Palletizing optimization algorithm based on multi-layer palletizing of materials

Abstract

The present invention relates to the field of logistics management technology, and in particular to a palletizing optimization algorithm based on multi-layer palletizing of materials, comprising the following steps: S1, material information collection and database storage: collecting information of materials to be palletized; S2, palletizing model establishment and preliminary scheme generation: establishing a three-dimensional model of the material; S3, optimization algorithm calculation and stability verification: stability analysis of the generated palletizing scheme; S4, simulation verification and scheme optimization: optimizing the palletizing scheme through an iterative optimization algorithm; S5, adaptive deformation prediction and compensation algorithm: predicting the deformation of different materials during the palletizing process; S6, palletizing execution and real-time monitoring: real-time monitoring of abnormal conditions during the palletizing process; S7, result feedback and continuous improvement: recording data and results during the palletizing process. The present invention can obtain and adjust the deformation data of the material in real time during the palletizing process, thereby ensuring the stability and accuracy of the material palletizing.

Description

Technical Field

[0001] The invention relates to the technical field of logistics management, and in particular to a palletizing optimization algorithm based on multi-layer palletizing of materials. Background Art

[0002] With the development of modern logistics and manufacturing industries, material palletizing, as an important warehousing and transportation operation, has received more and more attention. The traditional palletizing process usually relies on manual operation, which is time-consuming, labor-intensive and inefficient. In order to solve this problem, automated palletizing equipment and systems have emerged. By utilizing robotics technology and advanced algorithms, automated and intelligent palletizing of materials has been realized. However, in actual applications, the diversity and complexity of materials make the palletizing process face many challenges, such as the variability of material size, weight, shape and material, as well as problems such as dumping, sliding and deformation that may occur during the palletizing process.

[0003] Although the existing palletizing technology has improved the palletizing efficiency and automation level to a certain extent, it still has some significant shortcomings and deficiencies. First, the existing technology is not comprehensive and accurate in collecting material information, and lacks precise measurement and recording of material characteristics (such as size, weight, shape and material), resulting in insufficient basic data for the palletizing solution. Secondly, the existing optimization algorithm has limitations in calculating the optimal palletizing path and stability verification, and it is difficult to effectively deal with complex material stacking situations, which easily leads to instability and inefficiency in the palletizing process. In addition, the existing technology lacks real-time prediction and compensation mechanism for material deformation during the palletizing process, which makes the palletizing structure prone to instability when the force changes, affecting the palletizing effect and safety.

[0004] The purpose of the present invention is to provide a palletizing optimization algorithm based on multi-layer palletizing of materials, which solves the deficiencies in the prior art and improves the stability, accuracy and efficiency of the palletizing process through comprehensive and accurate material information collection, optimization algorithm and stability verification, and adaptive deformation prediction and compensation mechanism. Summary of the invention

[0005] Based on the above objectives, the present invention provides a palletizing optimization algorithm based on multi-layer palletizing of materials.

[0006] The palletizing optimization algorithm based on multi-layer palletizing of materials includes the following steps:

[0007] S1, material information collection and database storage: collect information about the materials to be palletized, including size, weight, shape and material, through sensors and scanning equipment, and store the material information in the database;

[0008] S2, palletizing model establishment and preliminary plan generation: Based on the collected material information, a three-dimensional model of the material is established, and a preliminary palletizing plan is generated according to the preset palletizing rules, including the placement, sequence and number of stacking layers of the materials;

[0009] S3, optimization algorithm calculation and stability verification: Based on the preliminary palletizing plan, the optimization algorithm is used to calculate the best palletizing path, and the stability of the generated palletizing plan is analyzed;

[0010] S4, simulation verification and solution optimization: simulate the palletizing solution in a virtual environment to verify the feasibility and stability of the solution, record and analyze the simulation results, and optimize the palletizing solution through iterative optimization algorithm based on the simulation results;

[0011] S5, Adaptive deformation prediction and compensation algorithm: predicts the deformation of different materials during the palletizing process and compensates by adjusting the palletizing solution in real time, including:

[0012] S51, Material deformation characteristics analysis: Through physical simulation and historical material information analysis, the deformation characteristics of different materials during the palletizing process are identified;

[0013] S52, real-time deformation monitoring: During the palletizing process, sensors are used to monitor the deformation of materials in real time and collect deformation data;

[0014] S53, deformation prediction: Based on real-time monitoring deformation data and historical logistics information, a deformation prediction model is constructed to predict the deformation trend of materials;

[0015] S54, dynamic scheme adjustment: according to the predicted deformation trend, the stacking scheme is adjusted in real time to optimize the placement angle and position of the materials;

[0016] S55, deformation compensation execution: in the actual palletizing operation, the adjusted palletizing plan is executed;

[0017] S6, palletizing and real-time monitoring: input the final optimized palletizing solution into the palletizing robot or automation equipment, perform actual palletizing operations, and monitor abnormal conditions during the palletizing process in real time to make timely adjustments;

[0018] S7, Result Feedback and Continuous Improvement: Record the data and results of the palletizing process, form a palletizing report, and store the feedback information in the database.

[0019] Furthermore, the material information collection and database storage in S1 include:

[0020] S11, dimension collection: using a laser rangefinder or a 3D scanner to measure the external dimensions of the material and generate three-dimensional dimension data of the material;

[0021] S12, weight collection: measure the weight of the material through the weighing sensor and record the weight data;

[0022] S13, shape acquisition: using 3D scanning equipment to obtain the shape contour and surface features of the material;

[0023] S14, material collection: analyzing the material characteristics of the material through a material identification sensor (such as a spectrometer or an infrared sensor) and recording the material information;

[0024] S15, data storage: the collected size, weight, shape and material information is transmitted to the database through the data interface for storage.

[0025] Furthermore, the palletizing model establishment and preliminary solution generation in S2 include:

[0026] S21, 3D model establishment: based on the collected material size, shape and material, a 3D model of the material is established using 3D modeling software;

[0027] S22, palletizing rule setting: according to the characteristics of the materials and the preset palletizing rules (such as heavy objects at the bottom, light objects at the top, fragile objects in the middle, etc.), set the basic palletizing rules;

[0028] S23, generating a preliminary plan: generating a preliminary palletizing plan based on the three-dimensional model of the material and the set palletizing rules, including the placement of the material, the palletizing sequence, and the number of stacking layers;

[0029] S24, solution verification: preliminary verification of the generated preliminary palletizing solution.

[0030] Furthermore, the optimization algorithm calculation and stability check in S3 include:

[0031] S31, optimization algorithm calculation: based on the preliminary palletizing plan, the optimal palletizing path is calculated using a genetic algorithm;

[0032] S32, stability check: stability analysis of the generated palletizing solution, including:

[0033] S321, center of gravity calculation: Calculate the center of gravity of each layer of material. The calculation formula is:

[0034]

[0035] Among them, G x and G y are the x- and y-coordinates of the center of gravity, w i is the weight of the i-th material, x i and iare the x-coordinate and y-coordinate of the i-th material respectively;

[0036] S322, overturning analysis: Using the moment balance principle, calculate the overturning moment of each material after stacking. The calculation formula is:

[0037]

[0038] Among them, M t is the total overturning moment, d i is the distance from the i-th material to the supporting surface;

[0039] S323, sliding analysis: Calculate the friction between materials through friction analysis. The calculation formula is:

[0040] F f =μ×N;

[0041] Among them, F f is the friction force, μ is the friction coefficient, and N is the positive pressure.

[0042] Furthermore, the simulation verification and solution optimization in S4 include:

[0043] S41, simulation environment construction: construct a simulation model that matches the actual palletizing scene in the virtual environment, including three-dimensional models of palletizing equipment, materials and palletizing areas;

[0044] S42, Palletizing Scheme Simulation: Execute the initially generated palletizing scheme in the simulation environment, simulate the material placement and stacking process through the physical engine, monitor the material stress, deformation and stability in real time, and verify the feasibility and stability of the scheme;

[0045] S43, data recording: recording the simulation data generated during the simulation process, including the placement position of the material, the number of stacking layers, the stress condition, and the deformation condition;

[0046] S44, Application of iterative optimization algorithm: Based on the simulation results, the iterative optimization algorithm is used to optimize the palletizing scheme, including:

[0047] S441, fitness function update: adjust the weight coefficients (α, β, γ) in the fitness function according to the simulation results to reflect the problems found in the simulation and the optimization goals;

[0048] S442, plan adjustment: Generate a new palletizing plan using the particle swarm optimization algorithm, adjust the placement, sequence and number of stacking layers of the materials, and generate a new plan;

[0049] S443, multiple simulations and optimizations: Repeatedly perform simulation of the palletizing scheme, data recording, and iterative optimization algorithm application until the simulation results meet the preset stability, space utilization, and efficiency requirements;

[0050] S45, solution determination: Based on multiple simulations and optimizations, the final optimized palletizing solution is determined.

[0051] Furthermore, the material deformation characteristic analysis in S51 includes:

[0052] S511, physical simulation analysis: Use finite element analysis method to perform physical simulation on materials, simulating the stress and deformation of materials during the palletizing process, including:

[0053] Establish material model: Establish finite element model of material according to its size, shape and material;

[0054] Apply boundary conditions and loads: According to the actual palletizing environment, apply corresponding boundary conditions and loads to simulate the stress state of materials during the palletizing process;

[0055] Solve deformation: Use the finite element solver to calculate the deformation of the material under different stress conditions. The calculation formula is:

[0056] Ku=P;

[0057] Where K is the stiffness matrix, u is the node displacement vector, and P is the external force load vector;

[0058] S512, historical material information analysis: using historical material information to perform statistical analysis on the deformation characteristics of different materials, specifically including:

[0059] Data collection: Collect and organize material deformation data recorded during historical palletizing, including deformation, stress conditions and environmental conditions;

[0060] Data modeling: Based on historical material information, a statistical model of material deformation characteristics is established. The calculation formula is:

[0061] Δx=aP+bE+c;

[0062] Among them, Δx is the deformation, P is the external force load vector, E is the environmental conditions (such as temperature, humidity, etc.), a, b, c are constant coefficients;

[0063] Characteristic identification: The deformation characteristics of different materials under given force and environmental conditions are identified through cluster analysis to form a deformation characteristic library. The calculation formula is:

[0064]

[0065] Among them, J is the objective function (the sum of squares within the cluster), k is the number of clusters, x is the data point, and C j is the jth cluster, h j is the centroid of the jth cluster.

[0066] Furthermore, the real-time deformation monitoring in S52 includes:

[0067] S521, sensor deployment: deploy a variety of sensors on palletizing equipment and materials, including displacement sensors, strain gauges and pressure sensors, to monitor the deformation of materials during the palletizing process;

[0068] S522, deformation data collection: real-time measurement of material displacement changes through displacement sensors, monitoring of material strain through strain gauges, and monitoring of material surface stress through pressure sensors.

[0069] Furthermore, the deformation prediction model in S53 adopts an artificial neural network (ANN), and the artificial neural network (ANN) includes:

[0070] S531, input layer data preprocessing: normalize the input data, the calculation formula is:

[0071]

[0072] Among them, x i is the original input data, x i ′ is the normalized input data, μ x and σ x are the mean and standard deviation of the input data respectively;

[0073] S532, input layer: define the number of nodes in the input layer, determine the number of nodes in the input layer according to the characteristic dimensions of the monitored deformation data (such as displacement, strain and stress) and historical logistics information, and pass the processed input feature vector to the next layer, expressed as:

[0074] X=[x1′,x2′,...,x n ′];

[0075] Where X is the input feature vector;

[0076] S533, time series feature extraction: extract time series features through the convolution layer to capture the time series changes of deformation data. The calculation formula is:

[0077]

[0078] in, is the convolution output of the s-th scale, is the convolution kernel weight of the s-th scale, k s is the convolution kernel size of the sth scale, x t-i is the input at time ti, b (s) is bias;

[0079] S534, hidden layer structure design: By introducing adaptive activation functions and skip connections, the nonlinear expression ability and feature capture ability of the network are enhanced. The calculation formula is:

[0080]

[0081] in, is the activation value of the lth layer, W (l) is the weight matrix of the lth layer, b (l) is the bias of the lth layer, is the activation value of the first two layers;

[0082] The calculation formula of the adaptive activation function is:

[0083]

[0084] Among them, a and b are learnable parameters;

[0085] S535, Attention Mechanism: The attention mechanism is introduced in the convolutional layer. The attention mechanism adjusts the weight of the input feature by calculating the attention weight. The calculation formula is:

[0086]

[0087] Among them, Q t =W Q H t , K t =W K H t 、V t =W V H t are query, key and value matrices respectively, d k is the dimension of the key, softmax is the normalization function, A t Output for attention;

[0088] S536, loss function: define a weighted mean square error loss function and add regularization terms and attention mechanism loss terms to prevent overfitting. The calculation formula is:

[0089]

[0090] Among them, L is the loss function, N is the number of samples, and w i is the weight of the sample, is the predicted value, y i is the true value, λ is the regularization coefficient, ||W (l) || 2 is the L2 norm of the weight matrix of the first layer, β is the regularization coefficient of the attention mechanism, ||A (l) ||2 is the L2 norm of the attention matrix of the lth layer;

[0091] S537, output layer: Determine the number of output layer nodes according to actual needs, which is used to predict the deformation of the material. The calculation formula is:

[0092]

[0093] in, is the predicted output, W (o) is the weight matrix of the output layer, is the activation value of the last hidden layer, b (o) is the bias of the output layer.

[0094] Furthermore, the dynamic scheme adjustment in S54 includes:

[0095] S541, deformation trend prediction: Based on the artificial neural network model, the deformation trend of materials during the palletizing process is predicted to obtain the estimated deformation amount of each material;

[0096] S542, real-time adjustment of placement: dynamically adjust the placement of each material according to the predicted deformation trend, including:

[0097] Calculate the adjustment vector: According to the deformation prediction results, calculate the position adjustment vector Δp ​​of the material. The calculation formula is:

[0098] Δp=η×Δx;

[0099] Among them, Δp is the position adjustment vector, η is the adjustment coefficient, and Δx is the predicted deformation;

[0100] Update placement position: Update the original placement position p to the adjusted placement position p′. The calculation formula is:

[0101] p′=p+Δp;

[0102] S543, optimize the placement angle: optimize the material placement angle according to the deformation prediction result of the material, including:

[0103] Calculate the adjustment angle: According to the deformation prediction results, calculate the material placement angle adjustment Δθ. The calculation formula is:

[0104]

[0105] Among them, Δθ is the angle adjustment amount, ξ is the adjustment coefficient, is the predicted deformation angle change;

[0106] Update the placement angle: Update the original placement angle θ to the adjusted placement angle θ′. The calculation formula is:

[0107] θ′=θ+Δθ.

[0108] Furthermore, the palletizing and real-time monitoring in S6 include:

[0109] S61, input optimization plan: transmit the final optimized palletizing plan to the palletizing robot or automation equipment;

[0110] S62, executing the palletizing operation: the palletizing robot or automated equipment performs the actual palletizing operation according to the input optimization plan, and places the materials one by one according to the instructions in the optimization plan;

[0111] S63, real-time monitoring: During the palletizing process, sensors (such as cameras, laser rangefinders, strain gauges, etc.) installed on robots or automated equipment monitor each step of the palletizing and the status of the materials in real time;

[0112] S64, Abnormality Detection and Adjustment: Analyze the results of real-time monitoring, detect abnormal conditions during the palletizing process (such as material sliding, tilting, deformation, etc.), and make timely adjustments.

[0113] Beneficial effects of the present invention:

[0114] The present invention introduces a series of accurate material information collection and database storage technologies, utilizes laser rangefinders, 3D scanners, weighing sensors and material identification sensors to comprehensively collect material size, weight, shape and material information, and stores these data in a central database, thereby providing accurate and detailed material characteristic data for palletizing model establishment and optimization algorithms, significantly improving the accuracy, stability and efficiency of palletizing solutions, ensuring the reliability and feasibility of preliminary solutions, and providing a solid data foundation for subsequent optimization steps.

[0115] The present invention calculates the best stacking path through optimization algorithm calculation and stability verification, and performs structural stability analysis through finite element analysis and other methods to ensure that each layer of materials will not tip over or slide after being stacked. In the process of simulation verification and solution optimization, through multiple simulations and optimizations in a virtual environment, the final stacking solution generated not only has high feasibility and stability, but also can achieve the best effect in practical applications, thereby ensuring the comprehensive optimization of the stacking solution in terms of space utilization, stability and operating efficiency.

[0116] The present invention, through an artificial neural network model, accurately predicts the deformation of materials during the palletizing process, and compensates by adjusting the palletizing scheme in real time. Through real-time deformation monitoring, dynamic scheme adjustment and deformation compensation execution, the deformation data of materials can be acquired and adjusted in real time during the palletizing process, ensuring the stability and accuracy of material palletizing. In addition, by executing the palletizing and real-time monitoring steps, abnormal situations in the palletizing process can be identified and processed in real time, ensuring the safety and stability of the palletizing process, and the palletizing efficiency and quality are improved through dynamic adjustment, thereby significantly enhancing the automation and intelligence level of the palletizing process. BRIEF DESCRIPTION OF THE DRAWINGS

[0117] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0118] Figure 1 It is a schematic diagram of the optimization algorithm flow of an embodiment of the present invention;

[0119] Figure 2 Schematic diagram of the adaptive deformation prediction and compensation algorithm flow in an embodiment of the present invention. DETAILED DESCRIPTION

[0120] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0121] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0122] like Figure 1-Figure 2 As shown in the figure, the palletizing optimization algorithm based on multi-layer palletizing of materials includes the following steps:

[0123] S1, material information collection and database storage: collect information about the materials to be palletized, including size, weight, shape and material, through sensors and scanning equipment, and store the material information in the database;

[0124] S2, palletizing model establishment and preliminary plan generation: Based on the collected material information, a three-dimensional model of the material is established, and a preliminary palletizing plan is generated according to the preset palletizing rules, including the placement, sequence and number of stacking layers of the materials;

[0125] S3, optimization algorithm calculation and stability verification: Based on the preliminary palletizing plan, the optimization algorithm is used to calculate the best palletizing path, and the generated palletizing plan is analyzed for stability to ensure that each layer of materials will not fall or slide after being stacked, and the placement or order of the materials is adjusted if necessary;

[0126] S4, simulation verification and solution optimization: simulate the palletizing solution in a virtual environment to verify the feasibility and stability of the solution, and record and analyze the simulation results. Based on the simulation results, optimize the palletizing solution through iterative optimization algorithms to ensure the best space utilization and stability;

[0127] S5, Adaptive deformation prediction and compensation algorithm: predicts the deformation of different materials during the palletizing process and compensates by adjusting the palletizing solution in real time, including:

[0128] S51, Material deformation characteristics analysis: Through physical simulation and historical material information analysis, the deformation characteristics of different materials during the palletizing process are identified;

[0129] S52, real-time deformation monitoring: During the palletizing process, sensors are used to monitor the deformation of materials in real time and collect deformation data;

[0130] S53, deformation prediction: Based on real-time monitoring deformation data and historical logistics information, a deformation prediction model is constructed to predict the deformation trend of materials;

[0131] S54, dynamic scheme adjustment: according to the predicted deformation trend, the stacking scheme is adjusted in real time to optimize the placement angle and position of the materials to prevent the instability of the stacking structure caused by deformation;

[0132] S55, deformation compensation execution: in the actual palletizing operation, the adjusted palletizing plan is executed to ensure the stability and accuracy of material palletizing;

[0133] S6, palletizing and real-time monitoring: input the final optimized palletizing solution into the palletizing robot or automation equipment, perform actual palletizing operations, monitor abnormal conditions during the palletizing process in real time, and make timely adjustments to ensure the safety and stability of the palletizing process;

[0134] S7, Result Feedback and Continuous Improvement: Record the data and results of the palletizing process, form a palletizing report, and store the feedback information in the database for subsequent optimization and improvement to improve the effectiveness and reliability of future palletizing solutions;

[0135] Through the above steps, efficient, stable and precise material palletizing is achieved. The optimized palletizing solution not only improves space utilization and operating efficiency, but also significantly enhances the automation and intelligence level of the palletizing process, ensures the safety and reliability of the palletizing process, and continuously optimizes future palletizing solutions through continuous improvement.

[0136] Material information collection and database storage in S1 include:

[0137] S11, dimension collection: using a laser rangefinder or a 3D scanner to measure the external dimensions of the material and generate three-dimensional dimension data of the material;

[0138] S12, weight collection: measure the weight of the material through the weighing sensor and record the weight data;

[0139] S13, shape acquisition: using 3D scanning equipment to obtain the shape contour and surface features of the material;

[0140] S14, material collection: analyzing the material characteristics of the material through a material identification sensor (such as a spectrometer or an infrared sensor) and recording the material information;

[0141] S15, data storage: transmitting the collected size, weight, shape and material information to the database through the data interface for storage;

[0142] Through the above content, laser rangefinders, 3D scanners, weighing sensors and material identification sensors are used to comprehensively collect material size, weight, shape and material information, and store these data in a central database, providing accurate and detailed material characteristic data for palletizing model establishment and optimization algorithms, thereby significantly improving the accuracy, stability and efficiency of the palletizing solution.

[0143] The palletizing model establishment and preliminary solution generation in S2 include:

[0144] S21, 3D model establishment: Based on the collected material size, shape and material, use 3D modeling software to establish a 3D model of the material to ensure that the model accurately reflects the actual characteristics of the material, including:

[0145] Dimensional data processing: Import the material dimension data obtained by laser rangefinder or 3D scanner into 3D modeling software to generate the basic geometric shape of the material;

[0146] Shape data processing: Using the material surface shape data obtained by the 3D scanner, through point cloud processing and surface reconstruction algorithm, the detailed three-dimensional shape of the material is generated, including:

[0147] a. Point cloud data acquisition: Use the laser or optical detector of a 3D scanner to scan the surface of the material, store the three-dimensional coordinates (x, y, z) of each point acquired by the scanner as point cloud data, and import these point cloud data into the computer for further processing;

[0148] b. Point cloud preprocessing: Preprocess the original point cloud data, including denoising, removing redundant points, and filtering invalid data to ensure that the quality of the point cloud data is high enough;

[0149] c. Point cloud registration: Use the iterative closest point algorithm to align the point cloud data from multiple scanning perspectives. During the alignment process, through transformation operations such as rotation and translation, the point clouds from each perspective are accurately spliced ​​together to form a complete 3D point cloud model, which is expressed as:

[0150]

[0151] Among them, p i is a point in the source point cloud, q i is the nearest point in the target point cloud, R is the rotation matrix, T is the translation vector, and the point cloud data is aligned by iteratively updating R and T by minimizing the error E(R,T);

[0152] d. Surface reconstruction: Use the Poisson reconstruction algorithm to connect discrete point cloud data to generate a continuous triangular mesh. The generated triangular mesh model accurately represents the surface shape of the material, including complex geometric details, expressed as:

[0153]

[0154] where v is the vector field, ρ is the density function generated from the point cloud, It is the divergence of the vector field. Poisson reconstruction generates a continuous surface describing the surface shape of the material by solving the Poisson equation.

[0155] e. Model optimization: Further optimize the generated 3D surface model, remove unnecessary complex surfaces, and smooth the surface to ensure the smoothness and usability of the model, which is expressed as:

[0156] p i '=p i +λ∑ j∈N(i) (p j -p i );

[0157] Among them, p i' is the point position after smoothing, p i is the original position of the current point, p j is the position of the neighbor point, N(i) is the neighbor set of point i, and λ is the smoothing coefficient, which determines the degree of smoothing;

[0158] f. Output 3D shape model: Export the final 3D surface model to the modeling software to generate a complete 3D shape model of the material for subsequent palletizing solution optimization and simulation;

[0159] Material property mapping: Based on the material information collected by the material recognition sensor, the corresponding material properties (such as density, rigidity, friction coefficient, etc.) are applied in the 3D model to ensure that the physical properties of the model are consistent with the actual material;

[0160] Model calibration: Compare the generated 3D model with the actual material, and ensure that the size, shape and material properties of the 3D model accurately reflect the actual properties of the material through adjustment and calibration;

[0161] S22, palletizing rule setting: according to the characteristics of the materials and the preset palletizing rules (such as heavy objects at the bottom, light objects at the top, fragile objects in the middle, etc.), set the basic palletizing rules;

[0162] S23, preliminary plan generation: Based on the three-dimensional model of the material and the set palletizing rules, a preliminary palletizing plan is generated, including the placement of the material, the palletizing sequence and the number of stacking layers, to ensure that the preliminary plan can meet the basic requirements of space utilization and stability;

[0163] S24, Solution Verification: Perform preliminary verification on the generated preliminary palletizing solution to ensure its feasibility and provide basic data for subsequent optimization steps, including:

[0164] Structural stability analysis: Use finite element analysis (FEA) to conduct structural stability analysis on the preliminary palletizing solution to ensure that each layer of material will not tip over or slide after being stacked, including:

[0165] a. Establish finite element model: Based on the collected material size and shape data, establish the geometric model of each material, assign corresponding material properties to each material according to the material properties (such as density, Young's modulus, Poisson's ratio, etc.), and divide the material model into finite element units (such as tetrahedron and hexahedron units) for subsequent numerical analysis;

[0166] b. Apply boundary conditions and loads: Apply constraints (boundary conditions) and external loads (such as gravity and contact force) to each layer of material in the palletizing solution. These conditions are used to simulate the force conditions of the material during the palletizing process, including:

[0167] Gravity load: The weight of the material is the main load, and the calculation formula is:

[0168] F g =m·g;

[0169] Where m is the mass of the material, g is the acceleration due to gravity (9.81 m / s 2 );

[0170] Contact load: The contact load between materials is calculated by contact force to ensure that the materials will not slide or tip over due to uneven force;

[0171] c. Solve the equilibrium equation: The finite element method is used to solve the equilibrium equation of the material during the stacking process, and the stress, strain and displacement of the material are calculated, which can be expressed as:

[0172] [K]{u}={P};

[0173] Where K is the stiffness matrix, u is the node displacement vector, and P is the external force load vector;

[0174] K = ∫ V B T DBdV;

[0175] Where B is the displacement-strain matrix, D is the elastic matrix of the material, and V is the volume of the material;

[0176] Displacement solution: By solving the displacement vector u, the displacement of the material after being subjected to force can be obtained;

[0177] d. Stress and strain analysis: According to the displacement vector obtained by solving, the stress and strain distribution of the material is calculated to determine the stability of the material during the stacking process, which is expressed as:

[0178] ε=B·u;

[0179] Where ε is the strain vector, B is the displacement-strain matrix, and u is the displacement vector;

[0180] σ=D·ε;

[0181] Among them, σ is the stress vector, D is the elastic matrix of the material, and ε is the strain vector;

[0182] e. Overturning analysis: By analyzing the center of gravity and force of the material, determine whether the material may overturn, calculate the overturning moment of each layer of material, and ensure that it is less than the supporting moment, expressed as:

[0183]

[0184] Among them, r c is the center of gravity, m i is the mass of the material in the i-th layer, ri is the center of gravity of the material in the i-th layer, and M is the total mass of the material;

[0185]

[0186] Among them, M tilt is the overturning moment, F i is the weight of the material, d i is the water distance from the center of gravity to the support point;

[0187] M support =F support b.

[0188] Among them, F support is the support force, b is the distance between the support surface and the center of gravity, ensuring that M support >M tilt , to prevent capsizing;

[0189] f. Sliding analysis: By analyzing the friction between materials, ensure that the materials will not slide due to insufficient friction. The sliding trend is calculated based on the normal pressure and friction coefficient on the contact surface, expressed as:

[0190] F friction =μ·F normal ;

[0191] Among them, F friction is the friction between materials, μ is the friction coefficient, F normal is the normal force between the materials;

[0192] F slide =F horizontal -F friction ;

[0193] If F slide >0, the material may slip and needs to be adjusted;

[0194] g. Result analysis and adjustment: Based on the calculation results, determine the stability of the material stacking. If a tipping or sliding trend is found, adjust the placement and stacking order of the materials and recalculate their stability;

[0195] Load distribution verification: simulate the load distribution of each layer of materials to verify the overall balance and stability after stacking, including:

[0196] a. Establish a load distribution model: According to the geometric shape and size of the material, build a three-dimensional model of each layer of material. According to the material weight, stacking order and material properties, determine the dead weight and stacking pressure of each layer of material, and establish the material load distribution function to calculate the pressure distribution of each layer of material at different contact points;

[0197] b. Calculate the gravity load of each layer: Calculate the gravity load on each layer of materials during the stacking process. The gravity load is the main vertical load, which is transferred from the weight of the upper layer of materials to the lower layer of materials. It is expressed as:

[0198] F i =m i g;

[0199] Among them, F i is the gravity load of the material in the i-th layer (N), m i is the mass of the material in the i-th layer (kg), g is the acceleration due to gravity, and its value is 9.81m / s 2 ;

[0200] c. Calculation of contact surface pressure distribution: Calculate the pressure distribution on the contact surface of each layer of material. Considering that the material may be subjected to uneven force on the contact surface, the pressure distribution will affect the balance and stability of the material, which can be expressed as:

[0201]

[0202] Among them, P ij is the pressure of the i-th layer of material on the j-th contact point, F i is the total load of the material in the i-th layer, A ij It is the contact area of ​​the material at the contact point;

[0203] d. Stress analysis: Calculate the stress of each layer of material based on the pressure distribution. The stress distribution is used to determine whether the material will be locally deformed or unstable due to uneven stress, expressed as:

[0204]

[0205] Among them, σ ij is the stress on the contact surface, P ij is the pressure at the contact point, A ij is the contact area;

[0206] e. Balance analysis: Analyze the moment balance of each layer of materials to determine whether the materials are in a balanced state and prevent the materials from tilting or sliding. The moment balance of the materials is the key factor in maintaining the stability of the stacking structure, which is expressed as:

[0207]

[0208] Among them, M balance is the short-term balance of force in each layer of material, F ij is the load of the i-th layer of material on the j-th contact point, d ij is the horizontal distance from the contact point to the center of gravity of the material. If M balance =0, indicating that the material is in equilibrium. If Mbalance ≠0, tilt or slide may occur;

[0209] f. Sliding analysis: Analyze whether each layer of material will slide due to insufficient friction. The magnitude of friction determines the stability of the material on the contact surface, which is expressed as:

[0210] F friction =μ·F normal ;

[0211] Among them, F friction is the friction force, μ is the friction coefficient between the material and the contact surface, F normal is the normal force at the contact point;

[0212] g. Stability verification: By simulating the stress conditions of the materials during the stacking process, determine whether the materials are unevenly stressed, have excessive stress concentration, or are unstable. If unbalanced or unstable areas are found, adjust the placement, angle, or stacking order of the materials;

[0213] Space utilization assessment: Evaluate the space utilization of the preliminary palletizing solution to ensure that the storage space is maximized while ensuring stability;

[0214] Feasibility test: Conduct feasibility test in a simulation environment to simulate actual palletizing operations and verify the operability and execution efficiency of the palletizing solution;

[0215] Through the above content, the accuracy and feasibility of the preliminary palletizing plan can be ensured, which not only improves the accuracy and stability of the palletizing plan, but also maximizes the utilization of storage space and ensures the efficient execution of palletizing operations, providing a reliable data basis for subsequent optimization steps.

[0216] The optimization algorithm calculation and stability verification in S3 include:

[0217] S31, optimization algorithm calculation: Based on the preliminary palletizing plan, the genetic algorithm is used to calculate the optimal palletizing path, including:

[0218] Fitness function design: The fitness function F is defined to evaluate the pros and cons of each palletizing solution. The calculation formula is:

[0219] F = α × Space Utilization + β × Stability + γ × Efficiency;

[0220] Among them, α, β, and γ are weight coefficients respectively, SpaceUtilization represents space utilization, Stability represents palletizing stability, and Efficiency represents palletizing efficiency;

[0221] Initial population generation: Generate an initial population based on the preliminary palletizing plan, where each individual represents a palletizing path;

[0222] Selection, crossover and mutation: through selection, crossover and mutation operations, iteratively generate new populations and gradually optimize the palletizing path;

[0223] Termination condition: Set the iteration termination condition. When the fitness function value converges or reaches the preset number of iterations, the optimal palletizing path is output;

[0224] S32, stability check: stability analysis of the generated palletizing solution, including:

[0225] S321, center of gravity calculation: Calculate the center of gravity of each layer of material to ensure that the center of gravity of the overall palletizing structure is located within the support surface. The calculation formula is:

[0226]

[0227] Among them, G x and G y are the x- and y-coordinates of the center of gravity, w i is the weight of the i-th material, x i and i are the x-coordinate and y-coordinate of the i-th material respectively;

[0228] S322, overturning analysis: Using the moment balance principle, calculate the overturning moment of each material after stacking to ensure that the total overturning moment is less than the supporting moment. The calculation formula is:

[0229]

[0230] Among them, M t is the total overturning moment, d i is the distance from the i-th material to the supporting surface;

[0231] S323, sliding analysis: Through friction analysis, calculate the friction between materials to ensure that no sliding occurs during the palletizing process. The calculation formula is:

[0232] F f =μ×N;

[0233] Among them, F f is the friction force, μ is the friction coefficient, and N is the positive pressure;

[0234] Through the above steps, the optimal palletizing path can be effectively calculated, and the generated palletizing solution can be ensured to be optimal in terms of space utilization, stability and operational efficiency.

[0235] The simulation verification and solution optimization in S4 include:

[0236] S41, Simulation environment construction: Build a simulation model that matches the actual palletizing scene in the virtual environment, including three-dimensional models of palletizing equipment, materials and palletizing areas, to ensure that the simulation environment is as close to the actual situation as possible;

[0237] S42, Palletizing Scheme Simulation: Execute the initially generated palletizing scheme in the simulation environment, simulate the material placement and stacking process through the physical engine, monitor the material stress, deformation and stability in real time, and verify the feasibility and stability of the scheme;

[0238] S43, data recording: recording the simulation data generated during the simulation process, including the placement position of the material, the number of stacking layers, the stress condition, and the deformation condition;

[0239] S44, Application of iterative optimization algorithm: Based on the simulation results, the iterative optimization algorithm is used to optimize the palletizing scheme, including:

[0240] S441, fitness function update: adjust the weight coefficients (α, β, γ) in the fitness function according to the simulation results to reflect the problems found in the simulation and the optimization goals. The calculation formula is:

[0241] β′=β+k×(Stability target -Stability current );

[0242] α′=α-k×(Stability target -Stability current ) / 2;

[0243] γ′=γ-k×(Stability target -Stability current ) / 2;

[0244] Among them, k is the adjustment coefficient, Stability target Stability current is the stability in the current simulation results;

[0245] S442, plan adjustment: Generate a new palletizing plan using the particle swarm optimization algorithm, adjust the placement, sequence and number of stacking layers of the materials, and generate a new plan;

[0246] The calculation formula of the particle swarm optimization algorithm is:

[0247]

[0248] x i (t+1)=x i (t)+v i(t+1);

[0249] Among them, v i (t) and x i (t) represents the velocity and position of the ith particle in the tth generation, ω is the inertia weight, c1 and c2 are acceleration constants, r1 and r2 are random numbers between [0,1], is the best historical position of the ith particle, g best is the global optimal position of all particles;

[0250] S443, multiple simulations and optimizations: Repeatedly perform simulation of the palletizing scheme, data recording, and iterative optimization algorithm application until the simulation results meet the preset stability, space utilization, and efficiency requirements;

[0251] S45, solution determination: Based on multiple simulations and optimizations, the final optimized palletizing solution is determined to ensure that the solution is highly feasible and stable in practical applications;

[0252] Through the above steps, the palletizing scheme is verified and optimized multiple times in the virtual environment to ensure that the final generated palletizing scheme has good feasibility and stability and achieves the best effect in practical applications. The application of fitness function and optimization algorithm ensures the comprehensive optimization of the palletizing scheme in terms of space utilization, stability and efficiency.

[0253] The material deformation characteristics analysis in S51 includes:

[0254] S511, physical simulation analysis: Use finite element analysis method to perform physical simulation on materials, simulating the stress and deformation of materials during the palletizing process, including:

[0255] Establish material model: Establish finite element model of the material according to the size, shape and material of the material. The finite element model can use nodes and units to represent the geometry and material properties of the material.

[0256] Apply boundary conditions and loads: According to the actual palletizing environment, apply corresponding boundary conditions and loads to simulate the stress state of materials during the palletizing process;

[0257] Solve deformation: Use the finite element solver to calculate the deformation of the material under different stress conditions. The calculation formula is:

[0258] Ku=P;

[0259] Where K is the stiffness matrix, u is the node displacement vector, and P is the external force load vector;

[0260] S512, historical material information analysis: using historical material information to perform statistical analysis on the deformation characteristics of different materials, specifically including:

[0261] Data collection: Collect and organize material deformation data recorded during historical palletizing, including deformation, stress conditions and environmental conditions;

[0262] Data modeling: Based on historical material information, a statistical model of material deformation characteristics is established. The calculation formula is:

[0263] Δx=aP+bE+c;

[0264] Among them, Δx is the deformation variable, P is the external force load vector, E is the environmental condition (such as temperature, humidity, etc.), a, b, c are constant coefficients obtained by data fitting;

[0265] Characteristic identification: The deformation characteristics of different materials under given force and environmental conditions are identified through cluster analysis to form a deformation characteristic library. The calculation formula is:

[0266]

[0267] Among them, J is the objective function (the sum of squares within the cluster), k is the number of clusters, x is the data point, and C j is the jth cluster, h j is the centroid of the jth cluster;

[0268] Through the above content, the deformation characteristics of different materials during the palletizing process can be accurately identified, providing data support and model basis for subsequent real-time deformation monitoring, prediction and compensation, and improving the stability and accuracy of the palletizing process.

[0269] Real-time deformation monitoring in S52 includes:

[0270] S521, sensor deployment: deploy a variety of sensors on palletizing equipment and materials, including displacement sensors, strain gauges and pressure sensors, to monitor the deformation of materials during the palletizing process;

[0271] S522, deformation data collection: real-time measurement of material displacement changes through displacement sensors, monitoring of material strain through strain gauges, and monitoring of material surface stress through pressure sensors;

[0272] Through the above steps, deformation data such as displacement, strain and stress of materials can be obtained in real time during the palletizing process, ensuring that the deformation is accurately recorded and monitored, providing instant data support for deformation prediction and compensation, and improving the accuracy and stability of the palletizing process.

[0273] The deformation prediction model in S53 uses an artificial neural network (ANN), which includes:

[0274] S531, input layer data preprocessing: In order to make the input data of different scales within the same range and improve the training efficiency and prediction accuracy of the neural network, the input data is normalized. The calculation formula is:

[0275]

[0276] Among them, x i is the original input data, x i ′ is the normalized input data, μ x and σ x are the mean and standard deviation of the input data respectively;

[0277] S532, input layer: define the number of nodes in the input layer, determine the number of nodes in the input layer according to the characteristic dimensions of the monitored deformation data (such as displacement, strain and stress) and historical logistics information, and pass the processed input feature vector to the next layer, expressed as:

[0278] X=[x1′,x2′,...,x n ′];

[0279] Among them, X is the input feature vector, which contains all the normalized input data;

[0280] S533, time series feature extraction: extract time series features through the convolution layer to capture the time series changes of deformation data. Multi-scale convolution kernels are used for feature extraction at different time scales to enhance the network's ability to capture deformation features. The calculation formula is:

[0281]

[0282] in, is the convolution output of the s-th scale, is the convolution kernel weight of the s-th scale, k s is the convolution kernel size of the sth scale, x t-i is the input at time ti, b (s) is bias;

[0283] S534, hidden layer structure design: By introducing adaptive activation function and skip connection, the nonlinear expression ability and feature capture ability of the network are enhanced. The adaptive activation function can dynamically adjust the shape of the activation function, and the skip connection enhances the stability of gradient transfer. The calculation formula is:

[0284]

[0285] in, is the activation value of the lth layer, W (l) is the weight matrix of the lth layer, b (l) is the bias of the lth layer, is the activation value of the first two layers;

[0286] The calculation formula of the adaptive activation function is:

[0287]

[0288] Among them, a and b are learnable parameters;

[0289] S535, Attention Mechanism: Introduce the attention mechanism in the convolutional layer to enhance the network's attention to important features and improve the accuracy of deformation prediction. The attention mechanism adjusts the weight of the input feature by calculating the attention weight. The calculation formula is:

[0290]

[0291] Among them, Q t =W Q H t , K t =W K H t 、V t =W V H t are query, key and value matrices respectively, d k is the dimension of the key, softmax is the normalization function, A t Output for attention;

[0292] S536, loss function: In order to better fit the deformation prediction, a weighted mean square error loss function is defined, and regularization terms and attention mechanism loss terms are added to prevent overfitting. The calculation formula is:

[0293]

[0294] Among them, L is the loss function, N is the number of samples, and w i is the weight of the sample, is the predicted value, y i is the true value, λ is the regularization coefficient, ||W (l) || 2 is the L2 norm of the weight matrix of the first layer, β is the regularization coefficient of the attention mechanism, ||A (l) || 2 is the L2 norm of the attention matrix of the lth layer;

[0295] S537, output layer: The number of output layer nodes is determined according to actual needs. Usually, it is a single node, which is used to predict the deformation of the material. The calculation formula is:

[0296]

[0297] in, is the predicted output, W (o) is the weight matrix of the output layer, is the activation value of the last hidden layer, b (o) is the bias of the output layer;

[0298] Through the above-mentioned artificial neural network, the deformation characteristics of materials during the palletizing process can be accurately captured and predicted. By preprocessing the input data and designing the loss function, the training efficiency and prediction accuracy of the model are improved, overfitting is prevented and the stability of the model is enhanced, making the deformation prediction more accurate, thereby improving the stability and space utilization of the palletizing process, optimizing the palletizing plan, and ensuring the efficiency and safety of the entire process.

[0299] Dynamic solution adjustments in S54 include:

[0300] S541, deformation trend prediction: Based on the artificial neural network model, the deformation trend of materials during the palletizing process is predicted to obtain the estimated deformation amount of each material;

[0301] S542, real-time adjustment of placement: According to the predicted deformation trend, dynamically adjust the placement of each material to ensure that the material will not fall or slide during the palletizing process, including:

[0302] Calculate the adjustment vector: According to the deformation prediction results, calculate the position adjustment vector Δp ​​of the material. The calculation formula is:

[0303] Δp=η×Δx;

[0304] Among them, Δp is the position adjustment vector, η is the adjustment coefficient, and Δx is the predicted deformation;

[0305] Update placement position: Update the original placement position p to the adjusted placement position p′. The calculation formula is:

[0306] p′=p+Δp;

[0307] S543, Optimize the placement angle: Based on the deformation prediction results of the material, optimize the material placement angle to minimize the impact of deformation on the stability of palletizing, including:

[0308] Calculate the adjustment angle: According to the deformation prediction results, calculate the material placement angle adjustment Δθ. The calculation formula is:

[0309]

[0310] Among them, Δθ is the angle adjustment amount, ξ is the adjustment coefficient, is the predicted deformation angle change;

[0311] Update the placement angle: Update the original placement angle θ to the adjusted placement angle θ′. The calculation formula is:

[0312] θ′=θ+Δθ;

[0313] Through the above steps, the placement and angle of the materials can be optimized in real time according to the predicted deformation trend, the stability and accuracy of the palletizing process can be improved, and the feasibility and reliability of the palletizing solution can be ensured.

[0314] Palletizing and real-time monitoring in S6 include:

[0315] S61, input optimization plan: transmit the final optimized palletizing plan to the palletizing robot or automation equipment. The final optimized palletizing plan includes the placement position, angle and stacking order of each material;

[0316] S62, executing the palletizing operation: the palletizing robot or automated equipment performs the actual palletizing operation according to the input optimization plan, places the materials one by one according to the instructions in the optimization plan, and ensures that the materials are stacked at the set position and angle;

[0317] S63, real-time monitoring: During the palletizing process, sensors (such as cameras, laser rangefinders, strain gauges, etc.) installed on robots or automated equipment monitor each step of the palletizing and the state of the materials in real time, including monitoring the displacement, stress and stacking stability of the materials;

[0318] S64, abnormality detection and adjustment: Analyze the results of real-time monitoring, detect abnormal conditions during the palletizing process (such as material sliding, tilting, deformation, etc.), and make timely adjustments;

[0319] Anomaly detection and adjustment specifically include:

[0320] Abnormality detection: According to the preset threshold and algorithm, abnormal situations are identified, and alarms are triggered or palletizing operations are stopped. The calculation formula is:

[0321]

[0322] Among them, E t is an outlier, p t and θ t are the current material position and angle, and is the expected position and angle;

[0323] Adjustment plan: Adjust the plan according to abnormal conditions, dynamically modify the placement and angle of materials, re-optimize the palletizing path, and input the adjusted plan into the palletizing robot or automation equipment;

[0324] Through the above steps, abnormal situations in the palletizing process can be identified and handled in real time, ensuring the safety and stability of the palletizing process, and improving the palletizing efficiency and quality through dynamic adjustment.

[0325] Specific example:

[0326] 1. Scenario description: In an automated warehouse, heavy objects such as metal boxes, fragile glass boxes, plastic products, etc. need to be stacked in multiple layers. During the stacking process, the materials may deform due to differences in weight and material, which in turn affects the stability of the stacking. Therefore, the present invention uses sensors and multiple algorithms to monitor the deformation of materials in real time, predict deformation trends, and ensure the stability of the stacking by adjusting the placement and angle of the materials.

[0327] 2. Material information:

[0328] Material 1: Metal box, weight 50kg, size 50cm×30cm×30cm, made of steel.

[0329] Material 2: Glass box, weight 10kg, size 40cm×20cm×20cm, made of glass.

[0330] Material 3: plastic product, weight 5kg, size 30cm×30cm×30cm, material is plastic.

[0331] 3. Deformation prediction and adjustment steps: Use sensors to monitor the force and deformation of materials in real time, use artificial neural network (ANN) models to predict deformation trends, and adjust the placement and angle of materials in real time to ensure the safety of palletizing, including:

[0332] 3.1. Material deformation characteristics analysis: The size, weight, material and deformation characteristics of the material are collected through sensors. Assume that the data collected by the sensor is as follows:

[0333] Material 1: The displacement sensor measures a deformation of 0.5 mm, and the stress sensor measures a stress of 200 N / m 2 , the strain is 0.002.

[0334] Material 2: The displacement sensor measures a deformation of 2 mm, and the stress sensor measures a stress of 300 N / m 2 , the strain is 0.005.

[0335] Material 3: The displacement sensor measures a deformation of 1 mm, and the stress sensor measures a stress of 100 N / m 2 , the strain is 0.003.

[0336] 3.2. Deformation prediction based on artificial neural network: Assume that the artificial neural network model has been trained with historical data. The model inputs the following feature data:

[0337] Material size;

[0338] Material weight;

[0339] Real-time monitoring of stress and strain data;

[0340] Environmental conditions (such as temperature, humidity, etc.);

[0341] Prediction by artificial neural network model:

[0342] Material 1 (metal box): deformation Δx1 = 0.6 mm;

[0343] Material 2 (glass box): deformation Δx2 = 2.5 mm;

[0344] Material 3 (plastic product): deformation Δx3 = 1.2 mm;

[0345] 3.3. Dynamic solution adjustment: According to the deformation prediction results, the material placement and angle are adjusted in real time to compensate for the impact of deformation on stacking stability. The specific adjustment process is as follows:

[0346] (1) Adjustment of material 1 (steel metal box): Since the deformation of the metal box is small, only 0.6 mm, the main adjustment is the placement angle to balance the force.

[0347] Angle adjustment: Calculate the adjustment angle: Δθ1 = η × Δx1 = 0.05 × 0.6 = 0.03 degrees;

[0348] The adjusted placement angle θ1′ is: θ1′=θ1+Δθ1=90°+0.03°=90.03°.

[0349] (2) Adjustment of Material 2 (Glass Box): The deformation of the glass box is large and its position needs to be adjusted to ensure that the glass does not tilt or slide due to deformation.

[0350] Position adjustment: Calculate the position adjustment vector: Δp2 = η × Δx2 = 0.1 × 2.5 = 0.25 mm;

[0351] Adjusted placement position: p2'=p2+Δp2=0mm+0.25mm=0.25mm;

[0352] Adjustment of material 3 (plastic product): The deformation of the plastic product is 1.2 mm, so both the position and angle need to be adjusted.

[0353] Position adjustment: Δp3 = η × Δx3 = 0.1 × 1.2 = 0.12 mm;

[0354] Adjusted placement position: p3'=p3+Δp3=0mm+0.12mm=0.12mm;

[0355] Angle adjustment: Δθ3 = η × Δx3 = 0.05 × 1.2 = 0.06 degrees;

[0356] Adjusted placement angle: θ3' = θ3 + Δθ3 = 90° + 0.06° = 90.06°;

[0357] 3.4. Solution optimization: Through multiple adjustments and optimizations, ensure that the deformation and force of all materials are within the allowable range, and finally form a stable palletizing solution. The optimized stacking structure can achieve efficient space utilization and ensure that there will be no tilting, sliding or unstable stacking in subsequent palletizing operations.

[0358] 3.5. Final feedback: Feedback all adjusted data (position, angle, deformation, etc.) to the database, generate a palletizing report, and provide an optimization reference for the next palletizing operation.

[0359] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present invention is limited to these examples. Under the concept of the present invention, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the present invention as described above, which are not provided in detail for the sake of simplicity.

Claims

1. The palletizing optimization algorithm based on multi-layer palletizing of materials is characterized by: The following steps are involved: S1, material information collection and database storage: collect information about the materials to be palletized, including size, weight, shape and material, through sensors and scanning equipment, and store the material information in the database; S2, palletizing model establishment and preliminary plan generation: Based on the collected material information, a three-dimensional model of the material is established, and a preliminary palletizing plan is generated according to the preset palletizing rules, including the placement, sequence and number of stacking layers of the materials; S3, optimization algorithm calculation and stability verification: Based on the preliminary palletizing solution, the optimization algorithm is used to calculate the best palletizing path, and the generated palletizing solution is subjected to stability analysis. The optimization algorithm calculation and stability verification include: S31, optimization algorithm calculation: based on the preliminary palletizing plan, the optimal palletizing path is calculated using a genetic algorithm; S32, stability check: stability analysis of the generated palletizing solution, including: S321, center of gravity calculation: Calculate the center of gravity of each layer of material. The calculation formula is: Among them, G x and G y are the x- and y-coordinates of the center of gravity, w i is the weight of the i-th material, x i and i are the x-coordinate and y-coordinate of the i-th material respectively; S322, overturning analysis: Using the moment balance principle, calculate the overturning moment of each material after stacking. The calculation formula is: Among them, M t is the total overturning moment, d i is the distance from the i-th material to the supporting surface; S323, sliding analysis: Calculate the friction between materials through friction analysis. The calculation formula is: F f =μ×N; Among them, F f is the friction force, μ is the friction coefficient, and N is the positive pressure; S4, simulation verification and solution optimization: simulate the palletizing solution in a virtual environment to verify the feasibility and stability of the solution, and record and analyze the simulation results. Based on the simulation results, optimize the palletizing solution through an iterative optimization algorithm. The simulation verification and solution optimization include: S41, simulation environment construction: construct a simulation model that matches the actual palletizing scene in the virtual environment, including three-dimensional models of palletizing equipment, materials and palletizing areas; S42, Palletizing Scheme Simulation: Execute the initially generated palletizing scheme in the simulation environment, simulate the placement and stacking process of materials through the physical engine, monitor the force, deformation and stability of the materials in real time, and verify the feasibility and stability of the scheme; S43, data recording: recording the simulation data generated during the simulation process, including the placement position of the material, the number of stacking layers, the stress condition, and the deformation condition; S44, Application of iterative optimization algorithm: Based on the simulation results, the iterative optimization algorithm is used to optimize the palletizing scheme, including: S441, fitness function update: adjust the weight coefficient in the fitness function according to the simulation results to reflect the problems found in the simulation and the optimization goals; S442, plan adjustment: Generate a new palletizing plan using the particle swarm optimization algorithm, adjust the placement, sequence and number of stacking layers of the materials, and generate a new plan; S443, multiple simulations and optimizations: Repeatedly perform simulation of the palletizing scheme, data recording, and iterative optimization algorithm application until the simulation results meet the preset stability, space utilization, and efficiency requirements; S45, solution determination: based on multiple simulations and optimizations, determine the final optimized palletizing solution; S5, Adaptive deformation prediction and compensation algorithm: predicts the deformation of different materials during the palletizing process and compensates by adjusting the palletizing solution in real time, including: S51, Material deformation characteristics analysis: Through physical simulation and historical material information analysis, the deformation characteristics of different materials during the palletizing process are identified; S52, real-time deformation monitoring: During the palletizing process, sensors are used to monitor the deformation of materials in real time and collect deformation data; S53, deformation prediction: Based on real-time monitoring deformation data and historical logistics information, a deformation prediction model is constructed to predict the deformation trend of materials; S54, dynamic scheme adjustment: according to the predicted deformation trend, the stacking scheme is adjusted in real time to optimize the placement angle and position of the materials; S55, deformation compensation execution: in the actual palletizing operation, the adjusted palletizing plan is executed; S6, palletizing and real-time monitoring: input the final optimized palletizing solution into the palletizing robot or automation equipment, perform actual palletizing operations, and monitor abnormal conditions during the palletizing process in real time to make timely adjustments; S7, Result Feedback and Continuous Improvement: Record the data and results of the palletizing process, form a palletizing report, and store the feedback information in the database.

2. The palletizing optimization algorithm based on multi-layer palletizing of materials according to claim 1 is characterized in that: The material information collection and database storage in S1 include: S11, dimension collection: using a laser rangefinder or a 3D scanner to measure the external dimensions of the material and generate three-dimensional dimension data of the material; S12, weight collection: measure the weight of the material through the weighing sensor and record the weight data; S13, shape acquisition: using 3D scanning equipment to obtain the shape contour and surface features of the material; S14, material collection: analyzing the material characteristics of the material through the material recognition sensor and recording the material information; S15, data storage: the collected size, weight, shape and material information is transmitted to the database through the data interface for storage.

3. The palletizing optimization algorithm based on multi-layer palletizing of materials according to claim 2 is characterized in that: The palletizing model establishment and preliminary solution generation in S2 include: S21, 3D model establishment: based on the collected material size, shape and material, a 3D model of the material is established using 3D modeling software; S22, palletizing rule setting: according to the characteristics of the material and the preset palletizing rules, set the basic rules for palletizing; S23, generating a preliminary plan: generating a preliminary palletizing plan based on the three-dimensional model of the material and the set palletizing rules, including the placement of the material, the palletizing sequence, and the number of stacking layers; S24, solution verification: preliminary verification of the generated preliminary palletizing solution.

4. The palletizing optimization algorithm based on multi-layer palletizing of materials according to claim 1 is characterized in that: The material deformation characteristic analysis in S51 includes: S511, physical simulation analysis: Use finite element analysis method to perform physical simulation on materials, simulating the stress and deformation of materials during the palletizing process, including: Establish material model: Establish finite element model of material according to its size, shape and material; Apply boundary conditions and loads: According to the actual palletizing environment, apply corresponding boundary conditions and loads to simulate the stress state of materials during the palletizing process; Solve deformation: Use the finite element solver to calculate the deformation of the material under different stress conditions. The calculation formula is: Ku=P; Where K is the stiffness matrix, u is the node displacement vector, and P is the external force load vector; S512, historical material information analysis: using historical material information to perform statistical analysis on the deformation characteristics of different materials, specifically including: Data collection: Collect and organize material deformation data recorded during historical palletizing, including deformation, stress conditions and environmental conditions; Data modeling: Based on historical material information, a statistical model of material deformation characteristics is established. The calculation formula is: Δx=aP+bE+c; Among them, Δx is the deformation, P is the external force load vector, E is the environmental condition, and a, b, and c are constant coefficients; Characteristic identification: The deformation characteristics of different materials under given force and environmental conditions are identified through cluster analysis to form a deformation characteristic library. The calculation formula is: Among them, J is the objective function, k is the number of clusters, x is the data point, C j is the jth cluster, h j is the centroid of the jth cluster.

5. The palletizing optimization algorithm based on multi-layer palletizing of materials according to claim 4 is characterized in that: The real-time deformation monitoring in S52 includes: S521, sensor deployment: deploy a variety of sensors on palletizing equipment and materials, including displacement sensors, strain gauges and pressure sensors, to monitor the deformation of materials during the palletizing process; S522, deformation data collection: real-time measurement of material displacement changes through displacement sensors, monitoring of material strain through strain gauges, and monitoring of material surface stress through pressure sensors.

6. The palletizing optimization algorithm based on multi-layer palletizing of materials according to claim 1 is characterized in that: The dynamic scheme adjustment in S54 includes: S541, deformation trend prediction: Based on the artificial neural network model, the deformation trend of materials during the palletizing process is predicted to obtain the estimated deformation amount of each material; S542, real-time adjustment of placement: dynamically adjust the placement of each material according to the predicted deformation trend, including: Calculate the adjustment vector: According to the deformation prediction results, calculate the position adjustment vector Δp ​​of the material. The calculation formula is: Δp=η×Δx; Among them, Δp is the position adjustment vector, η is the adjustment coefficient, and Δx is the predicted deformation; Update placement position: Update the original placement position p to the adjusted placement position p′. The calculation formula is: p′=p+Δp; S543, optimize the placement angle: optimize the material placement angle according to the deformation prediction result of the material, including: Calculate the adjustment angle: According to the deformation prediction results, calculate the material placement angle adjustment Δθ. The calculation formula is: Among them, Δθ is the angle adjustment amount, ξ is the adjustment coefficient, is the predicted deformation angle change; Update the placement angle: Update the original placement angle θ to the adjusted placement angle θ′. The calculation formula is: θ′=θ+Δθ.

7. The palletizing optimization algorithm based on multi-layer palletizing of materials according to claim 1 is characterized in that: The palletizing and real-time monitoring in S6 include: S61, input optimization plan: transmit the final optimized palletizing plan to the palletizing robot or automation equipment; S62, executing the palletizing operation: the palletizing robot or automated equipment performs the actual palletizing operation according to the input optimization plan, and places the materials one by one according to the instructions in the optimization plan; S63, real-time monitoring: During the palletizing process, sensors installed on robots or automated equipment monitor each step of the palletizing and the status of the materials in real time; S64, Abnormality Detection and Adjustment: Analyze the results of real-time monitoring, detect abnormalities during the palletizing process, and make timely adjustments.

Images