Bulk material stacking method and system based on stacker-reclaimer

Abstract

The invention provides a bulk cargo stacking method and system based on a stacker-reclaimer, and belongs to the technical field of bulk cargo stacking and reclaiming, the bulk cargo stacking method comprises the steps that sensor data information and bulk cargo distribution information of a to-be-stacked area are acquired, and the data information comprises environment data and environment image information; processing the sensor data through the preprocessing model to obtain three-dimensional space data information; inputting the three-dimensional data and the bulk cargo distribution information into a bulk cargo stacking planning model, and calculating to obtain a stacking path and motion control data of the stacker-reclaimer; processing the accumulation path and the control data through an accumulation shape planning model to obtain accumulation shape data; the stacker-reclaimer executes bulk cargo stacking operation, and control data are optimized and adjusted through real-time feedback; and generating a bulk cargo stacking scheme. Through real-time optimization and intelligent adjustment, the stacking efficiency is effectively improved, material waste is reduced, and the shape and stability of a stacking body are ensured.

Description

Technical Field

[0001] The present invention belongs to the technical field of bulk material stacking and reclaiming, and particularly relates to a bulk material stacking method and system based on a stacker-reclaimer. Background Art

[0002] With the rapid development of modern industries such as mines, ports, and logistics, bulk material stacking operations play an important role in the processes of material stacking, loading, unloading, and transportation. According to the report of the International Material Handling Association, the global bulk material handling market size has exceeded one trillion US dollars, but the operation efficiency improvement rate is less than 3%, indicating that the traditional operation mode has difficulty meeting the requirements of modern industrial development.

[0003] Most traditional bulk material stacking methods rely on manual experience or mechanical manual operations, and usually cannot achieve efficient, accurate, and automated stacking operations. In the prior art, many stacking methods rely on fixed stacking patterns and simple path planning, and do not fully consider material characteristics, equipment working conditions, and real-time environmental factors, resulting in difficulty in dynamically adjusting and optimizing the stacking plan in complex scenarios.

[0004] Although the bulk material stacking methods in the prior art can meet some basic requirements, when facing complex bulk material stacking environments, they often have difficulty in dealing with problems such as variable bulk material shapes, uneven stacking positions, and real-time dynamic changes. In addition, traditional methods lack intelligent optimization of stacking shapes, path planning, and equipment control, resulting in problems such as large resource waste, unstable stacking, and low equipment utilization efficiency during the bulk material stacking process, seriously affecting the efficiency and safety of bulk material stacking operations. Summary of the Invention

[0005] In view of the problems in the prior art, the present invention provides a bulk material stacking method and system based on a stacker-reclaimer, which solves the problems of low stacking accuracy and operation efficiency caused by the inability to dynamically adjust the bulk material stacking path and the movement trajectory of the stacker-reclaimer in the prior art.

[0006] The technical solution adopted by the present invention is as follows: In a first aspect, the present application provides a bulk material stacking method based on a stacker-reclaimer, including the following steps: Step S1: Obtain the data information uploaded by the sensors in the area to be stacked and the bulk material distribution information, where the data information uploaded by the sensors includes the environmental data information and environmental image information of the area to be stacked; Step S2: Send the data information uploaded by the sensors in the area to be stacked to a preprocessing model for preprocessing to obtain the three-dimensional space data information of the area to be stacked; Step S3: Send the three-dimensional spatial data information of the area to be stacked, the data information uploaded by the sensors in the area to be stacked, and the bulk material distribution information to a preset bulk material stacking planning model for processing to obtain the stacking path of the bulk material and the motion control data of the stacker-reclaimer; Step S4: Send the stacking path of the bulk material and the motion control data of the stacker-reclaimer to a preset stacking shape planning model for processing to obtain the stacking shape data of the bulk material; Step S5: Combine the stacking shape data of the bulk material, the stacking path of the bulk material, and the motion control data of the stacker-reclaimer to perform bulk material stacking, and perform real-time optimization on the motion control data of the stacker-reclaimer based on the real-time obtained bulk material stacking result to obtain the final bulk material stacking plan.

[0007] Preferably, step S2 includes the following steps: Step S21: Denoise the data information uploaded by the sensors in the area to be stacked based on a preset mean shift clustering algorithm to obtain the smoothed historical material height and historical position data of the area to be stacked; Step S22: Merge the environmental image information of the area to be stacked based on a super-resolution reconstruction algorithm to obtain the high-resolution environmental image information of the area to be stacked; Step S23: Perform a weighted regression operation based on the smoothed historical material height and historical position data of the area to be stacked to determine the stacking height and position distribution information of the materials in the area to be stacked; Step S24: Process the high-resolution environmental image information of the area to be stacked, the stacking height and position distribution information of the materials in the area to be stacked based on a double integral weighting operation to obtain the three-dimensional spatial data information of the area to be stacked.

[0008] Preferably, step S3 includes the following steps: Step S31: Map the three-dimensional spatial data information of the area to be stacked, the data information uploaded by the sensors in the area to be stacked, and the bulk material distribution information into a preset high-dimensional space and perform processing through a fast dynamic random walk algorithm to obtain a preliminary path planning trajectory for bulk material stacking; Step S32: Perform fractal geometry optimization processing based on the three-dimensional spatial data information of the area to be stacked and the preliminary path planning trajectory for bulk material stacking, and select an optimized path planning trajectory for bulk material stacking by analyzing the complexity of the path and a preset simulated annealing algorithm; Step S33: Select a motion control strategy based on a preset multi-armed bandit algorithm for the optimized path planning trajectory for bulk material stacking and the preset parameter information of the stacker-reclaimer, and obtain the motion control data of the stacker-reclaimer by using the preset historical motion control strategy as a control parameter; Step S34: Perform real-time operations based on the optimized bulk material stacking path planning trajectory and the motion control data of the stacker-reclaimer, and perform real-time updates on the bulk material stacking path planning trajectory and the motion control data of the stacker-reclaimer based on the real-time operation feedback and the adaptive control algorithm to obtain the real-time updated bulk material stacking path planning trajectory and the motion control data of the stacker-reclaimer.

[0009] Preferably, step S4 includes the following steps: Step S41: Process the stacking path of the bulk material and the motion control data of the stacker-reclaimer based on a preset Poisson reconstruction algorithm to obtain the maximum stacking height and shape boundary of the bulk material; Step S42: Reduce the dimension of the bulk material distribution information and the stacking path of the bulk material based on a preset manifold learning method, and calculate the optimal stacking angle and direction through the stacking path of each bulk material and the bulk material distribution information; Step S43: Perform stability analysis based on the maximum stacking height, maximum stacking shape boundary, stacking angle, and stacking direction of the bulk material to obtain the stability analysis of the stacked body. If the analysis result shows that the stability is greater than a preset threshold, then use the maximum stacking height, maximum stacking shape boundary, stacking angle, and stacking direction of the bulk material as the stacking shape data of the bulk material.

[0010] Preferably, step S34 includes the following steps: Step S341: Perform real-time adjustment on the stacking path based on the data information of the real-time operation feedback and the particle swarm optimization algorithm to obtain the adjusted stacking path information; Step S342: Perform real-time adjustment on the data information of the real-time operation feedback and the motion control data of the stacker-reclaimer based on the adaptive control algorithm to obtain the adjusted motion control data of the stacker-reclaimer; Step S343: Perform a safety risk assessment based on the adjusted stacking path information and the motion control data of the stacker-reclaimer to obtain the risk assessment data of the bulk material stacking; Step S344: Perform another adjustment on the adjusted stacking path information and the motion control data of the stacker-reclaimer based on the risk assessment data of the bulk material stacking and a preset reinforcement learning algorithm to obtain the real-time updated bulk material stacking path planning trajectory and the motion control data of the stacker-reclaimer.

[0011] In a second aspect, the present application also provides a bulk material stacking system based on a stacker-reclaimer, including: An acquisition unit for acquiring the data information uploaded by the sensors in the area to be stacked and the bulk material distribution information, where the data information uploaded by the sensors includes the environmental data information and environmental image information of the area to be stacked; The first processing unit is used to send the data information uploaded by the sensors in the area to be stacked to a preprocessing model for preprocessing, so as to obtain the three-dimensional space data information of the area to be stacked; The second processing unit is used to send the three-dimensional space data information of the area to be stacked, the data information uploaded by the sensors in the area to be stacked, and the bulk material distribution information to a preset stacking planning model of bulk materials for processing, so as to obtain the stacking path of bulk materials and the motion control data of the stacker-reclaimer; The third processing unit is used to send the stacking path of bulk materials and the motion control data of the stacker-reclaimer to a preset stacking shape planning model for processing, so as to obtain the stacking shape data of bulk materials; The optimization unit is used to combine the stacking shape data of bulk materials, the stacking path of bulk materials, and the motion control data of the stacker-reclaimer to perform bulk material stacking, and perform real-time optimization on the motion control data of the stacker-reclaimer based on the real-time obtained bulk material stacking result, so as to obtain the final bulk material stacking plan.

[0012] Preferably, the first processing unit includes: The first processing subunit is used to perform denoising processing on the data information uploaded by the sensors in the area to be stacked based on a preset mean shift clustering algorithm, so as to obtain the historical material height and historical position data of the area to be stacked after smoothing; The second processing subunit is used to merge the environmental image information of the area to be stacked based on a super-resolution reconstruction algorithm, so as to obtain the high-resolution environmental image information of the area to be stacked; The third processing subunit is used to perform weighted regression operation based on the historical material height and historical position data of the area to be stacked after smoothing, so as to determine the stacking height and position distribution information of the materials in the area to be stacked; The fourth processing subunit is used to perform processing on the high-resolution environmental image information of the area to be stacked, the stacking height and position distribution information of the materials in the area to be stacked based on a double integral weighting operation, so as to obtain the three-dimensional space data information of the area to be stacked.

[0013] Preferably, the second processing unit includes: The fifth processing subunit is used to map the three-dimensional space data information of the area to be stacked, the data information uploaded by the sensors in the area to be stacked, and the bulk material distribution information into a preset high-dimensional space, and perform processing through a fast dynamic random walk algorithm, so as to obtain a preliminary path planning trajectory for bulk material stacking; The sixth processing subunit is used to perform fractal geometry optimization processing based on the three-dimensional space data information of the area to be stacked and the preliminary path planning trajectory for bulk material stacking. Among them, an optimized path planning trajectory for bulk material stacking is selected by analyzing the complexity of the path and a preset simulated annealing algorithm; The seventh processing subunit is configured to select a motion control strategy based on a preset multi-armed bandit algorithm for the optimized planning trajectory of bulk material stacking and the parameter information of a preset stacker-reclaimer, wherein the motion control data of the stacker-reclaimer is obtained by using the preset historical motion control strategy as a control parameter; The eighth processing subunit is configured to perform real-time operations based on the optimized bulk material stacking path planning trajectory and the motion control data of the stacker-reclaimer, and perform real-time updates on the bulk material stacking path planning trajectory and the motion control data of the stacker-reclaimer based on the real-time operation feedback and an adaptive control algorithm, so as to obtain the real-time updated bulk material stacking path planning trajectory and the motion control data of the stacker-reclaimer.

[0014] Preferably, the third processing unit includes: The ninth processing subunit is configured to process the stacking path of the bulk material and the motion control data of the stacker-reclaimer based on a preset Poisson reconstruction algorithm to obtain the maximum stacking height and shape boundary of the bulk material; The first calculation subunit is configured to reduce the dimension of the bulk material distribution information and the stacking path of the bulk material based on a preset manifold learning method, and calculate the optimal stacking angle and direction through the stacking path and the bulk material distribution information of each bulk material; The first analysis subunit is configured to perform stability analysis based on the maximum stacking height, maximum stacking shape boundary, stacking angle, and stacking direction of the bulk material to obtain the stability analysis of the stacked body. If the analysis result shows that the stability is greater than a preset threshold, the maximum stacking height, maximum stacking shape boundary, stacking angle, and stacking direction of the bulk material are used as the stacking shape data of the bulk material.

[0015] Preferably, the eighth processing subunit includes: The first adjustment subunit is configured to perform real-time adjustment on the stacking path based on the data information of the real-time operation feedback and a particle swarm optimization algorithm to obtain the adjusted stacking path information; The second adjustment subunit is configured to perform real-time adjustment on the data information of the real-time operation feedback and the motion control data of the stacker-reclaimer based on an adaptive control algorithm to obtain the adjusted motion control data of the stacker-reclaimer; The second analysis subunit is configured to perform a safety risk assessment based on the adjusted stacking path information and the motion control data of the stacker-reclaimer to obtain the risk assessment data of the bulk material stacking; The third adjustment subunit is configured to perform re-adjustment on the adjusted stacking path information and the motion control data of the stacker-reclaimer based on the risk assessment data of the bulk material stacking and a preset reinforcement learning algorithm to obtain the real-time updated bulk material stacking path planning trajectory and the motion control data of the stacker-reclaimer.

[0016] It can be seen from the above technical solutions that the present invention has the following advantages: The present invention obtains the environmental data and bulk material distribution information of the area to be stacked, and processes the data by combining intelligent algorithms to optimize the stacking path and the motion control of the stacker-reclaimer, thereby achieving efficient and precise bulk material stacking. Through a series of preprocessing models and planning models, such as mean shift clustering, super-resolution reconstruction, fractal geometry optimization, particle swarm optimization, and reinforcement learning, the present invention dynamically adjusts the bulk material stacking path and the motion trajectory of the stacker-reclaimer, can perform adaptive adjustment according to real-time feedback, and significantly improves the stacking accuracy and operation efficiency. At the same time, combined with the risk assessment mechanism and stability analysis, the present invention can also effectively identify and reduce the potential risks in the bulk material stacking process, ensuring the safety and reliability of the operation process. Compared with the prior art, the present invention not only improves the work efficiency, but also greatly reduces the resource waste and potential safety hazards caused by improper stacking path planning or inaccurate equipment motion control. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present invention, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0018] Figure 1 It is a schematic flow chart of the bulk material stacking method based on the stacker-reclaimer described in the embodiments of the present invention; Figure 2 It is a schematic structural diagram of the bulk material stacking system based on the stacker-reclaimer described in the embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] In the following detailed description, various embodiments of the present disclosure will be more fully described. The present disclosure may have various embodiments, and adjustments and changes may be made therein. However, it should be understood that there is no intention to limit the various embodiments of the present disclosure to the specific embodiments disclosed herein, but the present disclosure should be understood to cover all adjustments, equivalents, and / or alternative solutions falling within the spirit and scope of the various embodiments of the present disclosure.

[0020] This embodiment provides a bulk material stacking method based on a stacker-reclaimer.

[0021] See Figure 1 , the figure shows that this method includes step S1, step S2, step S3, step S4, and step S5.

[0022] Step S1: Obtain the data information uploaded by the sensors in the area to be stacked and the bulk material distribution information. The data information uploaded by the sensors includes the environmental data information and the environmental image information of the area to be stacked. It can be understood that in the bulk material stacking operation, accurately obtaining the data information uploaded by the sensors in the area to be stacked is the key to ensuring the efficient execution of subsequent steps. The core objective of this step is to comprehensively collect data on the area to be stacked through different types of sensors to provide the necessary environmental information and bulk material distribution information, thereby providing reliable basic data for subsequent stacking path planning and motion control. Specifically, the data information uploaded by the sensors includes the environmental data information and the environmental image information of the area to be stacked. These data help to depict the physical state and real-time changes of the operation area. The environmental data information is usually collected by a combination of multiple sensors such as lidar, radar sensors, temperature and humidity sensors, etc., and can provide height information of the area, obstacle distribution, basic levels of material stacking, and other environmental factors that may affect the stacking operation (such as humidity, temperature, etc.). The collection of these environmental data information not only helps to identify the structural characteristics of the stacking area but also provides accurate scene layout data for subsequent path planning. The environmental image information is usually captured by a high-resolution camera or other image sensors and can provide visual images of the area to help determine the distribution of bulk materials, dynamic changes in stacking, and possible risk points or obstacles.

[0023] By integrating and fusing this information, the system can perform real-time perception of the area to be stacked and capture the distribution, shape, and current stacking state of the bulk materials in the area. This comprehensive data collection method can effectively address the uncertainty problems caused by environmental complexity and incomplete data in traditional technologies and provide accurate input data for subsequent bulk material stacking path planning and stacker-reclaimer motion control. The technical effect of this step is to provide a comprehensive and accurate real-time data foundation, laying the foundation for the application of intelligent algorithms and the optimization of bulk material stacking operations.

[0024] Step S2: Send the data information uploaded by the sensors in the area to be stacked to the preprocessing model for preprocessing to obtain the three-dimensional space data information of the area to be stacked. It can be understood that through data preprocessing in this step, the accuracy and applicability of the data collected by the sensors are significantly improved, providing high-quality input data for subsequent bulk material stacking path planning and stacker-reclaimer control systems. This not only enables the system to handle complex operating environments but also improves the accuracy and efficiency of the entire stacking process, thereby reducing resource waste and operation errors caused by data errors and providing guarantee for the stable operation of the intelligent stacking system. In this step, Step S2 includes Step S21, Step S22, Step S23, and Step S24.

[0025] Step S21: Denoise the data information uploaded by the sensors in the to-be-stacked area based on a preset mean-shift clustering algorithm to obtain the smoothed historical material height and historical position data of the to-be-stacked area; It can be understood that the mean-shift algorithm eliminates the influence of random noise by performing a weighted average on the neighborhood around each data point. This process makes the local changes in the data smoother and avoids the interference of short-term fluctuations on the overall trend of the data. Through the clustering method, the algorithm identifies and retains the data points in the areas with higher density, while removing the outlier data points with lower density. These outlier points are often caused by sensor errors or external environmental interference. Therefore, removing them helps to improve the accuracy of the data. The data after mean-shift processing is smoother and more uniform, and can reflect the historical stacking position and height of the material. While maintaining the characteristics of the bulk material stacking area, these data eliminate the unnecessary fluctuations and errors caused by noise, providing more accurate material distribution data for the subsequent steps.

[0026] Step S22: Merge the environmental image information of the to-be-stacked area based on a super-resolution reconstruction algorithm to obtain the high-resolution environmental image information of the to-be-stacked area; It can be understood that in this step, by aligning the environmental images of the to-be-stacked area, ensuring that they can be fused in the same coordinate system, and then by fusing the information of multiple low-resolution images, the super-resolution algorithm can extract the unique detailed information in each image, avoiding the interference of duplicate information, thereby enhancing the detailed performance of the image. Finally, the system generates an image with a higher pixel density. This high-resolution image can display more details, especially in aspects such as the distribution and stacking form of the material, helping the stacking planning system to conduct more accurate analysis. Furthermore, the environmental perception problems that may originally be caused by low-resolution images are effectively solved, making the subsequent bulk material stacking planning and equipment control more accurate. The high-resolution image provides richer data for the intelligent algorithm, further enhancing the intelligent level and operation reliability of the entire system.

[0027] Step S23: Perform a weighted regression operation based on the smoothed historical material height and historical position data of the to-be-stacked area to determine the stacking height and position distribution information of the material in the to-be-stacked area; It can be understood that in this step, different weights need to be set according to the historical reliability of the data or the difference in spatial position. For example, the material data near the sensor position can be given a higher weight due to higher acquisition accuracy, while the data farther from the sensor or more affected by environmental factors can be given a lower weight. The setting of the weight can be dynamically adjusted through a predetermined rule or based on the statistical characteristics of the data (such as material density, sensor accuracy, etc.). In weighted regression, by taking the historical material height and position data as independent variables, the goal is to establish a mathematical model of material stacking based on these data. The regression model not only needs to fit the height and position of the material, but also be able to handle data fluctuations and uncertainties at different positions within the stacking area. Then, the weighted least squares method (WLS) is used for regression analysis to ensure different degrees of influence are assigned to different data points, so that the prediction of the stacking height and position is closer to the actual situation. Finally, after weighted regression, the stacking height and position distribution information of the material within the area to be stacked can be obtained. Among them, the weighted regression model has strong adaptability and can update and optimize the prediction data in real time during the material stacking process.

[0028] Step S24: Based on the double integral weighted operation, process the environmental image information of the high-resolution area to be stacked, the stacking height and position distribution information of the material in the area to be stacked, and obtain the three-dimensional space data information of the area to be stacked.

[0029] It can be understood that in this step, the high-resolution environmental image information is fused with the material stacking height and position data in the area to be stacked. The environmental image is two-dimensional information, while the stacking height and position data are one-dimensional or two-dimensional material distribution information. Through the double integral weighted operation, the present invention will assign different weights to different information according to the height, position distribution of the material and the detailed information of the image, combined with the relative importance of each data point. The core of the double integral operation is to perform a second integral on the image data and the material stacking data in the area to be stacked. The image data needs to be converted into depth information related to the material stacking height. This is then completed by integrating the corresponding material height information at each pixel point in the image. Then, a second integral is further performed on this information, combined with the distribution characteristics of the material in space, to generate the complete three-dimensional space data.

[0030] Step S3: Send the three-dimensional space data information of the area to be stacked, the data information uploaded by the sensors in the area to be stacked, and the bulk material distribution information to a preset bulk material stacking planning model for processing, and obtain the stacking path of the bulk material and the motion control data of the stacker-reclaimer. It can be understood that this step effectively optimizes the overall process of bulk material stacking operations through intelligent path planning and control data generation, improves work efficiency, accuracy, and safety, and provides strong technical support for subsequent stacking operations. In this step, step S3 includes step S31, step S32, step S33, and step S34.

[0031] Step S31: Map the three-dimensional spatial data information of the area to be stacked, the data information uploaded by sensors in the area to be stacked, and the bulk material distribution information into a preset high-dimensional space, and process them through a fast dynamic random walk algorithm to obtain a preliminary path planning trajectory for bulk material stacking. It can be understood that in this step, the three-dimensional spatial data, sensor data, and bulk material distribution information of the area to be stacked are mapped into a high-dimensional space. The model can perform more detailed analysis in a larger data space, thereby capturing more features, such as the stacking methods of different materials, the non-uniformity of material distribution, and the impact of obstacles on the path. Then, multiple starting points are randomly selected in the high-dimensional space, representing different starting positions or paths during the stacking process. Each particle randomly walks in the space according to a preset step size rule. At the same time, each time the direction and distance of movement are selected, they are evaluated based on the cost of the current path (such as stacking difficulty, space constraints, etc.) and dynamically adjusted. Through multiple iterations, each particle will continuously adjust the step size and direction during the walking process and gradually approach the optimal path. During this process, the movement trajectory of the particles will reflect the most suitable path planning within the stacking area.

[0032] Among them, the movement of each particle in each step is represented by the following formula:

[0033] Among them, represents the position point in the area to be stacked, represents the step size, i represents the spatial dimension, represents the time step mark in the random walk algorithm, represents the position of the gradient, represents the position point at the

[0034] Step S32: Perform fractal geometry optimization processing based on the three-dimensional spatial data information of the area to be stacked and the preliminary path planning trajectory of bulk material stacking. Among them, select an optimized path planning trajectory for bulk material stacking by analyzing the complexity of the path and a preset simulated annealing algorithm. It can be understood that this step processes the preliminary path of bulk material stacking through fractal geometry optimization to improve the efficiency of path planning and the stacking quality. This step combines the complexity analysis of the path and the simulated annealing algorithm to optimize the preliminary path planning trajectory, and finally obtains a more efficient and stable stacking path.

[0035] Among them, fractal geometry is used to describe objects or paths with self-similarity and complex structures. In path planning, the application of fractal geometry mainly identifies and optimizes redundant, overly tortuous, or unnecessary parts by analyzing the complexity of the path. The complexity of the path can be measured by calculating the fractal dimension of the path. The fractal dimension is a measure of the "filling" ability of the path in space and is calculated through the following formula:

[0036] Among them, represents the fractal dimension of the path, represents the number of small segments at a distance of on the path, represents the subdivision scale of the path.

[0037] Among them, the simulated annealing algorithm is a global optimization algorithm that draws on the annealing process in physics and aims to find the global optimal solution within a large solution space through the combination of random search and local search. In path planning, the simulated annealing algorithm selects a preliminary path and sets a relatively high initial temperature. Based on the current path, the path is randomly perturbed, that is, the positions of some points on the path or the curvature of the path are randomly adjusted. For each perturbed path, its "energy" value is calculated. Among them, the energy function of the path is as follows:

[0038] Among them, represents the energy function of the path, represents the length of the path, represents the complexity of the path, usually determined by the fractal dimension. represents the collision risk between the path and obstacles or the already stacked materials, , , represents the weighting coefficient.

[0039] Step S33: Select a motion control strategy based on the preset multi-armed bandit algorithm for the optimized planning trajectory of bulk material stacking and the parameter information of the preset stacker-reclaimer. Among them, by using the preset historical motion control strategy as a control parameter, the motion control data of the stacker-reclaimer is obtained; It can be understood that in this step, an effective motion control strategy is selected for the optimized bulk material stacking planning trajectory and the parameter information of the stacker-reclaimer through a preset multi-armed bandit algorithm. The key to this process lies in balancing the exploration and exploitation of different control strategies to obtain the optimal motion control scheme. During the execution of real-time tasks, the algorithm will adjust the strategy selection based on new data. For example, if the selected control strategy (i.e., the motion control strategy of the stacker-reclaimer) performs worse than expected in the new working environment, the algorithm will update the strategy based on the evaluation results and select the next optimal strategy. Through such an iterative optimization process, the stacker-reclaimer can select the most appropriate motion control strategy in each task cycle to ensure the efficient and stable execution of the bulk material stacking task.

[0040] In this step, the formula of the multi-armed bandit algorithm is as follows:

[0041] Where, represents the control strategy selected at time step k , represents the j th control strategy, represents the average reward of control strategy at time step k , represents the number of times strategy is selected, k represents the time step.

[0042] Step S34: Perform real-time operations based on the optimized bulk material stacking path planning trajectory and the motion control data of the stacker-reclaimer, and update the bulk material stacking path planning trajectory and the motion control data of the stacker-reclaimer in real time based on the real-time operation feedback and the adaptive control algorithm to obtain the real-time updated bulk material stacking path planning trajectory and the motion control data of the stacker-reclaimer.

[0043] It can be understood that adaptive control can dynamically adjust the path of the stacker-reclaimer according to real-time feedback during the execution process, ensuring that the stacker-reclaimer can adapt to environmental changes, reducing the deviation caused by path errors, and ensuring the accuracy of the bulk material stacking shape and height. By real-time adjusting the motion control data of the stacker-reclaimer, the algorithm can reduce motion errors and ensure the precise execution of each action, thereby improving work efficiency and task completion rate. In this step, step S34 includes step S341, step S342, step S343, and step S344.

[0044] Step S341: Based on the data information of the real-time operation feedback and the particle swarm optimization algorithm, perform real-time adjustment on the stacking path to obtain the adjusted stacking path information; It can be understood that the particle swarm optimization algorithm in this step is an optimization algorithm that simulates the foraging behavior of a bird flock. Its basic idea is to search the solution space through a group of "particles" to find the optimal solution to the problem. In this step, the particle swarm optimization algorithm will be used to optimize the stacking path. By simulating the search process of multiple particles, the selection of the path is continuously updated, so as to find the optimal stacking path that adapts to environmental changes and real-time feedback. Through the particle swarm optimization algorithm, the stacking path will be optimized based on real-time feedback. The optimized path will reflect the actual execution situation, environmental changes, and control requirements, so as to provide updated path information for the stacker-reclaimer, enabling it to better adapt to on-site changes and improve execution accuracy.

[0045] Step S342: Based on the adaptive control algorithm, perform real-time adjustment on the data information of the real-time operation feedback and the motion control data of the stacker-reclaimer to obtain the adjusted motion control data of the stacker-reclaimer; It can be understood that the adaptive control algorithm in this step is an algorithm that can automatically adjust the controller parameters according to the dynamic changes of the system. In this step, adaptive control is used to adjust the control input of the stacker-reclaimer according to the real-time feedback data to correct the possible errors in the execution process of the system and improve the robustness and stability of the system. The adaptive control adjustment process is as follows: First, the algorithm calculates the error between the current state and the target state of the stacker-reclaimer. Based on the error, the adaptive control algorithm adjusts the control input. Among them, the adjustment formula of the adaptive algorithm is as follows:

[0046] Among them, is the control input (i.e., the adjusted motion control data), including the speed, position, and acceleration control of the stacker-reclaimer; is the adaptive gain, representing the control gain that is adjusted over time and with the change of the error; represents the differential gain, represents the error between the current state and the target state of the stacker-reclaimer, represents the iteration number.

[0047] Finally, the adaptive control algorithm will dynamically adjust the gain and optimize the control strategy according to the real-time feedback of the system. These gain values will be self-adjusted according to the change of the current error, so that the control system can still operate efficiently when the environment changes or the system model is uncertain. With each iteration adjustment, the motion control data of the stacker-reclaimer will be updated in real time. This control data will guide the stacker-reclaimer to make corresponding actions, such as adjusting the motion trajectory, speed, acceleration, etc., to ensure that the stacker-reclaimer can accurately execute the path and handle any unexpected environmental changes during the bulk material stacking process.

[0048] Step S343: Conduct a safety risk assessment based on the adjusted stacking path information and the movement control data of the stacker-reclaimer to obtain risk assessment data for bulk material stacking; It can be understood that in this step, based on the calculation results of the Bayesian network, the obtained risk assessment data reflects the potential safety risks that may occur during the bulk material stacking process under the current stacking path and the control of the stacker-reclaimer. These assessment results can be used in a decision support system to further adjust the movement strategy of the stacker-reclaimer. For example, if the risk assessment value is high, it may be necessary to re-plan the stacking path, adjust the movement speed of the stacker-reclaimer, or take additional safety measures. In this step, step S343 includes step S3431 and step S3432.

[0049] Step S3431: Construct a Bayesian network based on the adjusted stacking path information and the movement control data of the stacker-reclaimer. Among them, use the adjusted stacking path information and the movement control data of the stacker-reclaimer as the nodes of the Bayesian network, and use the historical impact data of the preset stacking risk as the dependency relationship between the nodes to construct a directed acyclic graph structure, and then obtain a preliminary Bayesian network structure; It can be understood that in this step, first, based on the adjusted stacking path information and the movement control data of the stacker-reclaimer, a Bayesian network model is constructed to describe the various factors of the risk assessment and their mutual relationships. These nodes are connected through probabilistic dependency relationships, indicating how the various factors affect each other. For example, the complexity of the stacking path may directly affect the control difficulty of the stacker-reclaimer, thus increasing the operation risk. Among them, the directed acyclic graph structure of the Bayesian network can be represented in the following way:

[0050] Among them, is the conditional probability, indicating the probability of the overall risk given the path complexity and control data; represents the prior probability of the path risk, represents the prior probability of the movement control risk; represents the risk of the stacking path; represents the risk related to the movement control of the stacker-reclaimer, represents the safety risk during the overall bulk material stacking process, represents , and the probability of simultaneous occurrence.

[0051] Step S3432: Calculate the conditional probability of each node based on the preliminary Bayesian network structure, and use the Bayesian estimation method to calculate the probability values of each node under different conditions, and then obtain the risk probability values of bulk material stacking under different conditions.

[0052] It can be understood that the calculation formula for conditional probability in this step is as follows:

[0053] wherein, is the conditional probability, representing the probability of the overall risk given the path complexity and control data; represents and the probability of occurring simultaneously; represents 、 and the probability of occurring simultaneously.

[0054] Step S344: Based on the risk assessment data of bulk material stacking and a preset reinforcement learning algorithm, the adjusted stacking path information and the motion control data of the stacker-reclaimer are adjusted again to obtain the real-time updated bulk material stacking path planning trajectory and the motion control data of the stacker-reclaimer.

[0055] It can be understood that in this step, the current stacking path and the motion control data of the stacker-reclaimer are fed back through the risk assessment data to evaluate the safety, efficiency, and stability in the current state; then, the reinforcement learning algorithm (Q-learning algorithm) is used to adjust the stacking path and the motion control strategy according to the feedback, where the state space includes factors such as path complexity and machine motion state, and the action space includes path adjustment and control parameter changes; then, a reward function is defined, and the system is guided to select the optimal adjustment strategy through rewards or punishments to maximize long-term safety and efficiency; finally, the reinforcement learning algorithm continuously optimizes the path and control data, making the bulk material stacking process safer, more stable, and more efficient under changing environmental and operating conditions, and finally updating and optimizing the bulk material stacking path planning trajectory and the motion control data of the stacker-reclaimer in real time. The reward function is a comprehensive index obtained by weighted calculation of each risk assessment data.

[0056] Step S4: Send the stacking path of the bulk material and the motion control data of the stacker-reclaimer to a preset stacking shape planning model for processing to obtain the stacking shape data of the bulk material; It can be understood that in this step, by inputting the stacking path and control data into the stacking shape planning model and processing them, the final shape of the bulk material stacking can be accurately calculated. This process not only ensures the reasonable distribution of the stacked materials in space but also guarantees the stability of the stacked body, thereby reducing the risk of landslides or other unstable phenomena during the stacking process. In addition, the model optimizes the stacking angle and direction, improves the stacking efficiency and space utilization rate, and ensures that the final stacking result meets the functional requirements in practical applications. In this step, Step S4 includes Step S41, Step S42, and Step S43.

[0057] Step S41: Process the stacking path of the bulk material and the motion control data of the stacker-reclaimer based on a preset Poisson reconstruction algorithm to obtain the maximum stacking height and the shape boundary of the bulk material; It can be understood that the Poisson reconstruction algorithm in this step was initially used in 3D computer graphics. It is a method for reconstructing a 3D surface from point cloud data by solving the Poisson equation. In the bulk material stacking scenario, the application of the Poisson reconstruction algorithm is mainly to model the stacking path and the material distribution, so as to deduce the shape and size of the material in space. Based on the 3D surface model obtained by the Poisson reconstruction algorithm, the maximum stacking height and the shape boundary of the bulk material stack can be further calculated. These data are key indicators in the stacking process and help ensure the material stability and the rationality of the stack structure during the stacking process.

[0058] Step S42: Reduce the dimension of the bulk material distribution information and the stacking path of the bulk material based on a preset manifold learning method, and calculate the optimal stacking angle and direction through the stacking path and the bulk material distribution information of each bulk material; It can be understood that the manifold learning in this step is a class of nonlinear dimensionality reduction techniques, mainly used to discover the potential low-dimensional structure from high-dimensional data. In high-dimensional data, although the number of data points may be very large, its actual distribution may be confined to a low-dimensional manifold. The manifold learning method extracts the meaningful low-dimensional features through nonlinear mapping of the data, and then finds a more representative and effective structure in the complex high-dimensional data. The isometric feature mapping algorithm is adopted in this step. First, a distance matrix is established by calculating the geographical distance (i.e., the path length) between data points, and then these distances are processed by a global optimization algorithm to obtain a low-dimensional representation, so that the adjacency relationship of the data remains as stable as possible after dimensionality reduction. Through the analysis of the low-dimensional data, the isometric feature mapping algorithm can identify the possible best angles (i.e., the inclination angles of the material) and the suitable stacking directions (i.e., the orientation of the stacked material) during the stacking process, and combine the material distribution and path information to analyze the influence of different angles and directions on the stability of the stack. Through this process, unstable stacking methods can be avoided to ensure that the material will not shift or slide during the stacking process.

[0059] Step S43: Conduct a stability analysis based on the maximum stacking height, the maximum stacking shape boundary, the stacking angle, and the stacking direction of the bulk material to obtain the stability analysis of the stack. If the analysis result shows that the stability is greater than a preset threshold, then use the maximum stacking height, the maximum stacking shape boundary, the stacking angle, and the stacking direction of the bulk material as the stacking shape data of the bulk material.

[0060] It is understandable that this step conducts a detailed analysis of the stability during the bulk material stacking process to ensure that the generated stacking plan is safe and feasible in actual operation. This process is evaluated based on key parameters such as the maximum stacking height, the boundary of the maximum stacking shape, the stacking angle, and the stacking direction, and finally determines the stability of the bulk material stack. If the analysis result shows that the stability of the stack is greater than the preset threshold, then these parameters will be used as the final stacking shape data for the subsequent stacking process. The analysis method in this step is to judge the threshold through expert experience, such as giving expert scores to the maximum stacking height, the boundary of the maximum stacking shape, the stacking angle, and the stacking direction of the bulk material, and then calculating the score through weighted calculation. Finally, it is judged whether the score is greater than the preset threshold, and then whether it meets the requirements.

[0061] Step S5: Combine the stacking shape data of the bulk material, the stacking path of the bulk material, and the motion control data of the stacker-reclaimer to perform bulk material stacking, and optimize the motion control data of the stacker-reclaimer in real time based on the real-time obtained bulk material stacking result to obtain the final bulk material stacking plan.

[0062] It is understandable that the goal of the optimization control in this step is to correct the motion trajectory and control instructions of the stacker-reclaimer to the optimal state. This optimization process not only includes path correction but may also include adjustments to the working parameters of the stacker-reclaimer (such as grasping force, speed, lifting height, etc.). Based on the feedback of the real-time stacking result, the present invention can adjust the motion of the stacker-reclaimer to ensure that the stacking result can meet the predetermined shape and stability requirements.

[0063] Embodiment 2: As Figure 2 shown, this embodiment provides a bulk material stacking system based on a stacker-reclaimer. Refer to Figure 2 The system includes an acquisition unit 701, a first processing unit 702, a second processing unit 703, a third processing unit 704, and an optimization unit 705; The acquisition unit 701 is used to acquire the data information uploaded by the sensors in the area to be stacked and the bulk material distribution information. The data information uploaded by the sensors includes the environmental data information and the environmental image information of the area to be stacked; The first processing unit 702 is used to send the data information uploaded by the sensors in the area to be stacked to a preprocessing model for preprocessing to obtain the three-dimensional space data information of the area to be stacked; The first processing unit 702 includes: The first processing subunit is used to perform denoising processing on the data information uploaded by the sensors in the area to be stacked based on a preset mean shift clustering algorithm to obtain the historical material height and historical position data of the smoothed area to be stacked; A second processing subunit, configured to merge the environmental image information of the area to be stacked based on a super-resolution reconstruction algorithm to obtain high-resolution environmental image information of the area to be stacked; A third processing subunit, configured to perform a weighted regression operation based on the smoothed historical material height and historical position data of the area to be stacked to determine the stacking height and position distribution information of the materials in the area to be stacked; A fourth processing subunit, configured to process the high-resolution environmental image information of the area to be stacked, the stacking height and position distribution information of the materials in the area to be stacked based on a double integral weighting operation to obtain three-dimensional space data information of the area to be stacked; A second processing unit 703, configured to send the three-dimensional space data information of the area to be stacked, the data information uploaded by the sensors in the area to be stacked, and the bulk material distribution information to a preset stacking planning model of the bulk material for processing to obtain a stacking path of the bulk material and motion control data of a stacker-reclaimer; The second processing unit 703 includes: A fifth processing subunit, configured to map the three-dimensional space data information of the area to be stacked, the data information uploaded by the sensors in the area to be stacked, and the bulk material distribution information into a preset high-dimensional space and perform processing through a fast dynamic random walk algorithm to obtain a preliminary path planning trajectory for bulk material stacking; A sixth processing subunit, configured to perform fractal geometry optimization processing based on the three-dimensional space data information of the area to be stacked and the preliminary path planning trajectory for bulk material stacking, wherein an optimized path planning trajectory for bulk material stacking is selected by analyzing the complexity of the path and a preset simulated annealing algorithm; A seventh processing subunit, configured to select a motion control strategy based on a preset multi-armed bandit algorithm for the optimized path planning trajectory for bulk material stacking and preset parameter information of a stacker-reclaimer, wherein motion control data of the stacker-reclaimer is obtained by using a preset historical motion control strategy as a control parameter; An eighth processing subunit, configured to perform real-time operations based on the optimized path planning trajectory for bulk material stacking and the motion control data of the stacker-reclaimer, and perform real-time updates on the path planning trajectory for bulk material stacking and the motion control data of the stacker-reclaimer based on real-time operation feedback and an adaptive control algorithm to obtain real-time updated path planning trajectory for bulk material stacking and motion control data of the stacker-reclaimer; The eighth processing subunit includes: A first adjustment subunit, configured to perform real-time adjustment on the stacking path based on the data information of real-time operation feedback and a particle swarm optimization algorithm to obtain adjusted stacking path information; The second adjustment subunit is used to perform real-time adjustment on the data information of the real-time operation feedback and the motion control data of the stacker-reclaimer based on an adaptive control algorithm, and obtain the adjusted motion control data of the stacker-reclaimer; The second analysis subunit is used to perform a safety risk assessment based on the adjusted stacking path information and the motion control data of the stacker-reclaimer, and obtain the risk assessment data of the bulk material stacking; The third adjustment subunit is used to perform re-adjustment on the adjusted stacking path information and the motion control data of the stacker-reclaimer based on the risk assessment data of the bulk material stacking and a preset reinforcement learning algorithm, and obtain the real-time updated bulk material stacking path planning trajectory and the motion control data of the stacker-reclaimer; The third processing unit 704 is used to send the stacking path of the bulk material and the motion control data of the stacker-reclaimer to a preset stacking shape planning model for processing, and obtain the stacking shape data of the bulk material; The third processing unit 704 includes: The ninth processing subunit is used to process the stacking path of the bulk material and the motion control data of the stacker-reclaimer based on a preset Poisson reconstruction algorithm, and obtain the maximum stacking height and shape boundary of the bulk material; The first calculation subunit is used to perform dimensionality reduction on the bulk material distribution information and the stacking path of the bulk material based on a preset manifold learning method, and calculate the optimal stacking angle and direction through the stacking path of each bulk material and the bulk material distribution information; The first analysis subunit is used to perform a stability analysis based on the maximum stacking height, maximum stacking shape boundary, stacking angle and stacking direction of the bulk material, and obtain the stability analysis of the stacked body. If the analysis result shows that the stability is greater than a preset threshold, the maximum stacking height, maximum stacking shape boundary, stacking angle and stacking direction of the bulk material are used as the stacking shape data of the bulk material; The optimization unit 705 is used to combine the stacking shape data of the bulk material, the stacking path of the bulk material and the motion control data of the stacker-reclaimer to perform bulk material stacking, and perform real-time optimization on the motion control data of the stacker-reclaimer based on the real-time obtained bulk material stacking result, and obtain the final bulk material stacking scheme.

[0064] It can be understood that the systems, devices, modules or units described in the above embodiments can be specifically implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer, and the specific form of the computer can be a personal computer, a laptop computer, a personal digital assistant, a tablet computer, a wearable device, or a combination of any several of these devices.

[0065] In a typical configuration, a computer includes one or more processors (CPUs), an input / output interface, a network interface, and a memory.

[0066] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0067] Computer-readable media include permanent and non-permanent, removable and non-removable media that can be used to store information by any method or technology. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, disk storage, quantum memory, graphene-based storage media or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0068] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0069] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0070] The above describes specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the acts or steps recited in the claims may be performed in a different order than in the embodiments and still achieve the desired results. Additionally, the processes depicted in the figures do not necessarily require the particular order shown or sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0071] The terms used in one or more embodiments of this specification are for the purpose of describing particular embodiments only and are not intended to limit one or more embodiments of this specification. The singular forms "a", "the", and "said" used in one or more embodiments of this specification and the appended claims are also intended to include the plural forms unless the context clearly dictates otherwise. It should also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0072] It should be understood that although the terms first, second, third, etc. may be used in one or more embodiments of this specification to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, without departing from the scope of one or more embodiments of this specification, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "when" or "upon" or "in response to determining".

[0073] The above is only the preferred embodiment of one or more embodiments of this specification and is not intended to limit one or more embodiments of this specification. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of one or more embodiments of this specification shall be included within the scope of protection of one or more embodiments of this specification.

Claims

1. A bulk material stacking method based on a stacker-reclaimer, characterized in that: The following steps are involved: Step S1, acquiring data information and bulk material distribution information uploaded by sensors in the area to be deposited, wherein the data information uploaded by the sensors includes environmental data information and environmental image information of the area to be deposited; Step S2, sending the data information uploaded by the sensor of the area to be deposited to the preprocessing model for preprocessing to obtain the three-dimensional spatial data information of the area to be deposited; Step S3, sending the three-dimensional spatial data information of the area to be accumulated, the data information uploaded by the sensor of the area to be accumulated and the bulk material distribution information to a preset bulk material accumulation planning model for processing, so as to obtain the accumulation path of the bulk material and the motion control data of the stacker and reclaimer; Step S4, sending the bulk material stacking path and the motion control data of the stacker-reclaimer to a preset stacking shape planning model for processing to obtain the bulk material stacking shape data; Step S5: Bulk material is piled up in combination with the bulk material pile-up shape data, the bulk material pile-up path and the motion control data of the stacker-reclaimer, and the motion control data of the stacker-reclaimer is optimized in real time based on the bulk material pile-up result obtained in real time to obtain a final bulk material pile-up plan.

2. The bulk material stacking method based on a stacker-reclaimer according to claim 1, characterized in that: Step S2 includes the following steps: Step S21, performing denoising processing on the data information uploaded by the sensor of the area to be deposited based on a preset mean shift clustering algorithm to obtain smoothed historical material height and historical position data of the area to be deposited; Step S22, merging the environmental image information of the area to be deposited based on a super-resolution reconstruction algorithm to obtain high-resolution environmental image information of the area to be deposited; Step S23: performing a weighted regression operation based on the smoothed historical material height and historical position data of the area to be accumulated, to determine the accumulation height and position distribution information of the material in the area to be accumulated; Step S24: Based on the double integral weighted operation, the high-resolution environmental image information of the area to be deposited, the depositing height and position distribution information of the materials in the area to be deposited are processed to obtain the three-dimensional spatial data information of the area to be deposited.

3. The bulk material stacking method based on a stacker-reclaimer according to claim 1, characterized in that: Step S3 includes the following steps: Step S31, mapping the three-dimensional spatial data information of the area to be deposited, the data information uploaded by the sensors of the area to be deposited, and the bulk material distribution information into a preset high-dimensional space, and processing them through a fast dynamic random walk algorithm to obtain a preliminary path planning trajectory for bulk material depositing; Step S32: fractal geometry optimization processing is performed based on the three-dimensional spatial data information of the area to be deposited and the preliminary path planning trajectory of bulk material depositing, and an optimized bulk material depositing path planning trajectory is selected by analyzing the complexity of the path and a preset simulated annealing algorithm; Step S33, selecting a motion control strategy for the optimized planned trajectory of bulk material accumulation and the preset parameter information of the stacker-reclaimer based on the preset multi-armed bandit algorithm, and obtaining motion control data of the stacker-reclaimer by using the preset historical motion control strategy as a control parameter; Step S34: Perform real-time operations based on the optimized bulk material accumulation path planning trajectory and the motion control data of the stacker-reclaimer, and update the bulk material accumulation path planning trajectory and the motion control data of the stacker-reclaimer in real time based on real-time operation feedback and an adaptive control algorithm to obtain real-time updated bulk material accumulation path planning trajectory and the motion control data of the stacker-reclaimer.

4. The bulk material stacking method based on a stacker-reclaimer according to claim 1, characterized in that: Step S4 includes the following steps: Step S41, processing the bulk material stacking path and the motion control data of the stacker-reclaimer based on a preset Poisson reconstruction algorithm to obtain the maximum stacking height and shape boundary of the bulk material; Step S42: reducing the dimension of the bulk material distribution information and the bulk material stacking path based on a preset manifold learning method, and calculating the optimal stacking angle and direction through the stacking path and bulk material distribution information of each bulk material; Step S43, performing stability analysis based on the maximum stacking height, maximum stacking shape boundary, stacking angle and stacking direction of the bulk material to obtain a stability analysis of the stacking body; if the analysis result shows that the stability is greater than a preset threshold, the maximum stacking height, maximum stacking shape boundary, stacking angle and stacking direction of the bulk material are used as the stacking shape data of the bulk material.

5. The bulk material stacking method based on a stacker-reclaimer according to claim 3, characterized in that: Step S34 includes the following steps: Step S341, adjusting the stacking path in real time based on the data information of the real-time operation feedback and the particle swarm optimization algorithm to obtain adjusted stacking path information; Step S342: adjusting the data information of the real-time operation feedback and the motion control data of the stacker-reclaimer in real time based on the adaptive control algorithm to obtain the adjusted motion control data of the stacker-reclaimer; Step S343, performing safety risk assessment based on the adjusted stacking path information and the motion control data of the stacker-reclaimer to obtain risk assessment data for bulk material stacking; Step S344: Based on the risk assessment data of bulk material accumulation and the preset reinforcement learning algorithm, the adjusted accumulation path information and the motion control data of the stacker-reclaimer are adjusted again to obtain real-time updated bulk material accumulation path planning trajectory and motion control data of the stacker-reclaimer.

6. A bulk material stacking system based on a stacker-reclaimer, characterized in that: include: An acquisition unit, used to acquire data information and bulk material distribution information uploaded by a sensor of the area to be deposited, wherein the data information uploaded by the sensor includes environmental data information and environmental image information of the area to be deposited; A first processing unit, configured to send the data information uploaded by the sensor of the area to be deposited to a preprocessing model for preprocessing to obtain three-dimensional spatial data information of the area to be deposited; A second processing unit is used to send the three-dimensional spatial data information of the area to be accumulated, the data information uploaded by the sensor of the area to be accumulated and the bulk material distribution information to a preset bulk material accumulation planning model for processing, so as to obtain the bulk material accumulation path and the motion control data of the stacker and reclaimer; a third processing unit, configured to send the bulk material stacking path and the motion control data of the stacker-reclaimer to a preset stacking shape planning model for processing, so as to obtain the bulk material stacking shape data; The optimization unit is used to combine the bulk material stacking shape data, the bulk material stacking path and the motion control data of the stacker-reclaimer to stack the bulk materials, and optimize the motion control data of the stacker-reclaimer in real time based on the bulk material stacking results obtained in real time to obtain a final bulk material stacking solution.

7. The bulk material accumulation system based on a stacker-reclaimer according to claim 6, characterized in that: The first processing unit comprises: A first processing subunit is used to perform denoising on the data information uploaded by the sensor of the area to be deposited based on a preset mean shift clustering algorithm to obtain smoothed historical material height and historical position data of the area to be deposited; The second processing subunit is used to merge the environmental image information of the area to be deposited based on a super-resolution reconstruction algorithm to obtain high-resolution environmental image information of the area to be deposited; The third processing subunit is used to perform a weighted regression operation based on the smoothed historical material height and historical position data of the area to be accumulated, so as to determine the accumulation height and position distribution information of the materials in the area to be accumulated; The fourth processing subunit is used to process the high-resolution environmental image information of the area to be deposited, the stacking height and position distribution information of the materials in the area to be deposited based on a double integral weighted operation to obtain three-dimensional spatial data information of the area to be deposited.

8. The bulk material accumulation system based on a stacker-reclaimer according to claim 6, characterized in that: The second processing unit comprises: A fifth processing subunit is used to map the three-dimensional spatial data information of the area to be deposited, the data information uploaded by the sensor of the area to be deposited, and the bulk material distribution information into a preset high-dimensional space, and process them through a fast dynamic random walk algorithm to obtain a preliminary path planning trajectory for bulk material depositing; The sixth processing subunit is used to perform fractal geometry optimization processing based on the three-dimensional spatial data information of the area to be deposited and the preliminary path planning trajectory of bulk material depositing, wherein an optimized bulk material depositing path planning trajectory is selected by analyzing the complexity of the path and a preset simulated annealing algorithm; a seventh processing subunit, configured to select a motion control strategy for the optimized planned trajectory of bulk material accumulation and the preset parameter information of the stacker-reclaimer based on a preset multi-armed bandit algorithm, wherein the motion control data of the stacker-reclaimer is obtained by using the preset historical motion control strategy as a control parameter; The eighth processing subunit is used to perform real-time operations based on the optimized bulk material accumulation path planning trajectory and the motion control data of the stacker-reclaimer, and to update the bulk material accumulation path planning trajectory and the motion control data of the stacker-reclaimer in real time based on real-time operation feedback and an adaptive control algorithm to obtain real-time updated bulk material accumulation path planning trajectory and the motion control data of the stacker-reclaimer.

9. The bulk material accumulation system based on a stacker-reclaimer according to claim 6, characterized in that: The third processing unit comprises: A ninth processing subunit, for processing the bulk material stacking path and the motion control data of the stacker-reclaimer based on a preset Poisson reconstruction algorithm to obtain the maximum stacking height and shape boundary of the bulk material; A first calculation subunit is used to reduce the dimension of the bulk material distribution information and the bulk material stacking path based on a preset manifold learning method, and calculate the optimal stacking angle and direction through the stacking path and bulk material distribution information of each bulk material; The first analysis subunit is used to perform stability analysis based on the maximum stacking height, maximum stacking shape boundary, stacking angle and stacking direction of the bulk material to obtain a stability analysis of the stacking body. If the analysis result is that the stability is greater than a preset threshold, the maximum stacking height, maximum stacking shape boundary, stacking angle and stacking direction of the bulk material are used as the stacking shape data of the bulk material.

10. The bulk material accumulation system based on a stacker-reclaimer according to claim 8, characterized in that: The eighth processing subunit comprises: A first adjustment subunit is used to adjust the stacking path in real time based on data information of real-time operation feedback and a particle swarm optimization algorithm to obtain adjusted stacking path information; The second adjustment subunit is used to adjust the data information of the real-time operation feedback and the motion control data of the stacker-reclaimer in real time based on the adaptive control algorithm to obtain the adjusted motion control data of the stacker-reclaimer; The second analysis subunit is used to perform safety risk assessment based on the adjusted stacking path information and the motion control data of the stacker-reclaimer to obtain risk assessment data for bulk material stacking; The third adjustment subunit is used to readjust the adjusted stacking path information and the motion control data of the stacker-reclaimer based on the risk assessment data of the bulk material stacking and the preset reinforcement learning algorithm to obtain the real-time updated bulk material stacking path planning trajectory and the motion control data of the stacker-reclaimer.

Images