Automatic stacking and reclaiming method and system of a stacker-reclaimer

Abstract

The application provides a kind of automatic stacker-reclaimer method and system thereof, relating to engineering control technical field, including obtaining multi-period, multi-angle stockpile environment information;After pre-processing the stockpile environment information, static geometric features and dynamic change features are extracted through the processed stockpile environment information to construct a dynamic three-dimensional model;Feature extraction is performed on the real-time mining data of the dynamic three-dimensional model, and a change law prediction model is constructed through BP neural network;Based on the adaptive ant algorithm and the change law prediction model, the dynamic three-dimensional model is planned for material taking path, the running angle of the stacker-reclaimer is adjusted in combination with the material taking motion constraint to obtain the walking path;Automatic simulation of material taking is carried out according to the walking path, and the automatic stacking and taking scheme is obtained by comparing and analyzing the simulation results and the preset performance indicators. The problem that the stacker-reclaimer is difficult to recognize and effectively grab the material according to the accurate path, causing part of the material to be missed or the material to be taken at all is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of engineering control technology, in particular to an automatic stacking and reclaiming method and system of a stacker-reclaimer. BACKGROUND

[0002] In the field of engineering control technology, a stacker-reclaimer is usually used for loading and unloading bulk materials in ports, mines and power plants. During the operation, due to the influence of material properties (such as humidity, particle size, etc.), weather changes and other unpredictable factors, the shape and distribution of the stack will change dynamically. The traditional stacker-reclaimer shows certain limitations when dealing with dynamic environments, mainly reflected in the inability to optimize its own operation process according to the changes in real-time working conditions. For example, when encountering bad weather or the nature of the material changes, the stack may appear irregular accumulation or hardening, causing the stacker-reclaimer to be difficult to identify and effectively grab the material according to the accurate path, resulting in the problem of missing part of the material or being unable to reclaim the material at all. Therefore, there is an urgent need for an automatic stacking and reclaiming method and system of a stacker-reclaimer to solve the problem of the stacker-reclaimer being difficult to identify and effectively grab the material according to the accurate path, resulting in the problem of missing part of the material or being unable to reclaim the material at all. SUMMARY

[0003] The purpose of the present application is to provide an automatic stacking and reclaiming method and system of a stacker-reclaimer to improve the above problems. In order to achieve the above purpose, the technical solution adopted by the present application is as follows:

[0004] In a first aspect, the present application provides an automatic stacking and reclaiming method of a stacker-reclaimer, comprising:

[0005] obtaining multi-period and multi-angle stack environment information;

[0006] After preprocessing the stack environment information, static geometric features and dynamic change features are extracted from the processed stack environment information to construct a dynamic three-dimensional model;

[0007] Features are extracted from real-time excavation data of the dynamic three-dimensional model, and a change law prediction model is constructed through a BP neural network;

[0008] Based on the adaptive ant algorithm and the change law prediction model, the dynamic three-dimensional model is planned for a reclaiming path, the running angle of the stacker-reclaimer is adjusted in combination with the reclaiming motion constraint, and a walking path is obtained;

[0009] According to the walking path, an automatic simulation of reclaiming is carried out, and through comparison and analysis of the simulation results and the preset performance indicators, an automatic stacking and reclaiming scheme is obtained.

[0010] In a second aspect, the present application further provides an automatic stacking and reclaiming system of a stacker-reclaimer, comprising:

[0011] an acquisition module configured to acquire multi-period and multi-angle stockpile environment information;

[0012] an extraction module configured to extract static geometric features and dynamic change features from the pre-processed stockpile environment information to construct a dynamic three-dimensional model;

[0013] a construction module configured to extract features from real-time mining data of the dynamic three-dimensional model and construct a change law prediction model through a BP neural network;

[0014] a planning module configured to plan a material taking path for the dynamic three-dimensional model based on an adaptive ant algorithm and the change law prediction model, adjust a running angle of the material taking machine in combination with material taking movement constraints, and obtain a walking path;

[0015] an analysis module configured to simulate material taking automation according to the walking path, compare and analyze simulation results and preset performance indicators, and obtain an automatic stockpile and material taking scheme.

[0016] In a third aspect, the present application further provides an automatic stockpile and material taking equipment of a stockpile and material taking machine, which comprises:

[0017] a memory configured to store a computer program;

[0018] a processor configured to execute the computer program to implement the steps of the automatic stockpile and material taking method of the stockpile and material taking machine.

[0019] In a fourth aspect, the present application further provides a medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement the steps of the automatic stockpile and material taking method based on the stockpile and material taking machine.

[0020] The present application has the following beneficial effects:

[0021] The present application extracts static geometric features and dynamic change features from stockpile environment information to construct a dynamic three-dimensional model. The dynamic three-dimensional model can intuitively and accurately present the shape, structure and changes over time of the stockpile, providing strong support for subsequent analysis and planning. Further, the present application uses a BP neural network to construct a change law prediction model to accurately predict the change trend of the stockpile, thereby planning and adjusting the material taking strategy in advance to improve the material taking efficiency and accuracy. The adaptive ant algorithm and the change law prediction model are used to plan a material taking path for the dynamic three-dimensional model to obtain an optimal walking path, effectively avoiding collisions and reducing energy consumption, thereby significantly improving the working efficiency of the material taking machine. The present technical solution solves the problem that the existing stockpile and material taking machine is difficult to identify and effectively grasp materials according to an accurate path, avoiding the situation that part of the materials are missed or cannot be taken.

[0022] Other features and advantages of the present application will be set forth in the descriptions that follow, and in part will be apparent from the description, or can be learned by practice of the application. The purposes and other advantages of the present application will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments will be briefly introduced as follows. It should be understood that the following drawings only show some of the embodiments of the present application, and therefore should not be considered as a limitation to the scope. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0024] Figure 1 Flow chart of the automatic stacking and reclaiming method of the stacker-reclaimer described in the embodiments of the present application;

[0025] Figure 2 Structural schematic diagram of the automatic stacking and reclaiming equipment of the stacker-reclaimer described in the embodiments of the present application.

[0026] Markings in the drawings: 800, automatic stacking and reclaiming equipment of the stacker-reclaimer; 801, processor; 802, memory; 803, multimedia component; 804, I / O interface; 805, communication component. DETAILED DESCRIPTION

[0027] In order to make the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions of the embodiments of the present application will be described clearly and completely in the following with reference to the drawings of the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, but not all the embodiments. The components of the embodiments of the present application described and shown in the drawings can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. All other embodiments obtained by those skilled in the art without creative labor on the basis of the embodiments of the present application belong to the scope of protection of the present application.

[0028] It should be noted that: similar labels and letters represent similar items in the following drawings, therefore, once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. Meanwhile, in the description of the present application, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0029] Embodiment 1:

[0030] The embodiment provides an automatic stacking and reclaiming method of a stacker-reclaimer.

[0031] Referring to Figure 1 , the method comprises steps S1 to S5, including:

[0032] S1: obtaining multi-period and multi-angle stockpile environment information;

[0033] In this step, the stockpile environment information obtained from different times and different angles by a laser scanner, a camera and other devices is obtained.

[0034] S2: after preprocessing the stockpile environment information, static geometric features and dynamic change features are extracted from the processed stockpile environment information to construct a dynamic three-dimensional model;

[0035] To clarify the specific acquisition method of the dynamic three-dimensional model, steps S21 to S24 are included in step S2, specifically including:

[0036] S21: the multi-period and multi-angle stockpile environment information is respectively subjected to point cloud and image processing and spatial alignment to obtain point cloud data and image data;

[0037] In this step, the multi-period and multi-angle stockpile environment information is subjected to filtering, noise reduction and down-sampling processing or grayscale, edge detection and other preprocessing, and the preprocessed multi-period and multi-angle stockpile environment information is aligned to the same spatial coordinate system.

[0038] S22: the point cloud data is segmented according to a region growing algorithm to extract static geometric features;

[0039] In this step, the point cloud data is segmented into regions with similar geometric features according to the region growing algorithm, and the regions with similar geometric features are extracted to obtain the volume, surface area, shape and other static geometric features of the stockpile.

[0040] S23: time series analysis is performed based on the point cloud data and the image data, and the motion trajectory of the stockpile at different times is calculated in combination with an optical flow method to obtain dynamic change features;

[0041] To clarify the specific acquisition method of the dynamic change features, steps S231 to S234 are included in step S23, specifically including:

[0042] S231: a three-dimensional model is constructed based on the point cloud data and the image data, the three-dimensional models at different times are compared to calculate a stockpile volume change rate and a shape change rate;

[0043] In this step, the stockpile volume change rate is:

[0044]

[0045] In the above formula (1), r V represents the volume change rate of the stockpile, V t represents the volume of the stockpile after time t, and V0 represents the initial volume;

[0046] The shape change rate of the stockpile is:

[0047]

[0048] In the above formula (2), r S represents the shape change rate of the stockpile, P t represents the position of the stockpile after time t, and P0 represents the initial position of the surface point;

[0049] Wherein, by comparing the three-dimensional models at different times, the volume change rate and the shape change rate of the stockpile are calculated to provide quantitative indicators for dynamic monitoring.

[0050] S232: Analyzing the image sequence of the stockpile at different times according to the optical flow method, and calculating the motion trajectory of the stockpile surface;

[0051] In this step, the image sequence of the stockpile at different times is analyzed according to the optical flow method, and the motion vector of the pixel point obtained by the optical flow method is combined with the position information of the stockpile in the image to determine the motion trajectory of each point on the surface of the stockpile.

[0052] Wherein, the optical flow method can capture the pixel motion trajectory on the surface of the stockpile, reflecting the dynamic change of the stockpile.

[0053] S233: Fusion based on the volume change rate, shape change rate and motion trajectory to form the dynamic characteristics of the stockpile;

[0054] In this step, the volume change rate, shape change rate and motion trajectory are fused to form the feature vector of the dynamic state of the stockpile.

[0055] F = [r V , r S , v x , v y ] (3)

[0056] In the above formula (3), F represents the feature vector of the dynamic state of the stockpile, r V represents the volume change rate of the stockpile, r S represents the shape change rate of the stockpile, v x and v y represent the motion velocity components extracted in the optical flow method;

[0057] Wherein, the step comprehensively describes the dynamic change of the stockpile by fusing multi-dimensional features, the stockpile volume change rate is used to measure the increasing or decreasing speed of the stockpile volume, and the stockpile shape change rate is used to describe the degree of change of the stockpile shape over time.

[0058] S234: State estimation and prediction of the stockpile dynamic characteristics according to the Kalman filter, to obtain dynamic change characteristics.

[0059] In this step, the Kalman filter is used to perform real-time state estimation and prediction of the stockpile dynamic characteristics, effectively reducing noise interference and improving the accuracy of dynamic characteristics.

[0060] S24: Fusion based on the static geometric characteristics and the dynamic change characteristics to obtain a dynamic three-dimensional model.

[0061] In this step, the volume, shape and other static geometric characteristics are fused with the motion trajectory, change rate and other dynamic change characteristics, and a dynamic three-dimensional model is constructed by a three-dimensional modeling algorithm based on the fused characteristics.

[0062] The feature fusion expression is:

[0063] M = w s .F s +w d .F d (4)

[0064] In the above formula (4), M represents the fused model, F s represents static geometric characteristics, F d represents dynamic change characteristics, w s and w d represent weight coefficients.

[0065] The dynamic three-dimensional model is used to reflect the change of the stockpile state in real time, and the dynamic three-dimensional model is updated in real time to reflect the static and dynamic information of the stockpile.

[0066] S3: Feature extraction of real-time mining data of the dynamic three-dimensional model, and construction of a change law prediction model through a BP neural network.

[0067] To clearly define the specific acquisition method of the change law prediction model, step S3 includes S31 to S33, specifically:

[0068] S31: Preprocessing of real-time mining data of the dynamic three-dimensional model, analysis of the preprocessed data through the initial BP neural network, and extraction of key features of the stockpile.

[0069] In this step, an initial BP neural network is constructed, the number of nodes of the input layer, the hidden layer and the output layer is set, and the weights and the bias are randomly initialized. The preprocessed data is input into the initial BP neural network, the network output is calculated through forward propagation, and the key features of the stockpile are extracted.

[0070] The forward propagation expression of the BP neural network is:

[0071]

[0072] In the above formula (5), δ j represents the error of the jth neuron, represents the partial derivative, E represents the loss function, aj represents the activation value of the jth neuron, f'(z j represents the derivative of the activation function f with respect to the input z j , Δw ij represents the adjustment amount of the weight w ij , η represents the learning rate, δ j represents the error of the jth neuron, x i represents the ith input value input to the jth neuron.

[0073] The initial BP neural network is used to extract key features sensitive to dynamic changes of the stockpile. The preprocessing includes normalization, denoising and other processing to eliminate abnormal values and noise in the data.

[0074] S32: The key features of the stockpile are input into the BP neural network for training. Through multiple iterations to optimize the network parameters, a trained BP neural network is obtained.

[0075] In this step, the key features of the stockpile are input as input to train the BP neural network. During the training process, the error is calculated through forward propagation, and the network parameters are updated through back propagation. After multiple iterations of optimization, the trained BP neural network is obtained

[0076] During the training process, the training cannot be ended until the network error reaches the preset threshold or the training round reaches the set value. The trained BP neural network can better fit the change rule of the stockpile features.

[0077] S33: Based on the trained BP neural network, the change rule of the features of the dynamic three-dimensional model is predicted, and a change rule prediction model is obtained by modeling the real-time dynamic changes of the stockpile using a time series analysis method.

[0078] In this step, the trained BP neural network predicts the feature change patterns of the dynamic three-dimensional model, predicting the future feature changes of the stockpile, including volume change trends, shape evolution, etc.; a time series model is established by performing time series analysis on the real-time dynamic change data of the stockpile; then the prediction results of the BP neural network are fused with the analysis results of the time series model, taking into account the historical dynamic change data of the stockpile and the current feature information, to obtain the change pattern prediction model.

[0079] This step is used to plan and adjust the material handling strategy in advance, which solves the problem that a single model in the existing technology cannot fully capture the dynamic changes of the material pile.

[0080] S4: Based on the adaptive ant algorithm and the change pattern prediction model, the material picking path is planned for the dynamic three-dimensional model, and the running angle of the material picking machine is adjusted in combination with the material picking motion constraints to obtain the walking path;

[0081] In this step, the adaptive ant algorithm and the change pattern prediction model can prevent the material handling machine from colliding.

[0082] To clarify the specific method for obtaining the walking path, step S4 includes S41 to S44, specifically:

[0083] S41: Based on the ant colony algorithm, perform a multi-region search on the dynamic 3D model to obtain an initial path;

[0084] In this step, the ant algorithm is used to perform multi-region search, simulating the mechanism of ants releasing pheromones on the path. The probability expression of the ant's path selection is used to guide the ant colony to find the optimal solution and obtain the initial path.

[0085] The probability expression for the ant's path selection is:

[0086]

[0087] In equation (6) above, P ij τ represents the probability that an ant moves from point i to point j. ij η represents the pheromone concentration along path (i,j). ij Let α represent the control pheromone, β represent the parameters of the heuristic information weights, ∑k∈allowed represent the summation over all allowed points k, and τ represent the heuristic information weights. ik η represents the pheromone concentration along the path of an ant from point i to point k. ik This represents the heuristic information along the path of the ant from point i to point k;

[0088] S42: iteratively optimizing the initial path according to the ant algorithm and an online learning mechanism to obtain an optimized path;

[0089] To make the specific acquisition method of the optimized path clear, step S42 includes S421 to S423, specifically:

[0090] S421: performing quality evaluation on the initial path according to the performance indicators of the initial path by using the adaptive ant algorithm to obtain an initial path score;

[0091] In this step, the initial path score expression is:

[0092] S = w1·L + w2·E + w3·T (7)

[0093] In the above formula (7), S represents the initial path score, L represents the path length, E represents the path energy consumption, T represents the path time, and w1, w2, and w3 represent the weights of the corresponding indicators.

[0094] The performance indicators of the initial path include path length, path smoothness, path, target task matching degree, time consumption, and / or energy consumption, etc.

[0095] S422: dynamically adjusting the pheromone distribution in the adaptive ant algorithm based on the online learning mechanism and the initial path score to obtain an updated pheromone distribution;

[0096] In this step, the pheromone update rule is:

[0097]

[0098] In the above formula (8), represents the pheromone concentration on path (i, j) at the t+1th iteration, and p represents the pheromone evaporation coefficient, represents the pheromone concentration on path (i, j) at the tth iteration, and Δτ ij represents the amount of newly added pheromone on path (i, j);

[0099] If the initial path score is high, it means that the path quality is good, so the concentration of pheromone on this path is appropriately increased, and in subsequent iterations, other ants are more likely to choose this path. Conversely, if the initial path score is low, it means that the path quality is poor, so the concentration of pheromone on this path is reduced to reduce the probability of being selected. By dynamically adjusting the pheromone distribution, the adaptability of the algorithm to environmental changes is enhanced, and the efficiency and effectiveness of path optimization are improved.

[0100] S423: obtaining an optimized path by multiple iterations based on the updated pheromone distribution.

[0101] In this step, based on the updated pheromone distribution, the ants start a new round of path search, and each ant will decide the direction of the next step according to the concentration of pheromone and heuristic information (such as the direction of the path, the distance to the target point, etc.) when selecting the path; through multiple iterations of optimization, the path is gradually improved, and finally the optimized path is obtained. The initial path is gradually improved through the iteration process to adapt to the dynamically changing environment and requirements.

[0102] S43: Calculate the running angle of the reclaimer according to the optimized path, generate an angle adjustment strategy in combination with the reclaimer motion constraint, and evaluate to obtain a path evaluation result;

[0103] To clarify the specific way of obtaining the path evaluation result, steps S431 to S433 are included in step S43, specifically:

[0104] S431: Calculate the running angle of the reclaimer at each path point according to the optimized path to obtain running angle data;

[0105] In this step, the running angle expression is:

[0106]

[0107] In the above formula (9), θ i represents the running angle of the reclaimer at the i-th path point, Δx i represents the displacement of the reclaimer in the x direction, and Δy i represents the displacement of the reclaimer in the y direction.

[0108] Wherein, by analyzing the angle between the tangent direction of the optimized path at this point and the target direction, the running angle of the reclaimer at each path point is calculated.

[0109] The tangent direction of the optimized path at this point is analyzed to calculate the running angle of the reclaimer at each path point. Further, the running angle of the reclaimer at the path point is determined by the angle between the tangent direction and the horizontal direction (x axis).

[0110] S432: Adjust the running angle of the reclaimer based on the running angle data and the reclaimer motion constraint to obtain an angle adjustment strategy;

[0111] In this step, the angle adjustment strategy expression is:

[0112] θ′ i = max(min(θ i , θ max ), θ min ) (10)

[0113] In the above formula (10), θ' represents the adjusted operating angle, θ i represents the adjusted operating angle, θ i represents the operating angle of the reclaimer at the i-th path point, θ max represents the maximum operating angle allowed by the reclaimer, θ min represents the minimum operating angle allowed by the reclaimer, θ

[0114] By limiting the maximum and minimum operating angle, it is ensured that the operating angle of the reclaimer at each path point is neither too large nor too small, thereby avoiding mechanical damage or operational errors due to the angle exceeding the allowed range; the reclaimer movement constraint includes the mechanical structure limitation of the reclaimer itself (maximum rotation angle, minimum rotation angle, rotation speed limitation, etc.) and the process requirement of the reclamation operation (stable angle range required during reclamation), and by ensuring that the operating angle of the reclaimer meets the movement constraint, the safety and reliability of the reclamation process are improved.

[0115] S433: Adjust the optimized path based on the angle adjustment strategy, and evaluate the adjusted path based on the online learning mechanism to obtain a path evaluation result.

[0116] In this step, the path evaluation result is:

[0117] S' = w1.L' + w2.E' + w3.T' (11)

[0118] In the above formula (11), S ′ represents the path evaluation result, L ′ represents the length of the adjusted path, E ′ represents the energy consumption of the adjusted path, T' represents the time of the adjusted path, and w1, w2, and w3 represent the weights of the corresponding indicators.

[0119] By dynamically evaluating the adjusted path through the online learning mechanism, the efficiency and effectiveness of path optimization are improved.

[0120] S44: Predict path dynamic changes based on the path evaluation result and the change law prediction model, and generate a walking path.

[0121] In this step, based on the path evaluation result, the change law prediction model predicts the future trend of change according to the historical dynamic change data and the current state of the stockpile, and generates a walking path that can adapt to future changes, thereby improving the flexibility and adaptability of reclamation.

[0122] S5: Perform automatic reclamation simulation according to the walking path, and compare and analyze the simulation results with the preset performance indicators to obtain an automatic stockpile reclamation scheme.

[0123] To determine the specific acquisition method of the automatic stacking and reclaiming scheme, step S5 includes S51 to S53, specifically:

[0124] S51: input the walking path into the automation control system, simulate the walking, rotating and pitching actions of the reclaimer, and generate simulation operation data;

[0125] In this step, the walking action simulates the movement of the reclaimer in the stockpile area, controls the moving direction and distance of the reclaimer according to the coordinate information of the walking path; the rotating action simulates the adjustment of the reclaimer's orientation at different positions to align with the reclaiming position or the target path direction; the pitching action simulates the up and down adjustment of the reclaimer's reclaiming arm or related components to adapt to different height materials and reclaiming operation requirements. The simulation operation can pre-act and evaluate the working state of the reclaimer before actual operation, find potential problems in advance, and reduce risks and errors in actual operation.

[0126] S52: compare and analyze the simulation operation data and the performance indicators to obtain a comparison result;

[0127] In this step, the differences between the simulation operation data and the performance indicators are found through comparison to obtain a comparison result, which can clearly show the gap between the reclaimer's running effect and the expected target when the reclaimer runs according to the current walking path, and provide a clear direction and basis for subsequent control strategy adjustment.

[0128] S53: adjust and optimize the automation control strategy of the reclaimer based on the comparison result to obtain an automatic stacking and reclaiming scheme.

[0129] In this step, the adjustment strategy includes modifying speed, path or operation sequence, etc.

[0130] Embodiment 2:

[0131] This embodiment provides an automatic stacking and reclaiming device for a reclaimer, which comprises:

[0132] An acquisition module is configured to acquire multi-period and multi-angle stockpile environment information;

[0133] An extraction module is configured to extract static geometric features and dynamic change features from the processed stockpile environment information to construct a dynamic three-dimensional model;

[0134] A construction module is configured to extract features from real-time mining data of the dynamic three-dimensional model and construct a change law prediction model through a BP neural network;

[0135] To determine the specific acquisition method of the construction module, specifically:

[0136] a first analysis unit, configured to preprocess real-time mining data of the dynamic three-dimensional model, analyze the preprocessed data by using an initial BP neural network, and extract key features of the stockpile;

[0137] a training unit, configured to input the key features of the stockpile into the BP neural network for training, optimize network parameters through multiple iterations, and obtain a trained BP neural network;

[0138] a first prediction unit, configured to predict a change rule of features of the dynamic three-dimensional model based on the trained BP neural network, and model real-time dynamic changes of the stockpile by using a time series analysis method, to obtain a change rule prediction model.

[0139] a planning module, configured to plan a material taking path of the dynamic three-dimensional model based on an adaptive ant algorithm and the change rule prediction model, adjust a running angle of the material taking machine in combination with a material taking motion constraint, and obtain a walking path.

[0140] To clearly define the specific acquisition manner of the planning module, the following is specifically provided:

[0141] a searching unit, configured to perform multi-region search on the dynamic three-dimensional model based on the ant algorithm, and obtain an initial path;

[0142] a first optimization unit, configured to iteratively optimize the initial path according to the ant algorithm and an online learning mechanism, and obtain an optimized path;

[0143] a calculation unit, configured to calculate the running angle of the material taking machine according to the optimized path, generate an angle adjustment strategy in combination with a material taking motion constraint, and evaluate the angle adjustment strategy, to obtain a path evaluation result;

[0144] a second prediction unit, configured to predict path dynamic changes based on the path evaluation result and the change rule prediction model, and generate a walking path.

[0145] an analysis module, configured to simulate material taking automation according to the walking path, compare and analyze simulation results and preset performance indicators, and obtain an automatic stockpile and material taking scheme.

[0146] To clearly define the specific acquisition manner of the analysis module, the following is specifically provided:

[0147] a simulation unit, configured to input the walking path into an automatic control system, simulate walking, rotating, and pitching actions of the material taking machine, and generate simulation running data;

[0148] a second analysis unit, configured to compare and analyze the simulation running data and the performance indicators, and obtain a comparison result;

[0149] A second optimization unit is configured to adjust and optimize the automation control strategy of the reclaimer based on the comparison result, to obtain an automatic stacking and reclaiming scheme.

[0150] It should be noted that the specific manner in which the various modules perform operations in the above-described apparatuses has been described in detail in the embodiments of the method, and will not be described in detail here.

[0151] Embodiment 3:

[0152] Corresponding to the above method embodiments, the present embodiment also provides a reclaimer automatic stacking and reclaiming device. The reclaimer automatic stacking and reclaiming device described below can be correspondingly referred to the reclaimer automatic stacking and reclaiming method described above.

[0153] Figure 2 is a block diagram of a reclaimer automatic stacking and reclaiming device 800 according to an exemplary embodiment. As shown, the reclaimer automatic stacking and reclaiming device 800 can include a processor 801, a memory 802. The reclaimer automatic stacking and reclaiming device 800 can also include one or more of a multimedia component 803, an I / O interface 804, and a communication component 805. Figure 2

[0154] ​The processor 801 is configured to control the overall operation of the stacker-reclaimer automatic stockpiling and reclaiming device 800 to complete all or part of the steps of the stacker-reclaimer automatic stockpiling and reclaiming method described above. The memory 802 is configured to store various types of data to support the operation of the stacker-reclaimer automatic stockpiling and reclaiming device 800. For example, the data can include instructions for any application or method operating on the stacker-reclaimer automatic stockpiling and reclaiming device 800, and application-related data, such as contact data, sent and received messages, pictures, audio, video, and the like. The memory 802 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk. The multimedia component 803 can include a screen and an audio component. The screen can be a touch screen, for example, and the audio component is configured to output and / or input audio signals. For example, the audio component can include a microphone configured to receive external audio signals. The received audio signals can be further stored in the memory 802 or transmitted through the communication component 805. The audio component also includes at least one speaker configured to output audio signals. The I / O interface 804 provides an interface between the processor 801 and other interface modules, which can be a keyboard, a mouse, a button, and the like. The buttons can be virtual buttons or physical buttons. The communication component 805 is configured to enable wired or wireless communication between the stacker-reclaimer automatic stockpiling and reclaiming device 800 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, near field communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 805 can include a Wi-Fi module, a Bluetooth module, an NFC module.

[0155] In an exemplary embodiment, the automated stacker-reclaimer device 800 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the automated stacker-reclaimer method described above.

[0156] Example 4:

[0157] Corresponding to the above method embodiments, this embodiment also provides a medium. The medium described below can be referred to in correspondence with the automatic stacking and reclaiming method of the stacker-reclaimer described above.

[0158] A medium storing a computer program, which, when executed by a processor, implements the steps of the automatic stacking and reclaiming method of the stacker-reclaimer described in the above method embodiments.

[0159] The medium can specifically be any medium capable of storing program code, such as a USB flash drive, external hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0160] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. 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.

[0161] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. An automatic stacking and reclaiming method for a stacker-reclaimer, characterized in that, include: Obtain multi-time period and multi-angle information on the stockpile environment; After preprocessing the stockpile environment information, a dynamic three-dimensional model is constructed by extracting static geometric features and dynamic change features from the processed stockpile environment information. The specific methods for obtaining the dynamic 3D model include: Point cloud and image processing were performed on the material pile environment information from multiple time periods and angles, and spatial alignment was performed to obtain point cloud data and image data. The point cloud data is segmented using a region growing algorithm to extract static geometric features; Time series analysis is performed based on the point cloud data and the image data, and the motion trajectory of the material pile at different times is calculated by combining the optical flow method to obtain dynamic change characteristics. A dynamic three-dimensional model is obtained by fusing the static geometric features and the dynamic change features. Feature extraction is performed on the real-time mining data of the dynamic 3D model, and a change pattern prediction model is constructed using a BP neural network; The specific methods for obtaining the change pattern prediction model include: The real-time mining data preprocessing of the dynamic three-dimensional model is performed by analyzing the preprocessed data through the initial BP neural network to extract key features of the stockpile; The key features of the stockpile are input into the BP neural network for training. The network parameters are optimized through multiple iterations to obtain the trained BP neural network. Based on the trained BP neural network, the feature change law of the dynamic three-dimensional model is predicted, and the real-time dynamic change of the material pile is modeled using time series analysis method to obtain the change law prediction model. Based on the adaptive ant algorithm and the change pattern prediction model, the material picking path is planned for the dynamic three-dimensional model, and the running angle of the material picking machine is adjusted in combination with the material picking motion constraints to obtain the walking path; The specific methods for obtaining the walking path include: Based on the adaptive ant colony algorithm, a multi-region search is performed on the dynamic 3D model to obtain an initial path; The initial path is iteratively optimized using the adaptive ant colony algorithm and online learning mechanism to obtain the optimized path. The operating angle of the material handling machine is calculated based on the optimized path, and an angle adjustment strategy is generated and evaluated in conjunction with the material handling motion constraints to obtain the path evaluation result. Based on the path evaluation results and the change pattern prediction model, the dynamic changes of the path are predicted, and a walking path is generated. The automated material handling process is simulated based on the described walking path. By comparing and analyzing the simulation results with preset performance indicators, an automated stacking and reclaiming scheme is obtained.

2. The automatic stacking and reclaiming method for a stacker-reclaimer according to claim 1, characterized in that, The operating angle of the reclaimer is calculated based on the optimized path, and an angle adjustment strategy is generated and evaluated in conjunction with the reclaiming motion constraints to obtain the path evaluation result, including: The running angle of the material reclaimer at each path point is calculated based on the optimized path to obtain running angle data; The operating angle of the material handling machine is adjusted based on the operating angle data and the material handling motion constraints to obtain an angle adjustment strategy; The optimized path is adjusted based on the angle adjustment strategy, and the adjusted path is evaluated in conjunction with the online learning mechanism to obtain the path evaluation result.

3. The automatic stacking and reclaiming method for a stacker-reclaimer according to claim 1, characterized in that, Based on the described walking path, an automated material handling simulation is performed. By comparing and analyzing the simulation results with preset performance indicators, an automated stacking and reclaiming scheme is obtained, including: The walking path is input into the automated control system to simulate the walking, rotating, and pitching movements of the material handling machine and generate simulated operation data. The comparison results are obtained by comparing and analyzing the simulated operation data with the performance indicators. Based on the comparison results, the automation control strategy of the material reclaimer is adjusted and optimized to obtain an automatic stacking and reclaiming scheme.

4. An automatic stacker-reclaimer system, using the automatic stacker-reclaimer method as described in any one of claims 1-3, characterized in that, include: The acquisition module is used to acquire material pile environmental information from multiple time periods and angles; The extraction module is used to preprocess the material pile environment information and then extract static geometric features and dynamic change features from the processed material pile environment information to construct a dynamic three-dimensional model. The module is used to extract features from the real-time mining data of the dynamic 3D model and to build a change pattern prediction model through a BP neural network. The planning module is used to plan the material handling path for the dynamic three-dimensional model based on the adaptive ant algorithm and the change pattern prediction model, and to adjust the running angle of the material handling machine in combination with the material handling motion constraints to obtain the walking path; The analysis module is used to simulate the automated material handling process based on the walking path, and to obtain an automated stacking and reclaiming scheme by comparing and analyzing the simulation results with preset performance indicators.

5. The automatic stacker-reclaimer system according to claim 4, characterized in that, The building module includes: The first analysis unit is used to preprocess the real-time mining data of the dynamic three-dimensional model, and to extract key features of the stockpile by analyzing the preprocessed data through the initial BP neural network. The training unit is used to input the key features of the stockpile into the BP neural network for training, and to optimize the network parameters through multiple iterations to obtain the trained BP neural network. The first prediction unit is used to predict the feature change law of the dynamic three-dimensional model based on the trained BP neural network, and to model the real-time dynamic changes of the material pile using time series analysis method to obtain the change law prediction model.

6. The automatic stacker-reclaimer system according to claim 4, characterized in that, The planning module includes: The search unit is used to perform multi-region search on the dynamic 3D model based on the adaptive ant algorithm to obtain an initial path; The first optimization unit is used to iteratively optimize the initial path according to the adaptive ant algorithm and the online learning mechanism to obtain an optimized path; The calculation unit is used to calculate the operating angle of the material reclaimer based on the optimized path, generate an angle adjustment strategy in combination with the material reclaiming motion constraints, and evaluate it to obtain the path evaluation result. The second prediction unit is used to predict the dynamic changes of the path based on the path evaluation results and the change pattern prediction model, and generate a walking path.

7. The automatic stacker-reclaimer system according to claim 4, characterized in that, The analysis module includes: The simulation unit is used to input the walking path into the automation control system, simulate the walking, rotating and pitching movements of the material handling machine, and generate simulation operation data. The second analysis unit is used to compare and analyze the simulated running data with the performance indicators to obtain comparison results; The second optimization unit is used to adjust and optimize the automatic control strategy of the material reclaimer based on the comparison results, so as to obtain an automatic stacking and reclaiming scheme.

Images