A cage control system and control method based on visual feature extraction and deformation compensation

Abstract

The present application belongs to the technical field of mine hoist, and proposes a cage control system and control method based on visual feature extraction and deformation compensation, which comprises a cage integrated terminal installed on the cage of mine, a wireless relay network arranged along the shaft, and a ground control center arranged on the ground; the cage integrated terminal comprises an explosion-proof box body; the present application sets the integrated terminal on the cage, and configures a weighing sensor, an acceleration sensor, and an image acquisition and processing module, so that the cage has real-time sensing capability for its own load state and running state; meanwhile, the image feature extraction module is combined to dataize the internal situation of the cage, realizing the change from traditional indirect detection relying on the hoist side to active sensing at the cage end; and in the case of abnormal wireless bridge communication, the key state data can still be sent through feature extraction and LORA communication module, thereby ensuring the continuous availability of the internal personnel and material state of the cage.

Description

Technical Field

[0001] This invention belongs to the field of mine hoisting technology, specifically relating to a cage control system and control method based on visual feature extraction and deformation compensation. Background Technology

[0002] Mine hoisting systems are one of the key transportation systems in mine production. Cages, as crucial carriers for transporting personnel, materials, and equipment up and down within the shaft, are widely used in underground mining scenarios such as coal and metal mines. Cages are typically connected to the hoist via steel wire ropes and reciprocate up and down along a guide structure within the shaft, facilitating transportation between the shaft opening and bottom, and between intermediate levels. With increasing mine depths and advancements in automation, higher demands are placed on the safety, positioning accuracy, and information management capabilities of cage operations.

[0003] However, existing cage systems still have significant shortcomings in practical applications: On the one hand, the cages rely heavily on indirect detection from the hoist side during operation, lacking real-time acquisition and utilization of the cage's own operating status (such as load and motion status), resulting in significant impact on positioning accuracy due to the elastic deformation of the wire rope during multi-level docking, easily leading to docking deviations; on the other hand, the shaft environment is complex, and wireless signals are easily affected by metal structures and spatial obstructions, resulting in poor stability of video and data transmission inside the cage, posing a risk of communication interruption, which in turn affects the continuous monitoring of personnel and material status inside the cage; in addition, existing systems rely heavily on single-condition control between the cage and the shaft safety door, lacking a multi-condition interlocking mechanism, which poses certain safety hazards. Summary of the Invention

[0004] This invention provides a cage control system and method based on visual feature extraction and deformation compensation, which solves the problems mentioned in the background art, such as insufficient acquisition of cage operating status information, significant influence of wire rope elastic deformation on docking positioning accuracy, poor stability of wireless communication in the shaft environment, and insufficient safety due to the single interlocking control between the cage and the shaft safety door.

[0005] The technical solution adopted in this invention is: a cage control system based on visual feature extraction and deformation compensation, including a cage integrated terminal installed on a mine cage, a wireless relay network deployed along the shaft, and a ground control center set up on the ground.

[0006] The cage integrated terminal includes an explosion-proof enclosure, and a cage main control board, power supply module, sensor group, camera group, LORA communication module, wireless bridge module and wireless charging receiver installed in the explosion-proof enclosure; wherein, the sensor group includes at least a load cell for detecting the load and an accelerometer for measuring the cage's motion state.

[0007] The wireless relay network is used to establish a wireless communication link between the cage integrated terminal and the ground control center. The wireless communication link includes a broadband channel for transmitting video data and a narrowband channel for transmitting status data.

[0008] The ground control center is electrically connected to the mine hoist electrical control system and the mine entrance safety door system, respectively. It is used to receive status data sent by the cage integrated terminal and output control signals for hoist position compensation and safety door interlock control.

[0009] The wireless charging receiver assembly is installed on the outer wall of the mine cage. It includes two guide rods installed on the outer wall of the cage and arranged vertically. A movable cover plate is slidably installed on the two guide rods. A return spring is sleeved on both ends of the guide rods. A charging receiver panel is installed in the middle of the outer wall of the cage. A push plate is fixed on the outer surface of the movable cover plate. When the cage moves to the wellhead or the bottom of the well, the movable cover plate and the push plate can move downward or upward under the action of external force to fully expose the charging receiver panel. This allows the wireless charging transmitter installed at the wellhead or the bottom of the well to be positioned opposite the charging receiver panel for electromagnetic induction charging. A locking device is provided on the movable cover plate to prevent the movable cover plate from moving up and down when it covers the charging receiver panel.

[0010] The explosion-proof enclosure is also equipped with an image feature extraction module, which is connected to the camera group and the main control board of the cage respectively.

[0011] The cage main control board is used to control the image feature extraction module to extract features from the image data collected by the camera group when the wireless bridge module communication is abnormal, and to send the extraction results to the ground control center through the LORA communication module.

[0012] The ground control center is equipped with a 3D display terminal, which is used to visualize and reconstruct the internal state of the cage based on the received feature data.

[0013] The ground control center is equipped with a deformation compensation module, which is connected to the cage integrated terminal and the mine hoist electrical control system, respectively, and is used to generate position compensation based on the received load data and motion status data.

[0014] The ground control center is used to apply position compensation to the operating control parameters of the hoist, so as to adjust the deceleration position and stopping position of the cage.

[0015] The ground control center is equipped with an interlocking control module, which outputs an opening permission signal for the wellhead safety door. Its output condition requires the simultaneous fulfillment of the following three signals:

[0016] Signal A: The mechanical limit switch installed at the docking position is triggered by the cage;

[0017] Signal B: The cage integrated terminal sends a current operating status of "stopped" via the LORA communication module.

[0018] Signal C: The mechanical brake of the mine hoist electrical control system is locked;

[0019] When all three signals are satisfied, the interlocking control module outputs an opening control signal to the wellhead safety door system.

[0020] This application also provides a cage control method based on visual feature extraction and deformation compensation, which applies the above-mentioned cage control system and includes the following steps:

[0021] Step S1: When the cage is stopped at the wellhead or bottom, the power supply module of the cage is replenished with power through the wireless charging device, and the monitoring and communication module is activated according to the load change inside the cage.

[0022] Step S2: During the operation of the cage, load data and motion status data are collected, and the cage position compensation is calculated based on the data. The position compensation is then applied to the hoist control parameters to achieve leveling control of the cage.

[0023] Step S3: Monitor the wireless communication status during cage operation. When broadband communication is abnormal, initiate image feature extraction and send feature data through narrowband communication to achieve continuous monitoring of the cage's internal status.

[0024] The specific process for calculating the dynamic tensile deformation of the wire rope in step S2 is as follows:

[0025] Based on the total weight measured by the load cell, the static tensile force caused by the static load is calculated.

[0026] Based on the running acceleration measured by the accelerometer, the dynamic additional stretch caused by inertial force is calculated;

[0027] The static stretching amount and the dynamic additional stretching amount are added together to obtain the total elongation error value, which is then sent to the main control PLC of the hoist in the form of a pulse signal.

[0028] The specific operation for extracting the personnel posture coordinates inside the cage in step S3 is as follows:

[0029] At the instant the video broadband channel is disconnected, the image feature extraction module inside the cage captures the image frame from the current camera;

[0030] Image processing algorithms are used to identify the human skeleton in the image and extract the three-dimensional spatial coordinates of the head, torso, and major joints of the limbs relative to the bottom plate of the cage.

[0031] These three-dimensional spatial coordinates are packaged into plain text data frames according to the communication protocol format of the LORA module and sent. After being received by the ground, the coordinates are assigned to a preset 3D virtual character to reproduce the standing or falling state of the person in the cage.

[0032] The beneficial effects of this invention are as follows:

[0033] This invention, by installing an integrated terminal on the cage and configuring it with a weighing sensor, an acceleration sensor, and an image acquisition and processing module, enables the cage to have real-time perception capabilities of its own load and operating status. At the same time, combined with an image feature extraction module, it digitizes the internal conditions of the cage, realizing a transformation from traditional indirect detection relying on the hoist side to active perception at the cage end. Moreover, even in the event of a wireless bridge communication failure, it can still extract key status data and send it through the LORA communication module, thereby ensuring the continuous availability of the status of personnel and materials inside the cage.

[0034] This invention, by setting up a deformation compensation module in the ground control center, combines load data obtained from weighing sensors and motion state data obtained from acceleration sensors to comprehensively calculate the elongation of the wire rope under static load and dynamic inertia, and generates a position compensation amount that is applied to the hoist control parameters. This enables dynamic correction of the cage's deceleration and stopping positions, thereby effectively reducing the impact of wire rope elastic deformation on docking accuracy and improving the leveling accuracy and operational stability of the cage during multi-level docking.

[0035] This invention utilizes a hierarchical communication link composed of a wireless relay network, dividing the communication channel into a broadband video data channel and a narrowband status data channel. Simultaneously, an interlocking control module is set up in the ground control center. Based on the simultaneous fulfillment of three conditions—mechanical limit switch signal, cage running status signal, and hoist brake status signal—a safety door opening command is output. This improves the stability of data transmission in the complex environment of the mine shaft while realizing multi-condition interlocking control to replace the single signal control method, effectively avoiding accidental door opening and improving the overall safety of mine hoisting operations. Attached Figure Description

[0036] Figure 1 This is a perspective view of the cage of the present invention;

[0037] Figure 2 This is a front view of the cage of the present invention;

[0038] Figure 3 This is a partial schematic diagram of the present invention;

[0039] Figure 4 The system flow of this invention Figure 1 ;

[0040] Figure 5 The system flow of this invention Figure 2 .

[0041] in:

[0042] 1. Cage body; 2. Movable cover; 3. Push plate; 4. Guide rod; 5. Charging receiver panel; 6. Return spring; 7. Explosion-proof enclosure; 8. Spring plunger. Detailed Implementation

[0043] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0044] As shown in the figure, a cage control system based on visual feature extraction and deformation compensation consists of three main parts: a cage integrated terminal, a wireless relay network, and a ground control center. The cage integrated terminal is fixedly installed on the top or side wall structure of the mine cage body 1. The wireless relay network is deployed in layers along the shaft height at intervals on the shaft wall or guide rail structure. The ground control center is located in the shaft opening control room and is electrically connected to the mine hoist electrical control system and the shaft opening safety door system via cables to realize centralized processing and control output of the cage's operating status.

[0045] The cage-integrated terminal is encapsulated in an explosion-proof enclosure 7, which is fixed to the cage's metal structure by bolts or welding. Inside, the electronic modules are arranged in a layered configuration. Specifically, the cage's main control board is located in the middle mounting layer, the power supply module is located at the bottom to lower the center of gravity, and the communication and image processing modules are located in the upper area to reduce electromagnetic interference. The modules are connected via cabling or an industrial bus, and the cage's main control board centrally manages data acquisition, processing, and communication scheduling. The power supply module preferably includes a rechargeable battery pack and a power management unit. The power management unit regulates, limits, and distributes the input power to supply power to the sensor group, camera group, and communication module.

[0046] In the sensor group, the load cell is preferably installed between the bottom plate of the cage and the supporting beam to obtain the total load data of the cage by measuring the force change of the supporting structure. The acceleration sensor is fixedly installed on the cage frame structure to collect the acceleration change signal of the cage in the vertical direction. Both types of sensors are connected to the cage main control board through the signal acquisition interface. The data they collect is transmitted to the main control board in real time at a set sampling frequency. The main control board performs filtering, noise reduction and data fusion processing to improve the stability and accuracy of the data.

[0047] The camera group comprises multiple distributed cameras, installed on the top and side walls of the cage to cover the interior space. Under normal circumstances, the video data collected is transmitted in real-time via a wireless bridge module. This wireless bridge module establishes a connection with the shaft relay node using broadband wireless communication, thereby enabling continuous transmission of video data to the ground control center. Simultaneously, an image feature extraction module is installed inside the explosion-proof enclosure 7. This module has a built-in image processing unit. When the communication signal of the wireless bridge module falls below a set threshold or is completely interrupted, the cage's main control board triggers the image feature extraction module to start, analyzing and processing the currently buffered video frames.

[0048] Specifically, in the image feature extraction process, the image is first preprocessed, including grayscale conversion, noise reduction, and target region extraction. Then, the human key point recognition algorithm is used to identify the people in the image and extract the spatial position data of multiple joint points, including the head, shoulders, elbows, and knees. A three-dimensional coordinate model is established with the bottom plate of the cage as the reference coordinate system. Subsequently, the extracted coordinate data is encoded and compressed according to the preset communication protocol to form a small feature data frame, which is sent to the ground control center through the LORA communication module. This allows the posture of the people inside the cage to be reflected even under low bandwidth conditions.

[0049] The wireless relay network includes multiple relay nodes positioned at different heights within the wellbore. Each relay node is fixed to the well wall or guide structure via a bracket. Adjacent relay nodes form a relay communication link. The relay node closest to the wellhead communicates directly with the ground control center. During operation, the cage integrated terminal establishes a wireless connection with the nearest relay node and switches connections between different relay nodes as the cage's position changes to ensure the continuity of the communication link. The communication link is divided into a broadband channel and a narrowband channel. The broadband channel is used for video transmission, while the narrowband channel is used for status data and feature data transmission, thus maintaining critical data communication even under conditions of limited bandwidth or strong interference.

[0050] The wireless charging receiver is located in the middle of the outer wall of the cage. In its structure, two guide rods 4 are vertically parallel. The movable cover 2 is sleeved on the guide rods 4 through a sliding sleeve structure and can slide up and down along the guide rods 4. The return springs 6 set at the upper and lower ends of the guide rods 4 apply elastic force to the movable cover 2, so that it remains in a state of blocking the charging receiver panel 5 when there is no external force. When the cage moves to the preset position at the wellhead or bottom, the fixed pushing mechanism set at the wellhead or bottom contacts the pushing plate 3 on the movable cover 2, thereby forcing the movable cover 2 to move along the guide rods 4, so that the charging receiver panel 5 is exposed and aligned with the wireless charging transmitter to realize electromagnetic induction charging. After charging is completed, it automatically resets under the action of the return spring 6. At the same time, the locking device limits and locks the movable cover 2 to prevent shaking during operation. In addition, a locking device is provided on the movable cover plate 2, which is configured to prevent the movable cover plate 2 from moving up and down when it covers the charging receiving panel 5. The locking device is a spring plunger 8 installed on the movable cover plate 2. A groove is provided at the middle position of the guide rod 4 near the charging receiving panel 5. The steel ball of the spring plunger 8 can be placed in the groove. When the movable cover plate 2 is in the state of covering the charging receiving panel 5, the steel ball in the spring plunger 8 is embedded in the groove under the action of elasticity, thereby forming a positioning limit for the movable cover plate 2, so that the movable cover plate 2 remains stable during the normal operation of the cage and avoids up and down sliding due to vibration or inertia. When the cage moves to the wellhead or bottom position, the movable cover 2 moves along the guide rod 4 under the action of the external pushing mechanism. During the movement, the steel ball of the spring plunger 8 overcomes the elastic force and disengages from the groove under the action of external force, so that the movable cover 2 is released from the limit and continues to slide, thereby realizing the opening of the charging receiving panel 5. After the external force is removed, the movable cover 2 returns to the initial position under the action of the reset spring 6, and the steel ball is re-embedded in the groove, completing the re-locking.

[0051] The ground control center is equipped with a data processing unit, a 3D display terminal, a deformation compensation module, and an interlocking control module. The data processing unit is used to process the received load data, motion data, and feature data in a unified manner. The 3D display terminal establishes a virtual model based on the received human body key point coordinate data. By driving the preset 3D human body model, the dynamic display of the posture of the personnel inside the cage is realized, so that the status of the personnel can still be intuitively reflected even when the video is interrupted.

[0052] The deformation compensation module is used to calculate the elongation of the wire rope. The specific process is as follows: First, the elongation of the wire rope under static stress is calculated based on the total load measured by the weighing sensor. Then, the additional tensile force caused by inertial force is calculated based on the running acceleration measured by the acceleration sensor. Subsequently, the two sets of data are superimposed to obtain the total elongation error, and this error is converted into the corresponding displacement compensation amount. This is sent to the hoist main control PLC through the control interface to adjust the deceleration trigger position and final stopping position of the hoist, so as to achieve precise leveling of the cage.

[0053] Furthermore, the deformation compensation module is used to calculate the elongation of the wire rope under stress and generate the corresponding position compensation amount. The specific calculation process is as follows:

[0054] First, based on the total load of the cage measured by the weighing sensor... Combined with the material elastic modulus of the steel wire rope Effective cross-sectional area and the current length of the stressed section Calculate the static tensile strength of the wire rope under static load. The calculation formula is as follows:

[0055]

[0056] Wherein, the length of the force-bearing segment The location can be dynamically obtained or determined by looking up a table based on the current well depth of the cage.

[0057] Based on this, the cage's running acceleration is measured by the acceleration sensor. Combined with the total mass of the cage and load Calculate the additional tensile force caused by inertial force. And further calculate the corresponding dynamic additional tensile amount. Its expression is:

[0058]

[0059] Subsequently, the static tensile amount and the dynamic additional tensile amount are superimposed to obtain the total elongation error of the wire rope. ,Right now:

[0060]

[0061] Furthermore, the total elongation error The displacement compensation amount is converted into a value that the hoist control system can recognize. Preferably, the compensation amount is converted into the corresponding encoder pulse count or displacement correction value and sent to the hoist main control PLC through the communication interface. This is used to feedforward correction of the hoist's deceleration trigger position and final stopping position, so that the cage can achieve precise leveling under different loads and operating conditions.

[0062] The ground control center is equipped with an interlocking control module used to output an opening permission signal for the wellhead safety door. Its output condition requires the simultaneous fulfillment of the following three signals:

[0063] Signal A: The mechanical limit switch installed at the docking position is triggered by the cage;

[0064] Signal B: The cage integrated terminal sends a current operating status of "stopped" via the LORA communication module.

[0065] Signal C: The mechanical brake of the mine hoist electrical control system is locked;

[0066] When the above three signals are met simultaneously, the interlocking control module outputs an opening control signal to the wellhead safety door system. That is, the opening control signal is only output when the cage reaches the designated stopping position, the running status is stopped, and the brake is locked. Otherwise, the safety door remains closed, thus forming a multi-condition interlocking control mechanism.

[0067] Based on the above system, this embodiment also provides a control method. During operation, when the cage stops at the wellhead or bottom, the power supply module is recharged via a wireless charging component. Simultaneously, the monitoring and communication module is automatically activated based on load changes. Load and motion data are continuously collected during cage operation, and the wire rope elongation error is calculated in real time. The compensation amount is fed back to the hoist control system to achieve dynamic leveling control. At the same time, the wireless communication status is monitored in real time. When a broadband communication anomaly is detected, the system automatically switches to a feature data transmission mode, and continuous monitoring of the cage's internal status is achieved through image feature extraction and narrowband communication.

[0068] A cage control method based on visual feature extraction and deformation compensation, which applies the above-mentioned cage control system, includes the following steps:

[0069] Step S1: When the cage is stopped at the wellhead or bottom, the power supply module of the cage is replenished with power through the wireless charging device, and the monitoring and communication module is activated according to the load change inside the cage.

[0070] Step S2: During the operation of the cage, load data and motion status data are collected, and the cage position compensation is calculated based on the data. The position compensation is then applied to the hoist control parameters to achieve leveling control of the cage.

[0071] Step S3: Monitor the wireless communication status during cage operation. When broadband communication is abnormal, initiate image feature extraction and send feature data through narrowband communication to achieve continuous monitoring of the cage's internal status.

[0072] The specific process for calculating the dynamic tensile deformation of the wire rope in step S2 is as follows:

[0073] Based on the total weight measured by the load cell, the static tensile force caused by the static load is calculated.

[0074] Based on the running acceleration measured by the accelerometer, the dynamic additional stretch caused by inertial force is calculated;

[0075] The static stretching amount and the dynamic additional stretching amount are added together to obtain the total elongation error value, which is then sent to the main control PLC of the hoist in the form of a pulse signal.

[0076] The specific operation for extracting the personnel posture coordinates from inside the cage in step S3 is as follows:

[0077] At the instant the video broadband channel is disconnected, the image feature extraction module inside the cage captures the image frame from the current camera;

[0078] Image processing algorithms are used to identify the human skeleton in the image and extract the three-dimensional spatial coordinates of the head, torso, and major joints of the limbs relative to the bottom plate of the cage.

[0079] These three-dimensional spatial coordinates are packaged into plain text data frames according to the communication protocol format of the LORA module and sent. After being received by the ground, the coordinates are assigned to a preset 3D virtual character to reproduce the standing or falling state of the person in the cage.

[0080] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A cage control system based on visual feature extraction and deformation compensation, characterized in that, This includes cage integration terminals installed on mine cages, wireless relay networks deployed along the shaft, and ground control centers located on the ground. The cage integrated terminal includes an explosion-proof enclosure (7), and a cage main control board, power supply module, sensor group, camera group, LORA communication module, wireless bridge module and wireless charging receiver assembly installed in the explosion-proof enclosure (7); wherein, the sensor group includes at least a weighing sensor for detecting the load and an acceleration sensor for measuring the cage's motion state; The wireless relay network is used to establish a wireless communication link between the cage integrated terminal and the ground control center. The wireless communication link includes a broadband channel for transmitting video data and a narrowband channel for transmitting status data. The ground control center is electrically connected to the mine hoist electrical control system and the mine entrance safety door system, respectively. It is used to receive status data sent by the cage integrated terminal and output control signals for hoist position compensation and safety door interlock control.

2. The cage control system based on visual feature extraction and deformation compensation according to claim 1, characterized in that: The wireless charging receiver assembly is installed on the outer wall of the mine cage. It includes two guide rods (4) installed on the outer wall of the cage and arranged vertically. A movable cover plate (2) is slidably installed on the two guide rods (4). A return spring (6) is sleeved on both ends of the guide rods (4). A charging receiver panel (5) is installed in the middle of the outer wall of the cage. A push plate (3) is fixed on the outer surface of the movable cover plate (2). When the cage moves to the wellhead or the bottom of the well, the movable cover plate (2) and the push plate (3) can move downward or upward under the action of external force so that the charging receiver panel (5) is fully exposed. The wireless charging transmitter installed at the wellhead or the bottom of the well is positioned opposite to the charging receiver panel (5) to perform electromagnetic induction charging. A locking device is provided on the movable cover plate (2) to prevent the movable cover plate (2) from moving up and down when it covers the charging receiver panel (5).

3. The cage control system based on visual feature extraction and deformation compensation according to claim 1, characterized in that: The explosion-proof enclosure (7) is also equipped with an image feature extraction module, which is connected to the camera group and the cage main control board respectively; The cage main control board is used to control the image feature extraction module to extract features from the image data collected by the camera group when the wireless bridge module communication is abnormal, and to send the extraction results to the ground control center through the LORA communication module.

4. A cage control system based on visual feature extraction and deformation compensation according to claim 3, characterized in that: The ground control center is equipped with a 3D display terminal, which is used to visualize and reconstruct the internal state of the cage based on the received feature data.

5. A cage control system based on visual feature extraction and deformation compensation according to claim 1, characterized in that: The ground control center is equipped with a deformation compensation module, which is connected to the cage integrated terminal and the mine hoist electrical control system. It is used to generate position compensation based on the received load data and motion status data.

6. A cage control system based on visual feature extraction and deformation compensation according to claim 5, characterized in that: The ground control center is used to apply position compensation to the operating control parameters of the hoist, so as to adjust the deceleration position and stopping position of the cage.

7. A cage control system based on visual feature extraction and deformation compensation according to claim 1, characterized in that: The ground control center is equipped with an interlocking control module used to output an opening permission signal for the wellhead safety door. Its output condition requires the simultaneous fulfillment of the following three signals: Signal A: The mechanical limit switch installed at the docking position is triggered by the cage; Signal B: The cage integrated terminal sends a current operating status of "stopped" via the LORA communication module. Signal C: The mechanical brake of the mine hoist electrical control system is locked; When all three signals are satisfied, the interlocking control module outputs an opening control signal to the wellhead safety door system.

8. A cage control method based on the cage control system according to any one of claims 1 to 7, characterized in that, Includes the following steps: Step S1: When the cage is stopped at the wellhead or bottom, the power supply module of the cage is replenished with power through the wireless charging device, and the monitoring and communication module is activated according to the load change inside the cage. Step S2: During the operation of the cage, load data and motion status data are collected, and the cage position compensation is calculated based on the data. The position compensation is then applied to the hoist control parameters to achieve leveling control of the cage. Step S3: Monitor the wireless communication status during cage operation. When broadband communication is abnormal, initiate image feature extraction and send feature data through narrowband communication to achieve continuous monitoring of the cage's internal status.

9. The cage control method according to claim 8, characterized in that, The specific process for calculating the dynamic tensile deformation of the wire rope in step S2 is as follows: Based on the total weight measured by the load cell, the static tensile force caused by the static load is calculated. Based on the running acceleration measured by the accelerometer, the dynamic additional stretch caused by inertial force is calculated; The static stretching amount and the dynamic additional stretching amount are added together to obtain the total elongation error value, which is then sent to the main control PLC of the hoist in the form of a pulse signal.

10. The cage control method according to claim 8, characterized in that, The specific operation for extracting the personnel posture coordinates inside the cage in step S3 is as follows: At the instant the video broadband channel is disconnected, the image feature extraction module inside the cage captures the image frame from the current camera; Image processing algorithms are used to identify the human skeleton in the image and extract the three-dimensional spatial coordinates of the head, torso, and major joints of the limbs relative to the bottom plate of the cage. These three-dimensional spatial coordinates are packaged into plain text data frames according to the communication protocol format of the LORA module and sent. After being received by the ground, the coordinates are assigned to a preset 3D virtual character to reproduce the standing or falling state of the person in the cage.

Images