretical energy consumption E_ref = basic energy consumption coefficient × square of maximum flip angle + flow integral term 2. Parameter adaptive adjustment: When E_act > 1.15E_ref: The hydraulic valve opening rate in the next cycle is reduced to 80% of the original value Enable regeneration circuit: convert the hydraulic cylinder downward energy into electrical energy storage Step 5: Job termination and data archiving When the bag turning angle reaches 90°, the hydraulic system will execute the slow-down procedure (speed ≤ 0.5° / s) Store key parameters in the database: actual center of gravity trajectory, P_risk peak, energy consumption deviation rate Updating random forest model weights based on historical data Finally, it should be noted that: first, in the description of the present application, it should be pointed out that unless otherwise specified and limited, the terms "installation", "connection", "connection" should be broadly understood, which can be mechanical connection or electrical connection, or the communication between two elements, or direct connection, "up", "down", "left", "right" and the like are only used to indicate the relative positional relationship, when the absolute position of the described object changes, the relative positional relationship may change; Secondly: the drawings of the disclosed embodiments of the present application only involve the structures involved in the disclosed embodiments, other structures can refer to the usual design, and in the case of no conflict, the same embodiment and different embodiments of the present application can be combined with each other; Finally: the above only describes the preferred embodiments of the present application and is not used to limit the present application, any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A control system for a land bag turning machine, characterized in that: include: The multimodal sensing unit monitors the status of the bag turning operation in real time. It consists of a digital tilt sensor installed at the rotating joint of the bag turning arm, a multi-point pressure sensor integrated into the hydraulic branch, an array weighing sensor fixed to the bottom of the material loading platform, and a 3D vision sensor facing the material pile. The central control unit is connected to the multi-modal sensor unit data and has a built-in adaptive decision module to generate hydraulic control instructions by fusing multi-source sensor data; The hydraulic actuator receives instructions from the central control unit and includes an electro-hydraulic proportional valve group and a hydraulic cylinder with displacement feedback to dynamically adjust the lifting trajectory of the bag turning arm; The human-computer interaction terminal is equipped with an industrial-grade touch screen and a physical emergency stop switch, which can display the material distribution heat map, overturning risk level and hydraulic system status parameters in real time.

2. The control system for a land bag turning machine according to claim 1, characterized in that: The adaptive decision-making module includes a dynamic load analysis submodule and a hydraulic parameter optimization submodule. The dynamic load analysis submodule calculates the three-dimensional coordinates of the material center of gravity and the uniformity of mass distribution in real time based on the pressure distribution data of the weighing sensor and the point cloud data of the 3D vision sensor. The overturning risk prediction submodule inputs the material center of gravity coordinates, real-time flip angle and equipment structural parameters into a pre-trained machine learning model, and outputs a dynamic overturning probability value. The hydraulic parameter optimization submodule adjusts the working pressure of the hydraulic system in stages according to the overturning probability value, and activates a progressive pressure control strategy when the probability value exceeds a preset safety threshold.

3. The control system for a land bag turning machine according to claim 2, characterized in that: The progressive pressure control strategy is executed according to the following logic: first-level response, when the overturning probability value is in the range of 0.25 to 0.4, the target working pressure of the hydraulic cylinder is reduced to 90%-95% of the rated pressure; second-level response, when the overturning probability value is in the range of 0.4 to 0.6, the hydraulic pressure is reduced in steps every time the bale turning arm rotates 10 degrees, and the single pressure reduction amplitude is 3%-5% of the current pressure; third-level response, when the overturning probability value exceeds 0.6, the bale turning action is immediately suspended and the vibration compensation mechanism is activated to stabilize the material shape through high-frequency micro-vibration.

4. The control system for a land bag turning machine according to claim 1, characterized in that: The 3D vision sensor implements material status analysis through the following process: a time-of-flight depth camera is used to continuously collect three-dimensional point cloud data of the bag turning area, a spatial clustering algorithm is used to separate the material pile and the equipment background point cloud, and the material boundary slip characteristics are identified based on the rate of change of the point cloud normal vector. The boundary stability index and pile height uniformity parameters are output.

5. A control method for a land bag turning machine, applied to any system of claims 1-4, characterized in that: The following steps are involved: Step S1: Initialize the operating parameters and automatically match the preset flip speed curve according to the material type database; Step S2: Real-time monitoring of the center of gravity offset of the material. When the lateral offset exceeds 15% of the effective width of the load-bearing platform, online correction of the trajectory of the bag turning arm is triggered. Step S3: Analyze the hydraulic pressure fluctuation spectrum during the lifting process. When high-frequency fluctuations above 2 Hz are detected, it is determined to be material adhesion and a decoupling vibration signal is injected; Step S4: Compare the deviation between the actual energy consumption and the theoretical energy consumption model, and optimize the hydraulic valve control parameters of the subsequent operation cycle based on the deviation amplitude.

6. The control method for a land bag turning machine according to claim 5, characterized in that: The online correction of the motion trajectory in step S2 is achieved by first establishing a multi-rigid-body dynamic model of the bag-turning arm, taking the center of gravity offset vector as the input parameter, using an iterative optimization algorithm to calculate the real-time compensation angle of each joint of the robotic arm, converting the compensation angle value into a proportional valve opening instruction, and dynamically adjusting the hydraulic cylinder displacement through a PID controller.

7. The control method for a land bag turning machine according to claim 5, characterized in that: The decoupling vibration signal generation in step S3 includes: generating a variable frequency sinusoidal pressure command with an amplitude not exceeding 8% of the rated pressure of the hydraulic system, setting the vibration frequency to maintain a difference of more than 3 Hz from the natural frequency of the material to avoid resonance, and automatically determining the adhesion release state and stopping vibration based on the pressure fluctuation spectrum characteristics.

8. The control method for a land bag turning machine according to claim 5, characterized in that: The theoretical energy consumption model of step S4 is composed of the following elements: the integral term of the product of the real-time pressure and flow of the hydraulic system over time, and the square term of the maximum flip angle of a single operation. When the actual energy consumption continues to exceed the theoretical value by 15%, the opening rate of the hydraulic valve in the next working cycle is automatically reduced and the hydraulic energy recovery circuit is activated.

Images