Intelligent coal cleaning method for a car dumper

Abstract

The application provides a kind of intelligent coal cleaning method of dumper, comprising: using human body detection and tracking algorithm based on deep learning, real-time monitoring personnel position and behavior in the working area, if personnel enter dangerous area or unsafe behavior is detected, alarm mechanism is triggered, cantilever stops moving or shifts to safe position;Collect and analyze equipment operating parameters, vibration signal data, build support vector machine-based equipment health status evaluation model, realize early warning and diagnosis of equipment failure, if equipment anomaly is found, generate repair recommendations and solutions;Optimize cantilever motion planning and control strategy, reduce energy consumption and mechanical wear, real-time monitoring equipment state, and dynamically adjust cleaning parameters according to health evaluation results, ensure safe and efficient operation of equipment.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of information technology, and in particular to an intelligent car dumper residual coal cleaning method. BACKGROUND

[0002] During the empty car residual coal cleaning process of the car dumper, a new technical contradiction is introduced by the modularized cantilever design. The cantilever is composed of different functional modules, such as telescopic arms, folding arms, and rotating joints, to adapt to the complex and variable cleaning environment. However, the change in module combination leads to changes in the cantilever's center of gravity, length, and movement mode, requiring dynamic adjustment of the fulcrum structure. The fulcrum needs to provide a matching support torque and flexibility according to different cleaning scenarios to ensure the stability and safety of the cleaning operation.

[0003] This technical contradiction further extends to the spatial limitation problem of the fulcrum structure. The internal space of the car dumper is narrow, and the rapid movement and precise adjustment of the fulcrum face space constraints. How to realize the dynamic adjustment of the fulcrum in limited space while considering cleaning efficiency and stability is a technical problem that needs to be solved.

[0004] In addition, the differences in residual coal properties also pose new requirements for the design of the cantilever structure. Dry coal dust is prone to dust, and wet coal blocks are prone to adhesion. Different properties of residual coal require the use of module combinations and movement modes that are suitable for them. This requires the cantilever structure to dynamically adjust the module combination and movement parameters according to the real-time detection results of the residual coal properties to optimize the cleaning effect.

[0005] In summary, the modularized cantilever design improves the adaptability of cleaning while introducing the technical contradiction of dynamic adjustment of the fulcrum. This contradiction involves the spatial limitation of the fulcrum structure, the differences in residual coal properties, and the balance between cleaning efficiency and stability, which requires in-depth research and innovation in mechanical design, material selection, sensing detection, and control algorithms to promote the breakthrough and development of the empty car residual coal cleaning technology of the car dumper. SUMMARY

[0006] The present application provides an intelligent car dumper residual coal cleaning method, mainly including:

[0007] Obtain the internal space boundary information of the car dumper and the movement trajectory planning result of the cantilever, use the space collision detection algorithm based on octree to judge the feasibility of the fulcrum adjustment scheme, if there is a risk of space interference, further optimize the adjustment scheme until the space constraint condition is met;

[0008] Obtain the residual coal property information, if the residual coal is dry coal dust, select a high-speed pneumatic cleaning tool with a dust collection device, control the cantilever to move at a low speed and uniform speed, if the residual coal is wet coal block, select a cleaning tool with vibration and scraping function, control the cantilever to move at a low speed;

[0009] The residual coal distribution and the current pose information of the cantilever are acquired, the A-star algorithm is used to solve the optimal cleaning path under the obstacle avoidance constraint, the cleaning completion threshold based on image processing is set, it is judged whether the current area reaches the predetermined cleaning standard, if not, the area is returned for repeated cleaning until the requirement is met;

[0010] The human body detection and tracking algorithm based on deep learning is adopted to monitor the position and behavior of personnel in the working area in real time, if the personnel enter the dangerous area or there is unsafe behavior, the alarm mechanism is triggered to control the cantilever to stop moving or shift to a safe position;

[0011] The device running parameters and vibration signal data are collected and analyzed, the device health state evaluation model based on support vector machine is constructed, the early warning and diagnosis of device fault are realized, if the device is found to be abnormal, the maintenance suggestion and scheme are generated;

[0012] The motion planning and control strategy of the cantilever are optimized, the energy consumption and mechanical wear are reduced, the device state is monitored in real time, and the cleaning parameters are dynamically adjusted according to the health evaluation result, so that the device is safely and efficiently operated.

[0013] The technical scheme provided by the embodiment of the present application can include the following beneficial effects:

[0014] The present application discloses an intelligent residual coal cleaning method for a dumper. First, the octree space collision detection algorithm is used to judge the feasibility of the fulcrum adjustment scheme to ensure that the cantilever movement does not interfere with the space. Then, according to the nature of the residual coal, the appropriate cleaning tool and control strategy are selected, and the A-star algorithm is used to plan the optimal cleaning path. During the cleaning process, the present application evaluates the cleaning completion degree through image processing technology, and repeats the cleaning if necessary. At the same time, the deep learning algorithm is used to monitor the safety of personnel in the working area in real time, and the abnormality is found to alarm and take safety measures in time. In addition, the present application also establishes a device health state evaluation model based on support vector machine to realize fault warning and diagnosis. By optimizing the motion control strategy, the energy consumption and wear are reduced to ensure the safe and efficient operation of the device. The present application realizes the intelligence and automation of the residual coal cleaning of the dumper, improves the working efficiency and safety. BRIEF DESCRIPTION OF DRAWINGS

[0015] Fig. 1 The flowchart of the intelligent residual coal cleaning method for the dumper of the present application.

[0016] Fig. 2 The schematic diagram of the intelligent residual coal cleaning method for the dumper of the present application.

[0017] Fig. 3 The another schematic diagram of the intelligent residual coal cleaning method for the dumper of the present application. DETAILED DESCRIPTION

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0019] like Figs. 1-3 This embodiment of an intelligent tippler coal cleaning method specifically includes:

[0020] S101. Obtain the internal space boundary information of the tipper and the cantilever motion trajectory planning results. Use an octree-based spatial collision detection algorithm to determine the feasibility of the fulcrum adjustment scheme. If there is a risk of spatial interference, further optimize the adjustment scheme until the spatial constraints are met.

[0021] After obtaining the boundary information of the tippler's internal space and the cantilever motion trajectory planning results, the boundary information is preprocessed. A Gaussian filtering algorithm is used to smooth and denoise the boundary point cloud data, improving the quality of the boundary information. Then, an octree model of the tippler's internal space is constructed for subsequent spatial collision detection. The cantilever motion trajectory is discretized to obtain a series of cantilever pose points. For each cantilever pose point, the corresponding cantilever geometric model is obtained. The cantilever geometric model can be modeled using CAD software and converted into a triangular mesh model. Spatial collision detection is performed in the octree model to determine whether there is interference between the cantilever mesh model and the tippler's internal space. Spatial collision detection is implemented using an axis-aligned bounding box tree (AABBTree) method. If spatial interference exists, the interference position information is obtained. Based on the interference position and type, an A heuristic search algorithm is used to optimize the pivot adjustment scheme. The A algorithm starts from the current pivot position and aims to eliminate spatial interference, searching for the optimal pivot position within a preset feasible region. The heuristic function is designed as the Euclidean distance from the current pivot position to the target position, used to estimate the cost. The optimized pivot adjustment scheme is applied to the cantilever trajectory planning to regenerate the trajectory. Based on the new pivot positions, a numerical iterative algorithm using the Jacobian matrix is ​​employed to solve the inverse kinematics of the cantilever, calculating the motion parameters of each joint and generating new joint spatial trajectories. Then, through forward kinematics calculations, the joint spatial trajectories are transformed into Cartesian space to form a new cantilever trajectory. The collision detection and trajectory optimization process is repeated until there is no spatial interference along the entire trajectory, or the preset maximum number of iterations is reached. The output is the pivot adjustment scheme and the optimized cantilever trajectory that satisfy the spatial constraints.

[0022] For example, after obtaining the boundary information of the tippler's internal space and the cantilever motion trajectory planning results, the boundary information is preprocessed. A Gaussian filtering algorithm is used to smooth and denoise the boundary point cloud data, improving the quality of the boundary information. Then, an octree model of the tippler's internal space is constructed for subsequent spatial collision detection. The cantilever motion trajectory is discretized to obtain a series of cantilever pose points. For each cantilever pose point, the corresponding cantilever geometric model is obtained. The cantilever geometric model can be modeled using CAD software and converted into a triangular mesh model. Spatial collision detection is performed in the octree model to determine whether there is interference between the cantilever mesh model and the tippler's internal space. Spatial collision detection is implemented using an axis-aligned bounding box tree (AABB Tree) method. If spatial interference exists, the interference position information is obtained. Based on the interference position and type, an A heuristic search algorithm is used to optimize the fulcrum adjustment scheme. The A algorithm starts from the current fulcrum position and aims to eliminate spatial interference, searching for the optimal fulcrum position within a preset feasible region. The heuristic function is designed as the Euclidean distance from the current fulcrum position to the target position, used to estimate the cost. The optimized pivot adjustment scheme is applied to the cantilever trajectory planning to regenerate the trajectory. Based on the new pivot positions, a numerical iterative algorithm using the Jacobian matrix is ​​employed to solve the inverse kinematics of the cantilever, calculating the motion parameters of each joint and generating new joint spatial trajectories. Then, through forward kinematics calculations, the joint spatial trajectories are transformed into Cartesian space to form a new cantilever trajectory. The collision detection and trajectory optimization process is repeated until there is no spatial interference along the entire trajectory, or the preset maximum number of iterations is reached. The output is the pivot adjustment scheme and the optimized cantilever trajectory that satisfy the spatial constraints.

[0023] S102. Obtain information on the properties of the remaining coal. If the remaining coal is dry coal powder, select a high-speed pneumatic cleaning tool with a dust collection device and control the cantilever to move at a low and uniform speed. If the remaining coal is damp coal lumps, select a cleaning tool with vibration and scraping functions and control the cantilever to move at a low speed.

[0024] The process involves acquiring information on the moisture content and particle size of the remaining coal; determining the type of remaining coal based on this information; if the remaining coal is dry coal powder, selecting a high-speed pneumatic cleaning tool with a dust collection device from a pre-set cleaning tool database; determining the maximum speed threshold for the cantilever's low-speed uniform motion using finite element analysis based on the cantilever's physical parameters; installing the high-speed pneumatic cleaning tool at the end of the cantilever; controlling the cantilever to move at a low-speed uniform speed at the maximum speed threshold; and simultaneously starting the high-speed pneumatic cleaning tool for cleaning operations. If the remaining coal is damp coal lumps, selecting a cleaning tool with vibration and scraping functions from the cleaning tool database; installing the cleaning tool at the end of the cantilever; controlling the cantilever to move at the minimum speed; and simultaneously starting the vibration and scraping devices of the cleaning tool to loosen and remove the damp coal lumps. During the cleaning process, real-time images of the remaining coal surface are acquired using an industrial camera; image processing algorithms are used to identify the degree of cleaning of the remaining coal; and based on the image recognition results, the cantilever's movement speed and the cleaning tool's vibration frequency parameters are dynamically adjusted using a control algorithm until the remaining coal is completely cleaned.

[0025] For example, the moisture content and particle size of the remaining coal are obtained using temperature and humidity sensors and a laser particle size analyzer to determine whether the remaining coal is dry coal powder or damp coal lumps. If the remaining coal is dry coal powder, a high-speed pneumatic cleaning tool with a dust collection device is selected from a preset cleaning tool database. Based on the cantilever's length, weight, and other physical parameters, the maximum speed threshold for the cantilever's low-speed uniform motion is calculated using finite element analysis to ensure uniform force on the cantilever and avoid deformation. The high-speed pneumatic cleaning tool with a dust collection device is installed at the end of the cantilever, and the cantilever is controlled by a servo motor to move at a low-speed uniform speed at the calculated maximum speed threshold. Simultaneously, the high-speed pneumatic device of the cleaning tool is activated to perform the cleaning operation, and the dust collection device collects the coal powder that falls during cleaning. If the remaining coal is damp coal lumps, a cleaning tool with vibration and scraping functions is selected from a preset cleaning tool database. The cleaning tool with vibration and scraping functions is installed at the end of the cantilever, and the cantilever is controlled by a servo motor to move at the lowest speed. Simultaneously, the vibration and scraping devices of the cleaning tool are activated to loosen and remove the damp coal lumps. During the cleaning process, industrial cameras capture real-time images of the remaining coal surface. Algorithms such as edge detection and threshold segmentation from the OpenCV image processing library are used to identify the degree of cleanup. Based on the image recognition results, parameters such as the cantilever speed and the vibration frequency of the cleaning tools are dynamically adjusted using a PID control algorithm until the remaining coal is completely removed.

[0026] S103. Obtain the distribution of remaining coal and the current pose of the cantilever. Use the A* algorithm to solve the optimal cleaning path under obstacle avoidance constraints. Set a cleaning completion threshold based on image processing. Determine whether the current area meets the predetermined cleaning standard. If it does not meet the standard, return to the area for repeated cleaning until the requirements are met.

[0027] After acquiring the remaining coal distribution information and the current pose information of the cantilever, the acquired information is preprocessed to extract effective information that can be used for the A algorithm. The preprocessed information is input into the A algorithm, and obstacle avoidance constraints are considered to plan and generate the optimal cleaning path. According to the preset cleaning completion threshold, image processing techniques such as image segmentation and binarization in OpenCV are used to process the cleaned area image to extract the cleaned remaining coal area. The area ratio of the cleaned remaining coal area is calculated and used as the current cleaning completion level. It is determined whether the current cleaning completion level has reached the preset threshold. If it has, the area is considered to be cleaned and the process moves to the next area; if it has not, the A algorithm is used to generate a path back to the incomplete area, and the cantilever is controlled to return to that area. According to the return path, the cantilever is controlled to return to the incomplete area and the cleaning operation is repeated. The cleaned image is acquired again, the cleaning completion level is recalculated, and it is determined whether it has reached the target. The above cleaning, return, and judgment process is repeated until the cleaning completion level of the current area meets the preset threshold requirement, or the number of repetitions reaches the set upper limit. Areas where the cleaning completion rate reaches a preset threshold are marked as cleaned areas and removed from the remaining coal distribution information, which is then updated. Images of the cleaned areas are replaced with images of areas without coal, updating the remaining coal distribution map. Based on the updated remaining coal distribution information, the cleaning operation proceeds to the next area until all areas are cleaned.

[0028] For example, after acquiring the remaining coal distribution information and the current pose information of the cantilever, the acquired information is preprocessed to extract effective information that can be used for Algorithm A. The preprocessed information is input into Algorithm A, and obstacle avoidance constraints are considered to plan and generate the optimal cleaning path. According to the preset cleaning completion threshold, image processing techniques such as image segmentation and binarization in OpenCV are used to process the image of the cleaned area to extract the remaining coal area. The area ratio of the remaining coal area after cleaning is calculated and used as the current cleaning completion. It is determined whether the current cleaning completion has reached the preset threshold. If it has, the area is considered to be cleaned and the process moves to the next area; if it has not, Algorithm A is used to generate a path back to the incomplete area, and the cantilever is controlled to return to that area. According to the return path, the cantilever is controlled to return to the incomplete area and the cleaning operation is repeated. The cleaned image is acquired again, the cleaning completion is recalculated, and it is determined whether it meets the standard. The above cleaning, returning, and judging process is repeated until the cleaning completion of the current area meets the preset threshold requirement, or the number of repetitions reaches the set upper limit. Areas where the cleaning completion rate reaches a preset threshold are marked as cleaned areas and removed from the remaining coal distribution information, which is then updated. Images of the cleaned areas are replaced with images of areas without coal, updating the remaining coal distribution map. Based on the updated remaining coal distribution information, the cleaning operation proceeds to the next area until all areas are cleaned.

[0029] S104. Employs a deep learning-based human detection and tracking algorithm to monitor the location and behavior of personnel in the work area in real time. If personnel are detected entering a dangerous area or engaging in unsafe behavior, an alarm mechanism is triggered to control the cantilever to stop moving or move to a safe position.

[0030] The system acquires video images of the work area and uses a pre-trained human detection model to detect and locate personnel in the images, obtaining their position coordinates. These coordinates are then compared to a preset danger zone coordinate range to determine if the personnel have entered the danger zone. If the personnel's coordinates fall within the danger zone, an alarm mechanism is triggered. A human posture estimation model based on key point detection is used to detect key body parts of the personnel, obtaining the coordinates of key points on the human skeleton. These key point coordinates are input into a pre-trained behavior recognition model to classify and identify the personnel's actions. If an unsafe behavior is identified, an alarm mechanism is triggered. This alarm mechanism includes issuing an audible and visual alarm signal and sending the alarm information to the cantilever control system. Upon receiving the alarm information, the cantilever control system stops the current movement of the cantilever and records its position coordinates at the time of stopping. It then selects the nearest safe coordinate from a preset safe position coordinate list and moves the cantilever to that safe coordinate position. When it detects that a personnel has left the danger zone and has not re-entered it within a certain period, it controls the cantilever to return to the position coordinates recorded at the time of stopping and continues the previous work process.

[0031] For example, video images of the work area are captured by a camera, and computer vision technology is used to analyze the images in real time. A pre-trained human detection model, such as YOLO or Faster R-CNN, is used to detect and locate personnel in the images, obtaining their position coordinates. After obtaining the personnel's position coordinates, they are compared with the coordinate range of a preset danger zone. The coordinates of the danger zone are pre-marked and entered into the system by management personnel based on the actual situation of the work site. If the personnel's position coordinates are detected to enter the danger zone, an alarm mechanism is triggered. Simultaneously, a human pose estimation model based on key point detection, such as OpenPose, is used to detect key body parts of the personnel, obtaining the coordinate information of key points of the human skeleton. The key point coordinates are input into a pre-trained behavior recognition model, such as LSTM, to classify and recognize the personnel's actions. The behavior recognition model is trained with a large amount of sample data of normal and unsafe behaviors, and can determine whether the current personnel's behavior belongs to a predefined unsafe behavior category. If an unsafe behavior is identified, an alarm mechanism is also triggered. The alarm mechanism includes issuing audible and visual alarm signals to remind the workers to pay attention to safety, and simultaneously sending alarm information to the cantilever control system. Upon receiving an alarm message, the control system immediately stops the cantilever's current movement and records its position coordinates at the time of stopping. After the cantilever stops, the control system selects the nearest safe coordinate from a pre-set list of safe position coordinates and moves the cantilever to that position. These safe position coordinates are pre-marked and entered into the system by management personnel based on the layout of the work site, ensuring that the cantilever is away from hazardous areas and personnel at this position. When personnel are detected leaving the hazardous area and do not re-enter within a certain period, the site is confirmed to be safe. The control system issues a resumption command, controlling the cantilever to return to the previously recorded position coordinates and continue the previous work process. Throughout the process, cameras continuously capture video images of the work area, cyclically performing personnel detection, position determination, and behavior recognition steps to monitor personnel safety status in real time. If personnel are detected re-entering the hazardous area or engaging in unsafe behavior, an alarm and stop operation are immediately triggered to ensure the safety of the personnel.

[0032] S105. Collect and analyze equipment operating parameters and vibration signal data, construct an equipment health status assessment model based on support vector machine, realize early warning and diagnosis of equipment failure, and generate maintenance suggestions and solutions if equipment abnormality is found.

[0033] The process involves acquiring equipment operating parameters and vibration signal data, preprocessing the data to obtain a preprocessed dataset, dividing the preprocessed dataset into a training set and a test set, constructing an equipment health status assessment model on the training set using a support vector machine algorithm, and optimizing the hyperparameters of the assessment model through grid search, inputting the real-time collected operating parameters and vibration signal data of the equipment into the assessment model to obtain a health status score for the equipment, determining that the equipment is abnormal if the health status score is lower than a preset threshold, based on the abnormal parameters and signal characteristics of the equipment, combined with a pre-established expert knowledge base and fault case database, using an expert system for rule-based reasoning, and considering maintenance costs and equipment downtime to determine the optimal maintenance plan, sending the optimal maintenance plan to relevant personnel to guide them in equipment maintenance and repair, and updating the maintenance process and results in the expert knowledge base and fault case database, continuously monitoring the equipment operating status, periodically inputting equipment data into the assessment model to dynamically assess the equipment health status, adjusting the equipment maintenance strategy based on the assessment results, and triggering a fault diagnosis and maintenance decision-making process when the equipment health status declines.

[0034] For example, operating parameters and vibration signal data of the equipment are collected by sensors. The collected data is preprocessed, including data cleaning, feature extraction, and data normalization, to obtain a dataset suitable for modeling. The preprocessed dataset is divided into training and testing sets. A support vector machine algorithm is used to build an equipment health status assessment model on the training set. The model hyperparameters are optimized using methods such as grid search, and the model performance is evaluated on the testing set to ensure that the model can accurately assess the health status of the equipment. The operating parameters and vibration signal data collected in real time by the equipment are input into the trained assessment model, and the model outputs a health status score for the equipment. If the score is lower than a preset threshold, the equipment is judged to be abnormal. Based on the abnormal parameters and signal characteristics, combined with the expert knowledge base and fault case database, the expert system performs rule reasoning to generate targeted equipment maintenance suggestions and solutions. The feasibility analysis and optimization of the generated maintenance suggestions and solutions are performed using a multi-objective optimization algorithm, taking into account factors such as maintenance costs and equipment downtime to determine the optimal maintenance solution. The optimized maintenance solution is sent to relevant personnel to guide them in equipment maintenance and upkeep. At the same time, the maintenance process and results are fed back to the system to update the expert knowledge base and fault case database. Continuously monitor equipment operating status and periodically input equipment data into the evaluation model to dynamically assess equipment health. Adjust equipment maintenance strategies based on the evaluation results, and trigger fault diagnosis and maintenance decision-making processes when equipment health deteriorates. Through this closed-loop control strategy, predictive maintenance and lifespan management of equipment are achieved, improving equipment operational reliability and production efficiency.

[0035] S106. Optimize cantilever motion planning and control strategies to reduce energy consumption and mechanical wear, monitor equipment status in real time, and dynamically adjust cleaning parameters based on health assessment results to ensure safe and efficient operation of the equipment.

[0036] The system acquires cantilever vibration and temperature data, which are collected in real time by sensors. Based on this data, a support vector regression algorithm is used to construct a device health assessment model. The health status level of the device is determined using this model. If the health status level is below a preset threshold, the cleaning motion parameters, including operating speed, acceleration, and amplitude, are dynamically adjusted. Energy consumption data under different operating parameters is acquired. Based on this energy consumption data, a CART decision tree algorithm is used to establish a mapping model between the device operating parameters and energy consumption. The optimal combination of operating parameters for energy consumption is determined using this mapping model. The wear status assessment results of the device are then obtained. A Bayesian network algorithm is used to establish a correlation model between wear state and grease application amount and application cycle. Through this correlation model, lubrication parameters, including grease application amount and application cycle, are dynamically optimized. Based on equipment health status, energy consumption optimization results, and wear prediction information, a weighted summation method is used to determine the optimal cleaning control strategy, which includes optimized cantilever motion parameters and lubrication maintenance parameters. The optimal cleaning control strategy is then sent to the cantilever control unit. The cantilever control unit adjusts the cantilever motion in real time using closed-loop control based on the received motion parameters. Finally, the cantilever control unit performs periodic maintenance according to the optimized lubrication strategy based on the received lubrication maintenance parameters.

[0037] For example, motion control parameters are dynamically optimized based on the cantilever's motion state and equipment health status. First, sensors collect real-time data on cantilever vibration, temperature, and other status data. Using this data, a health assessment model is constructed based on a support vector regression algorithm to determine the equipment's health status level. If the equipment's health status is below a preset threshold, the cleaning motion parameters are dynamically adjusted to reduce operating speed and acceleration, decrease cantilever amplitude, and ensure safe equipment operation. Energy consumption and wear factors are comprehensively considered when optimizing motion control parameters. Energy consumption data is obtained by conducting energy consumption tests under different equipment operating parameters. Using this data, a CART decision tree algorithm is used to establish a mapping model between equipment operating parameters and energy consumption, finding the optimal combination of operating parameters for energy consumption. Simultaneously, based on the equipment's wear status assessment results, a Bayesian network algorithm is used to establish a correlation model between wear status and grease application amount and application cycle, dynamically optimizing lubrication parameters. After obtaining the equipment health status, energy consumption optimization results, and wear prediction information, a weighted summation method is used to weigh the three optimization objectives, resulting in the optimal cleaning control strategy that balances safety, energy saving, and wear reduction. The optimal control strategy includes optimized motion parameters such as cantilever operating speed, acceleration, and amplitude, as well as maintenance parameters such as grease application amount and application cycle. The optimized cleaning control parameters, including motion control parameters and lubrication maintenance parameters, are sent to the cantilever control unit. Based on the received parameters, the cantilever control unit adjusts the cantilever movement in real time through closed-loop control and performs regular maintenance according to the optimized lubrication strategy to ensure continuous, efficient, and safe operation of the equipment.

[0038] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for cleaning residual coal on an intelligent tipper, characterized in that, The method includes: The internal space boundary information of the tipper and the cantilever motion trajectory planning results are obtained. An octree-based spatial collision detection algorithm is used to determine the feasibility of the fulcrum adjustment scheme. If there is a risk of spatial interference, the adjustment scheme is further optimized until the spatial constraints are met. Obtain information on the properties of the remaining coal. If the remaining coal is dry coal powder, select a high-speed pneumatic cleaning tool with a dust collection device and control the cantilever to move at a low and uniform speed. If the remaining coal is wet coal lumps, select a cleaning tool with vibration and scraping functions and control the cantilever to move at a low speed. The system obtains the distribution of remaining coal and the current pose of the cantilever. It uses the A* algorithm to solve the optimal cleaning path under obstacle avoidance constraints. It sets a cleaning completion threshold based on image processing to determine whether the current area meets the predetermined cleaning standard. If it does not meet the standard, it returns to the area for repeated cleaning until the requirements are met. Using deep learning-based human detection and tracking algorithms, the system monitors the location and behavior of personnel in the work area in real time. If personnel are detected entering a dangerous area or engaging in unsafe behavior, an alarm mechanism is triggered to control the cantilever to stop moving or move to a safe position. Collect and analyze equipment operating parameters and vibration signal data, construct an equipment health status assessment model based on support vector machine, realize early warning and diagnosis of equipment failure, and generate maintenance suggestions and solutions if equipment abnormality is detected. Optimize cantilever motion planning and control strategies to reduce energy consumption and mechanical wear, monitor equipment status in real time, and dynamically adjust cleaning parameters based on health assessment results to ensure safe and efficient equipment operation.

2. The method according to claim 1, characterized in that, The process involves acquiring the internal spatial boundary information of the tipper and the cantilever motion trajectory planning results, employing an octree-based spatial collision detection algorithm to determine the feasibility of the fulcrum adjustment scheme. If spatial interference risks exist, the adjustment scheme is further optimized until the spatial constraints are met, including: After obtaining the internal space boundary information of the tipper and the cantilever motion trajectory planning results, the boundary information is preprocessed. Gaussian filtering algorithm is used to smooth and denoise the boundary point cloud data, thereby improving the quality of boundary information; Then, an octree model of the internal space of the tipper is constructed for subsequent spatial collision detection. Discretize the cantilever motion trajectory to obtain a series of cantilever pose points; For each cantilever pose point, obtain the corresponding cantilever geometric model; The cantilever geometry model can be modeled using CAD software and converted into a triangular mesh model; Spatial collision detection is performed in the octree model to determine whether there is interference between the cantilever mesh model and the internal space of the tipper. Spatial collision detection is implemented using an axis-aligned bounding box tree (AABBTree) method. If spatial interference exists, obtain the interference location information; Based on the location and type of interference, the A heuristic search algorithm is used to optimize the fulcrum adjustment scheme; Algorithm A starts from the current fulcrum position and aims to eliminate spatial interference by searching for the optimal fulcrum position within the preset feasible region. The heuristic function is designed as the Euclidean distance from the current pivot position to the target position, used to estimate the cost; The optimized fulcrum adjustment scheme is applied to the cantilever motion trajectory planning to regenerate the motion trajectory. Based on the new fulcrum position, a numerical iterative algorithm that cyclically utilizes the Jacobian matrix is ​​used to solve the inverse kinematics of the cantilever, calculate the motion parameters of each joint of the cantilever, and generate a new joint space trajectory. Then, through forward kinematics calculations, the joint space trajectory is transformed into Cartesian space to form a new cantilever motion trajectory; Repeat the above collision detection and trajectory optimization process until there is no spatial interference on the entire motion trajectory, or until the preset maximum number of iterations is reached; Output the fulcrum adjustment scheme that satisfies the spatial constraints and the optimized cantilever motion trajectory.

3. The method according to claim 1, characterized in that, The process of obtaining residual coal property information involves, if the residual coal is dry coal powder, selecting a high-speed pneumatic cleaning tool with a dust collection device and controlling the cantilever to move at a low, uniform speed; if the residual coal is damp coal lumps, selecting a cleaning tool with vibration and scraping functions and controlling the cantilever to move at a low speed, including: Obtain information on the moisture content and particle size distribution of the remaining coal; Based on the aforementioned property information, determine the type of the remaining coal; If the remaining coal is dry coal powder, a high-speed pneumatic cleaning tool with a dust collection device is selected from the preset cleaning tool database. Based on the physical parameters of the cantilever, the maximum speed threshold for the low-speed uniform motion of the cantilever is determined by finite element analysis. The high-speed pneumatic cleaning tool is installed at the end of the cantilever, and the cantilever is controlled to move at a low speed uniformly at the maximum speed threshold. At the same time, the high-speed pneumatic cleaning tool is started to perform cleaning operations. If the remaining coal is a damp coal lump, a cleaning tool with vibration and scraping functions is selected from the cleaning tool database, the cleaning tool is installed at the end of the cantilever, the cantilever is controlled to move at the lowest speed, and the vibration and scraping device of the cleaning tool is activated to loosen and remove the damp coal lump. During the cleaning process, industrial cameras are used to collect real-time images of the remaining coal surface. Image processing algorithms are used to identify the degree of cleaning of the remaining coal. Based on the image recognition results, the control algorithm dynamically adjusts the cantilever movement speed and the vibration frequency parameters of the cleaning tool until the remaining coal is completely cleaned.

4. The method according to claim 1, characterized in that, The process involves acquiring the remaining coal distribution and the current cantilever pose information, using the A* algorithm to solve for the optimal cleaning path under obstacle avoidance constraints, setting a cleaning completion threshold based on image processing, and determining whether the current area meets the predetermined cleaning standard. If it does not meet the standard, the process returns to that area for repeated cleaning until the requirements are met, including: After obtaining the remaining coal distribution information and the current pose information of the cantilever, the obtained information is preprocessed to extract the effective information that can be used for the A algorithm; The preprocessed information is input into Algorithm A, and the obstacle avoidance constraints are taken into account to plan and generate the optimal cleaning path. Based on the preset cleaning completion threshold, image processing techniques such as image segmentation and binarization in OpenCV are used to process the cleaned area image and extract the remaining coal area after cleaning. Calculate the area percentage of the remaining coal area after cleaning and use it as the current cleaning completion rate; Determine whether the current cleaning completion rate has reached the preset threshold. If it has, consider the area to be cleaned and proceed to the next area. If the target is not reached, the A algorithm is used to generate a path back to the area that has not been cleaned, and the cantilever is controlled to return to that area. Based on the return path, control the cantilever to return to the area that did not meet the standard and repeat the cleaning operation; The image after cleaning is acquired again, the cleaning completion rate is recalculated, and it is determined whether the standard is met. Repeat the above cleaning, return, and judgment process until the cleaning completion rate of the current area meets the preset threshold requirement, or the number of repetitions reaches the set upper limit; Areas that have reached a preset threshold for cleaning completion are marked as cleaned areas and removed from the remaining coal distribution information, and the remaining coal distribution information is updated. Replace the image of the cleaned area with an image of no coal, and update the remaining coal distribution map; Based on the updated information on the distribution of remaining coal, proceed to the next area for cleaning operations until all areas have been cleaned.

5. The method according to claim 1, characterized in that, The system employs a deep learning-based human detection and tracking algorithm to monitor the location and behavior of personnel within the work area in real time. If personnel are detected entering a dangerous area or engaging in unsafe behavior, an alarm mechanism is triggered, controlling the cantilever to stop moving or relocate to a safe position. This includes: Acquire video images within the work area, use a pre-trained human detection model to detect and locate people in the images, and obtain the location coordinate information of the people; The personnel location coordinates are compared with the preset danger zone coordinate range to determine whether the personnel have entered the danger zone. If a person's location coordinates enter the danger zone, an alarm mechanism will be triggered. A human pose estimation model based on key point detection is used to detect key body parts of a person and obtain the coordinate information of key points of the human skeleton. The key point coordinates are input into a pre-trained behavior recognition model to classify and recognize human behavior actions; If unsafe behavior is detected, an alarm mechanism is triggered; The alarm mechanism includes issuing audible and visual alarm signals and sending alarm information to the cantilever control system; After receiving the alarm information, the cantilever control system controls the cantilever to stop its current movement and records the position coordinates at the time of stopping. Select the nearest safe coordinate from the preset safe position coordinate list, and control the cantilever to move to that safe coordinate position; When personnel are detected leaving the danger zone and do not re-enter within a certain period of time, the control boom returns to the position coordinates recorded when it stopped, and the previous work process continues.

6. The method according to claim 1, characterized in that, The system collects and analyzes equipment operating parameters and vibration signal data to construct a support vector machine-based equipment health status assessment model. This model enables early warning and diagnosis of equipment faults. If an equipment abnormality is detected, maintenance suggestions and solutions are generated, including: Acquire equipment operating parameters and vibration signal data, preprocess the data, and obtain a preprocessed dataset; The preprocessed dataset is divided into a training set and a test set; A device health status assessment model is constructed on the training set using the support vector machine algorithm, and the hyperparameters of the assessment model are optimized by grid search. The real-time operating parameters and vibration signal data collected by the equipment are input into the evaluation model to obtain the health status score of the equipment; If the health status score is lower than a preset threshold, the device is determined to be abnormal. Based on the abnormal parameters and signal characteristics of the equipment, combined with a pre-established expert knowledge base and fault case database, an expert system is used to perform rule-based reasoning, and the optimal maintenance plan is determined by considering maintenance costs and equipment downtime. The optimal maintenance plan is sent to relevant personnel to guide them in equipment maintenance and repair. The maintenance process and results are fed back to the expert knowledge base and fault case database for updates. Continuously monitor the equipment's operating status, periodically input equipment data into the evaluation model, and dynamically assess the equipment's health status; Adjust equipment maintenance strategies based on assessment results, and trigger fault diagnosis and maintenance decision-making processes when equipment health deteriorates.

7. The method according to claim 1, characterized in that, The optimized cantilever motion planning and control strategy reduces energy consumption and mechanical wear, monitors equipment status in real time, and dynamically adjusts cleaning parameters based on health assessment results to ensure safe and efficient equipment operation, including: Acquire cantilever vibration and temperature status data, which are collected in real time by sensors; Based on the aforementioned status data, a support vector regression algorithm is used to construct an equipment health assessment model; The equipment health status level is determined using the equipment health assessment model. If the health status level of the device is lower than a preset threshold, the cleaning motion parameters are dynamically adjusted. The cleaning motion parameters include running speed, acceleration, and amplitude. Obtain energy consumption data under different device operating parameters; Based on the energy consumption data, a mapping model between equipment operating parameters and energy consumption is established using the CART decision tree algorithm; The optimal combination of operating parameters for energy consumption is determined using the mapping model. Obtain the wear condition assessment results of the equipment; Based on the wear state assessment results, a correlation model between wear state and grease filling amount and filling cycle is established using a Bayesian network algorithm; The correlation model is used to dynamically optimize lubrication parameters, including grease application amount and application cycle. Based on the equipment health status, energy consumption optimization results and wear prediction information, the optimal cleaning control strategy is determined by weighted summation method. The optimal cleaning control strategy includes optimized cantilever motion parameters and lubrication maintenance parameters. The optimal cleaning control strategy is sent to the cantilever control unit; The cantilever control unit adjusts the cantilever motion in real time through closed-loop control based on the received motion parameters. The cantilever control unit performs regular maintenance according to the received lubrication maintenance parameters and the optimized lubrication strategy.

Images