Intelligent loading scheduling optimization method for engineering equipment of intelligent port

Abstract

The invention particularly relates to an intelligent loading scheduling optimization method for engineering equipment of an intelligent port. Comprising the following steps: S1, acquiring equipment related information, weight information and project labels of current engineering equipment; s2, modeling a three-dimensional model of the engineering equipment; s3, acquiring real-time state information and cargo hold weight distribution information of the cargo hold of the bulk cargo ship; s4, modeling a real-time three-dimensional model of the cargo hold; s5, constructing a multi-objective optimization problem for the current engineering equipment, wherein the multi-objective optimization problem takes the maximum cargo space utilization rate, the maximum weight distribution uniformity and the highest aggregation degree with other engineering equipment of the same project as the objective; s6, solving a multi-objective optimization problem of the current engineering equipment to obtain an optimal loading scheme; s7, loading the current engineering equipment to a specified position in a cargo hold of the bulk carrier based on the optimal loading scheme of the current engineering equipment; and S8, the steps from S1 to S7 are repeated till loading of all engineering equipment is completed or the cargo hold of the bulk cargo ship is fully loaded. The loading, unloading and transporting efficiency of the intelligent port engineering equipment can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of engineering equipment loading in smart ports, and in particular to an intelligent loading scheduling optimization method for engineering equipment in smart ports. Background Art

[0002] With the continued growth of global trade and the rapid development of the logistics industry, smart ports, as a key direction in modern port development, are becoming a key driver of efficient port operations and sustainable development. Leveraging advanced technologies such as the Internet of Things, big data, and artificial intelligence, smart ports have automated, intelligent, and information-based port operations, significantly improving port cargo throughput and operational efficiency.

[0003] The transportation of construction equipment is an integral part of smart port logistics. Construction equipment primarily includes various types of transport vehicles, as well as excavators, cranes, loaders, road rollers, concrete pumps, and piling machinery. Construction equipment transportation involves the cross-regional transport of construction equipment via river or sea transport using bulk carriers. Construction equipment is typically placed in bulk directly on the cargo hold deck of a vessel, with some precision engineering equipment housed within simple protective frames.

[0004] However, the current process of transporting and loading engineering equipment on bulk carriers mainly has the following problems:

[0005] 1) Engineering equipment has irregular and heterogeneous features (such as the extended shape of the excavator arm, the irregular contour of the tower crane segment components, or the curved surface structure of the crawler chassis). Since loading is not planned based on the three-dimensional structural characteristics of the engineering equipment, a large number of gaps are easily left between the engineering equipment, reducing the space utilization rate of the cabin.

[0006] 2) Engineering equipment is very heavy. Failure to fully consider its weight distribution can easily lead to weight imbalance in the cargo hold, increasing the risk of the bulk carrier swaying and tilting during navigation, affecting navigation safety. Furthermore, uneven weight distribution can easily lead to serious consequences such as damage to the ship's structure, damage to cargo, and even capsizing.

[0007] 3) To reduce transportation costs, a mixed loading mode is adopted for equipment from multiple projects, and equipment from different projects often has different unloading locations and times. Existing transportation planning uses a linear loading and unloading strategy, stacking equipment from different projects according to loading sequence. This leads to cascading efficiency losses in unloading operations at the destination port. Specifically, equipment from the same project is stored in multiple compartments, increasing the frequency of cross-regional dispatching of unloading machinery, resulting in increased complexity and time costs during equipment unloading, and low loading and unloading efficiency.

[0008] Therefore, how to design a scheduling method that can improve the utilization rate of bulk carrier cargo hold space, the uniformity of equipment weight distribution in the cargo hold, and reduce the complexity and time cost of unloading during the transportation of engineering equipment is a technical problem that needs to be solved urgently. Summary of the Invention

[0009] In view of the deficiencies of the above-mentioned prior art, the technical problem to be solved by the present invention is: how to provide an intelligent loading and scheduling optimization method for engineering equipment in a smart port. By obtaining relevant information of engineering equipment and bulk carrier cargo holds, a multi-objective optimization problem is constructed and solved to obtain the optimal loading plan for engineering equipment, so as to improve the utilization rate of bulk carrier cargo hold space and the uniformity of equipment weight distribution, and reduce the complexity and time cost of unloading, thereby improving the efficiency of loading and unloading and transportation of engineering equipment in smart ports and reducing transportation costs.

[0010] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0011] A method for intelligent loading and scheduling optimization of engineering equipment in a smart port, comprising:

[0012] S1: Obtain equipment-related information, weight information, and project tags of the current engineering equipment;

[0013] S2: Perform three-dimensional modeling based on the equipment-related information of the current engineering equipment to obtain a three-dimensional model of the engineering equipment;

[0014] S3: Obtaining real-time status information and cargo hold weight distribution information of the bulk carrier;

[0015] S4: Modeling is performed based on the real-time status information of the cargo hold to obtain a real-time three-dimensional model of the cargo hold;

[0016] S5: Based on the real-time 3D model and weight distribution information of the cargo hold, as well as the 3D model, weight information, and project label of the current engineering equipment, a multi-objective optimization problem is constructed for the current engineering equipment, with the goals of maximizing cargo hold space utilization, maximizing weight distribution uniformity, and maximizing the degree of aggregation with other engineering equipment in the same project.

[0017] S6: Solve the multi-objective optimization problem of the current engineering equipment through intelligent algorithms to obtain the optimal loading plan; the optimal loading plan includes the placement of the current engineering equipment in the cargo hold;

[0018] S7: Loading the current engineering equipment to a corresponding position in the cargo hold based on the optimal loading plan for the current engineering equipment;

[0019] S8: Repeat steps S1 to S7 until all engineering equipment is loaded or the cargo hold of the bulk carrier is fully loaded.

[0020] Preferably, in step S2, a three-dimensional model of the engineering equipment is obtained by modeling through the following steps:

[0021] S201: Equipment related information includes point cloud data, multi-angle photos and external dimension data of engineering equipment;

[0022] S202: performing point cloud denoising, point cloud registration, and point cloud simplification on the point cloud data of the engineering equipment to obtain a point cloud model of the engineering equipment;

[0023] S203: extracting features from the multi-angle photos of the engineering equipment to obtain feature points of the photos at each angle; matching the feature points of the photos at each angle to obtain a photo feature point matching result;

[0024] S204: Selecting a reference photo from the multi-angle photos based on the photo feature point matching results, and geometrically correcting the other photos using the reference photo; fusing the color information and texture information in the geometrically corrected angled photos with the engineering equipment point cloud model, and interpolating the color information and texture information of points in the engineering equipment point cloud model that do not directly correspond to pixels in the photo using an interpolation algorithm to obtain a color point cloud model that fuses color and texture;

[0025] S205: Surface reconstruction is performed on the color point cloud model that integrates color and texture using a surface reconstruction algorithm to generate a three-dimensional mesh model of the engineering equipment;

[0026] S206: Performing size calibration and physical property assignment on the three-dimensional mesh model of the engineering equipment according to the external dimension data and weight information of the engineering equipment to obtain the three-dimensional model of the engineering equipment.

[0027] Preferably, in step S3, cargo hold weight distribution information is obtained through the following steps:

[0028] S301: Install several weight sensors at the bottom of the cargo hold, determine the installation location information of each weight sensor, and set a sensor identifier for each weight sensor;

[0029] S302: Collect weight values at corresponding positions through each weight sensor to obtain a data set containing a number of data points; each data point includes a corresponding sensor identifier, weight value, and collection time;

[0030] S303: Perform data cleaning, data alignment, and unit unification on the data set to obtain a preprocessed data set;

[0031] S304: Establish a two-dimensional coordinate system for the cargo hold with a corner point of the cargo hold as the origin and the length and width directions as coordinate axes;

[0032] S305: Determine the coordinates of each weight sensor in the cargo hold two-dimensional coordinate system based on the installation location information of each weight sensor, and construct a mapping table to associate the sensor identifier of each weight sensor with its coordinates in the cargo hold coordinate system;

[0033] S306: Divide the cargo hold into n×m small intervals along the length and width directions of the cargo hold, and construct a two-dimensional cargo hold weight distribution matrix of size n×m; the small intervals correspond one-to-one to the matrix cells in the cargo hold weight distribution matrix;

[0034] S307: Traverse each data point in the preprocessed data set, obtain the corresponding coordinates from the mapping table according to the sensor identifier of the data point; determine the matrix cell where the coordinates corresponding to the data point are located, and accumulate the weight value of the data point to the corresponding matrix cell;

[0035] S308: After processing all data points in step S307, the obtained cargo hold weight distribution matrix is used as cargo hold weight distribution information; each element in the cargo hold weight distribution matrix represents the weight of the corresponding small area in the cargo hold.

[0036] Preferably, in step S4, a real-time three-dimensional model of the cargo hold is obtained by modeling through the following steps:

[0037] S401: Real-time status information of the bulk carrier's cargo hold includes 3D point cloud data and photo images of loaded engineering equipment, as well as geometric dimension data of the cargo hold;

[0038] S402: performing point cloud preprocessing on the three-dimensional point cloud data of the loaded engineering equipment to obtain preprocessed three-dimensional point cloud data; performing point cloud feature extraction on the preprocessed three-dimensional point cloud data to obtain point cloud feature points;

[0039] S403: performing image preprocessing on the photo image of the loaded engineering equipment to obtain preprocessed image data; performing image feature extraction on the preprocessed image data to obtain image feature points;

[0040] S404: Matching the point cloud feature points with the image feature points to obtain successfully matched point cloud-image feature point pairs;

[0041] S405: Constructing a point cloud coordinate system based on the pre-processed 3D point cloud data; constructing a 3D coordinate system for the cargo hold with a corner point of the cargo hold as the origin, the length direction of the cargo hold as the X-axis, the width direction as the Y-axis, and the height direction as the Z-axis;

[0042] S406: Calculating a coordinate transformation matrix between the point cloud coordinate system and the cargo hold three-dimensional coordinate system based on the point cloud-image feature point pairs, the coordinates of the point cloud feature points in the point cloud coordinate system, and the coordinates of the image feature points in the cargo hold three-dimensional coordinate system;

[0043] S407: Generate a three-dimensional model framework of the cargo hold based on the geometric dimension data of the cargo hold;

[0044] S408: All point cloud data in the pre-processed three-dimensional point cloud data are converted into the cargo hold three-dimensional coordinate system through the coordinate conversion matrix; the converted point cloud data are integrated with the cargo hold three-dimensional model framework to obtain a preliminary cargo hold three-dimensional model that integrates the information of the loaded engineering equipment; the surface of the preliminary cargo hold three-dimensional model is reconstructed to obtain a real-time three-dimensional model of the cargo hold.

[0045] Preferably, in step S405, a point cloud coordinate system is constructed by the following steps:

[0046] S4051: Calculate the centroid of preprocessed 3D point cloud data

[0047] S4052: Calculate the centroid of each point in the preprocessed 3D point cloud data distance, and filter out the distance centroid The farthest target point p f ;

[0048] S4053: Centroid As the starting point and the target point p f Get a preliminary direction vector v for the end point i ; through the preliminary direction vector v i The projection of the preprocessed 3D point cloud data is divided into two regions, and the centroid of the point cloud data in the two regions is calculated respectively. and Connect centroids and Get the auxiliary direction vector v a ;

[0049] S4054: Calculate the covariance matrix of the preprocessed point cloud data and perform eigenvalue decomposition on the covariance matrix to obtain eigenvalues λ1≥λ2≥λ3 and their eigenvalue vectors v1, v2, and v3;

[0050] S4055: According to the preliminary direction vector v i and auxiliary direction vector v a Adjust the eigenvector to get the X coordinate of the point cloud system p 、Y p , Z p Axis direction vector;

[0051] S4056: Comprehensive X p 、Y p , Z p The reference point in the axis direction determines the origin of the point cloud coordinate system O p ;

[0052] S4057: Based on the origin of the point cloud coordinate system O p and X p 、Y p , Z p The point cloud coordinate system is constructed by the axis direction vector.

[0053] Preferably, in step S406, a coordinate transformation matrix is constructed by the following steps:

[0054] S4061: screening and normalizing the point cloud-image feature point pairs to obtain pre-processed feature point pairs;

[0055] S4062: Estimate the initial rotation matrix R0 and translation vector T0 using the singular value decomposition algorithm;

[0056] S4063: Iteratively optimize the initial rotation matrix R0 and translation vector T0 using an iterative closest point algorithm to obtain an optimized rotation matrix R and translation vector T;

[0057] S4064: Calculate the coordinate transformation matrix M = (R, T) based on the rotation matrix R and the translation vector T.

[0058] Preferably, in step S5, a multi-objective optimization problem is constructed by the following steps:

[0059] S501: unifying the engineering equipment 3D model of the current engineering equipment and the real-time 3D model of the cargo hold into the same coordinate system;

[0060] S502: Define the placement position coordinate variables of the current engineering equipment as (x, y, z) and the rotation angle variables as (α, β, γ);

[0061] S503: Calculate the remaining available volume V in the cargo hold based on the real-time three-dimensional model of the cargo hold a The cargo hold is divided into m×n×p three-dimensional grids; the spatial volume V occupied by the current engineering equipment in the cargo hold is calculated based on the three-dimensional model of the engineering equipment and the rotation angle variable. c ; Calculate the space utilization objective function f1;

[0062] The formula is:

[0063]

[0064] Where: v ijk Represents the volume of a single grid cell (i, j, k); s ijk Indicates the occupation status of the grid unit (i, j, k) by the current engineering equipment, 0 means unoccupied, 1 means occupied;

[0065] S504: Calculate the weight w of each grid cell (i, j, k) based on the cargo hold weight distribution information of the cargo hold ijk ; Update the weight w′ of each grid cell (i, j, k) based on the weight information of the current engineering equipment ijk , and calculate the average weight of all grid cells after update Calculate the weight distribution uniformity objective function f2;

[0066] The formula is:

[0067]

[0068] Where: N represents the total number of grid cells;

[0069] S505: Obtain the equipment coordinates of other engineering equipment of the same project loaded in the cargo hold according to the project tag of the current engineering equipment Calculate the Euclidean distance d between the current engineering equipment and each other engineering equipment in the same project based on the placement coordinates (x, y, z) of the current engineering equipment j , calculate the objective function f3 of the degree of equipment aggregation in the same project;

[0070] The formula is:

[0071]

[0072] S505: Construct the objective function of the multi-objective optimization problem as follows:

[0073] minF(x,y,z,α,β,γ)=(1-f1,f2,f3).

[0074] Preferably, in step S505, the constraints of the multi-objective optimization problem include:

[0075] 1) Space constraints

[0076] The current engineering equipment does not extend beyond the cargo hold boundary;

[0077] The current engineering equipment does not collide with fixed obstacles or placed engineering equipment in the cargo hold;

[0078] 2) Rotation angle constraint

[0079] The rotation angle of the current engineering equipment does not exceed the allowable range;

[0080] 3) Weight constraints

[0081] The total weight of all engineering equipment loaded in the cargo hold shall not exceed the load-bearing capacity of the cargo hold.

[0082] Preferably, in step S6, the multi-objective optimization problem is solved by a genetic algorithm, and the specific solving steps include:

[0083] S601: The placement position coordinates (x, y, z) and rotation angles (α, β, γ) of the current engineering equipment are encoded as genes to form an initial population. Each chromosome in the initial population corresponds to a loading plan, including the placement position coordinates (x, y, z) and rotation angles (α, β, y).

[0084] S602: Define fitness function;

[0085] S603: Calculating the fitness function value of each chromosome using the fitness function;

[0086] S604: Selecting chromosomes to enter the next generation based on the fitness function value of each chromosome;

[0087] S605: Randomly select two chromosomes and exchange some of their genes to generate new chromosomes;

[0088] S606: randomly changing the value of a gene with a certain mutation probability, that is, randomly changing the loading position or direction of a certain engineering equipment or replacing the engineering equipment;

[0089] S607: Repeat steps S603 to S606 until the preset number of iterations is reached or the fitness function value changes less than the set threshold within several consecutive generations, then stop the iteration;

[0090] S608: The loading plan corresponding to the chromosome with the largest fitness function value is taken as the optimal loading plan.

[0091] Preferably, in step S602, the fitness function is expressed as:

[0092] F=ω1f1+ω2f2-ω3f3;

[0093] Where: ω1, ω2, ω3 represent the set weights.

[0094] Compared with the prior art, the intelligent loading scheduling optimization method for engineering equipment in the smart port of the present invention has the following beneficial effects:

[0095] The present invention obtains equipment-related information of the current engineering equipment and performs three-dimensional modeling to obtain a three-dimensional model of the engineering equipment. Simultaneously, it obtains real-time status information of the bulk carrier's cargo hold and models it to obtain a real-time three-dimensional model of the cargo hold. When constructing a multi-objective optimization problem based on these two three-dimensional models, it can fully consider the irregular, heterogeneous characteristics of the engineering equipment and the spatial matching relationship between the engineering equipment and the cargo hold, accurately estimate the space occupancy of the engineering equipment, and provide a data basis for subsequent engineering equipment loading and scheduling planning, which is conducive to improving the space utilization rate of the cargo hold when loading engineering equipment. During the loading process, the real-time status information of the cargo hold is obtained in real time. Each time new engineering equipment is loaded, the multi-objective optimization problem is reconstructed based on the latest cargo hold status and the optimal loading solution is solved. Through dynamic adjustment, the irregular remaining space in the cargo hold is fully utilized, avoiding the problem of subsequent cargo hold space waste caused by unreasonable early planning in traditional loading methods. The most suitable placement position for the new engineering equipment can be found based on the real-time status information, thereby further improving the space utilization rate of the cargo hold when loading engineering equipment.

[0096] When constructing a multi-objective optimization problem, the present invention not only takes into account the irregular heterogeneous characteristics of the engineering equipment and the real-time status of the cargo hold, but also fully integrates the weight information of the current engineering equipment and the weight distribution information of the cargo hold, so that the optimal loading plan that makes the weight distribution uniform can be solved. That is, the placement of the engineering equipment can be reasonably arranged according to the weight of the engineering equipment and the weight distribution of the cargo hold, so that the center of gravity of the bulk carrier is kept within a reasonable range, thereby improving the weight distribution uniformity of the cargo hold when the engineering equipment is loaded. At the same time, the multi-objective optimization strategy of the present invention takes weight distribution uniformity as one of the important goals, and comprehensively considers it with other goals (space utilization and distance from other engineering equipment in the same project). Through intelligent algorithms, it will seek a balance between multiple goals, thereby better improving the weight distribution uniformity of the cargo hold while ensuring space utilization and unloading efficiency.

[0097] When constructing a multi-objective optimization problem, the present invention fully considers the distance between engineering equipment of the same project based on the project label of the engineering equipment, avoiding the problem of increasing the cross-regional scheduling frequency of unloading machinery due to the dispersed storage of engineering equipment of the same project, reducing the complexity and time cost of unloading engineering equipment, thereby improving the efficiency of loading, unloading and transportation of smart port engineering equipment and reducing transportation costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0098] In order to make the purpose, technical solutions and advantages of the invention more clear, the present invention will be further described in detail below with reference to the accompanying drawings, in which:

[0099] Figure 1 This is the logical block diagram of the intelligent loading and scheduling optimization method for engineering equipment in smart ports. DETAILED DESCRIPTION

[0100] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the invention claimed for protection, but only represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0101] It should be noted that similar reference numerals and letters denote similar items in the following figures. Therefore, once an item is defined in one figure, it does not require further definition or explanation in subsequent figures. In the description of the present invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer" indicate positions or relationships based on the positions or relationships shown in the figures, or the positions or relationships in which the inventive product is typically placed when in use. These terms are intended solely to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or components referred to must have a specific orientation, be constructed, or operate in a specific orientation, and are therefore not to be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third," etc., are used solely to distinguish descriptions and are not to be construed as indicating or implying relative importance. Furthermore, terms such as "horizontal" and "vertical" do not imply that a component is absolutely horizontal or overhanging, but rather may be slightly tilted. For example, "horizontal" simply means that its direction is more horizontal than "vertical," and does not mean that the structure must be completely horizontal, but rather may be slightly tilted. In the description of the present invention, it should also be noted that, unless otherwise expressly specified or limited, the terms "disposed," "installed," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; they may refer to mechanical connections or electrical connections; they may refer to direct connections or indirect connections through an intermediate medium; and they may refer to internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0102] The following is a further detailed description through specific implementation methods:

[0103] Example:

[0104] This embodiment discloses a method for intelligent loading scheduling optimization of engineering equipment in a smart port.

[0105] like Figure 1 As shown, a method for intelligent loading scheduling optimization of engineering equipment in a smart port includes:

[0106] S1: When the engineering equipment is ready to be loaded, obtain the equipment-related information, weight information and project label of the current engineering equipment;

[0107] In this embodiment, the outer packaging of the engineering equipment is a rectangular structure, and the weight and project label of the engineering equipment are indicated on the outer packaging of the engineering equipment, so the weight information and project label can be directly obtained by scanning the label on the outer packaging of the engineering equipment.

[0108] S2: Perform three-dimensional modeling based on the equipment-related information of the current engineering equipment to obtain a three-dimensional model of the engineering equipment;

[0109] S3: Obtaining real-time status information and cargo hold weight distribution information of the bulk carrier;

[0110] S4: Modeling is performed based on the real-time status information of the cargo hold to obtain a real-time three-dimensional model of the cargo hold;

[0111] S5: Based on the real-time 3D model and weight distribution information of the cargo hold, as well as the 3D model, weight information, and project label of the current engineering equipment, a multi-objective optimization problem is constructed for the current engineering equipment, with the goals of maximizing cargo hold space utilization, maximizing weight distribution uniformity, and maximizing the degree of aggregation with other engineering equipment in the same project.

[0112] S6: Using intelligent algorithms to solve the multi-objective optimization problem of the current engineering equipment, an optimal loading plan is obtained; the optimal loading plan includes the placement position (coordinates) of the current engineering equipment in the cargo hold and the rotation angle during placement;

[0113] S7: Loading the current engineering equipment to a designated location in the cargo hold of the bulk carrier based on the optimal loading plan for the current engineering equipment;

[0114] S8: Repeat steps S1 to S7 until all engineering equipment is loaded or the cargo hold of the bulk carrier is fully loaded.

[0115] The present invention obtains equipment-related information of the current engineering equipment and performs three-dimensional modeling to obtain a three-dimensional model of the engineering equipment. Simultaneously, it obtains real-time status information of the bulk carrier's cargo hold and models it to obtain a real-time three-dimensional model of the cargo hold. When constructing a multi-objective optimization problem based on these two three-dimensional models, it can fully consider the irregular, heterogeneous characteristics of the engineering equipment and the spatial matching relationship between the engineering equipment and the cargo hold, accurately estimate the space occupancy of the engineering equipment, and provide a data basis for subsequent engineering equipment loading and scheduling planning, which is conducive to improving the space utilization rate of the cargo hold when loading engineering equipment. During the loading process, the real-time status information of the cargo hold is obtained in real time. Each time new engineering equipment is loaded, the multi-objective optimization problem is reconstructed based on the latest cargo hold status and the optimal loading solution is solved. Through dynamic adjustment, the irregular remaining space in the cargo hold is fully utilized, avoiding the problem of subsequent cargo hold space waste caused by unreasonable early planning in traditional loading methods. The most suitable placement position for the new engineering equipment can be found based on the real-time status information, thereby further improving the space utilization rate of the cargo hold when loading engineering equipment.

[0116] When constructing a multi-objective optimization problem, the present invention not only takes into account the irregular heterogeneous characteristics of the engineering equipment and the real-time status of the cargo hold, but also fully integrates the weight information of the current engineering equipment and the weight distribution information of the cargo hold, so that the optimal loading plan that makes the weight distribution uniform can be solved. That is, the placement of the engineering equipment can be reasonably arranged according to the weight of the engineering equipment and the weight distribution of the cargo hold, so that the center of gravity of the bulk carrier is kept within a reasonable range, thereby improving the weight distribution uniformity of the cargo hold when the engineering equipment is loaded. At the same time, the multi-objective optimization strategy of the present invention takes weight distribution uniformity as one of the important goals, and comprehensively considers it with other goals (space utilization and distance from other engineering equipment in the same project). Through intelligent algorithms, it will seek a balance between multiple goals, thereby better improving the weight distribution uniformity of the cargo hold while ensuring space utilization and unloading efficiency.

[0117] When constructing a multi-objective optimization problem, the present invention fully considers the distance between engineering equipment of the same project based on the project label of the engineering equipment, avoiding the problem of increasing the cross-regional scheduling frequency of unloading machinery due to the dispersed storage of engineering equipment of the same project, reducing the complexity and time cost of unloading engineering equipment, thereby improving the efficiency of loading, unloading and transportation of smart port engineering equipment and reducing transportation costs.

[0118] In order to better introduce the technical solution of the present invention, this embodiment is described through the following parts.

[0119] 1. 3D model of engineering equipment

[0120] The three-dimensional model of engineering equipment is obtained through the following steps:

[0121] S201: Equipment related information includes point cloud data, multi-angle photos and external dimension data of engineering equipment;

[0122] In this embodiment, a high-precision 3D laser scanner is used to perform an omnidirectional scan of each piece of engineering equipment, acquiring point cloud data of the equipment's surface. A high-resolution camera is used to capture multi-angle photographs of the equipment. Sensors are used to measure the equipment's dimensions (e.g., length, width, height, and diameter), and their weight is recorded. Weight information can be obtained by direct weighing or by reading the weight label on the equipment's packaging.

[0123] S202: performing point cloud denoising, point cloud registration, and point cloud simplification on the point cloud data of the engineering equipment to obtain a point cloud model of the engineering equipment;

[0124] In this embodiment, point cloud denoising uses a statistical filtering algorithm to remove outliers and noise points from the point cloud, improving the quality of the point cloud data. Point cloud registration combines point cloud data acquired from different scanning angles to form a complete point cloud model. During the registration process, an iterative closest point (ICP) algorithm is used to continuously optimize the correspondence between point clouds to achieve precise registration. Point cloud simplification streamlines the registered point cloud data using a voxel grid downsampling algorithm. While ensuring the accuracy of the point cloud model, it reduces the amount of point cloud data and improves the efficiency of subsequent processing.

[0125] S203: extracting features from the multi-angle photos of the engineering equipment to obtain feature points of the photos at each angle; matching the feature points of the photos at each angle to obtain a photo feature point matching result;

[0126] In this example, computer vision techniques are first used to extract features from each photo, obtaining unique and stable local image feature points such as corners, edges, and spots. Feature points from different photos are then matched, and the overlap and relative positional relationships between the photos are determined by establishing correspondences between the feature points.

[0127] S204: Based on the photo feature point matching results, a reference photo (one with the richest feature points and the widest coverage area) is selected from the multi-angle photos, and the other photos are geometrically corrected using the reference photo; the color information and texture information in the geometrically corrected angled photo are fused with the engineering equipment point cloud model, and the color information and texture information of points in the engineering equipment point cloud model that do not directly correspond to photo pixels are interpolated using an interpolation algorithm to obtain a color point cloud model that fuses color and texture;

[0128] In this embodiment, the color information and texture information in the photo are fused with the point cloud model based on the matching results of the feature points in the photo; for each point in the point cloud model, its corresponding pixel position in the photo is found, and the color and texture information of the pixel is assigned to the point; for overlapping areas, a weighted average is performed based on the matching quality of the feature points and the quality of the photo to obtain a more accurate fusion result.

[0129] S205: Surface reconstruction is performed on the color point cloud model that integrates color and texture using a surface reconstruction algorithm to generate a three-dimensional mesh model of the engineering equipment;

[0130] S206: Performing size calibration and physical property assignment on the three-dimensional mesh model of the engineering equipment according to the external dimension data and weight information of the engineering equipment to obtain the three-dimensional model of the engineering equipment.

[0131] In this embodiment, the size of the three-dimensional mesh model can be calibrated and physical properties can be assigned based on the external dimensions and weight information. The vertex positions of the mesh model can also be adjusted to make its size consistent with the actual measurement value.

[0132] This method integrates the color and texture information from engineering equipment photos with the engineering equipment point cloud model, significantly improving the realism and detail of the 3D model. This gives the 3D model not only precise geometry but also a realistic appearance, more closely resembling the visual effects of actual engineering equipment. This provides a more accurate and vivid foundation for subsequent loading scheduling and simulation displays. Simultaneously, the 3D model is calibrated and attributes such as dimensions and weight are assigned to facilitate subsequent calculations and planning related to engineering equipment loading.

[0133] 2. Cargo hold weight distribution information

[0134] Obtain cargo hold weight distribution information by following the steps below:

[0135] S301: Install several weight sensors at the bottom of the cargo hold, determine the installation location information of each weight sensor, and set a sensor identifier for each weight sensor;

[0136] S302: Collect weight values at corresponding positions through each weight sensor to obtain a data set containing a number of data points; each data point includes a corresponding sensor identifier, weight value, and collection time;

[0137] S303: Perform data cleaning, data alignment, and unit unification on the data set to obtain a preprocessed data set;

[0138] In this embodiment, data cleaning involves checking for outliers in the data set, such as weight measurements that clearly fall outside a reasonable range. Outliers can be corrected using interpolation or the average of adjacent sensor data, or simply removed. Data alignment involves time-aligning the data. Unit unification involves converting all weight measurements to a standard unit (e.g., kg).

[0139] S304: Establish a two-dimensional coordinate system for the cargo hold with a corner point of the cargo hold as the origin and the length and width directions as coordinate axes;

[0140] S305: Determine the coordinates of each weight sensor in the cargo hold two-dimensional coordinate system based on the installation location information of each weight sensor, and construct a mapping table to associate the sensor identifier of each weight sensor with its coordinates in the cargo hold coordinate system;

[0141] S306: Divide the cargo hold into n×m small intervals along the length and width directions of the cargo hold, and construct a two-dimensional cargo hold weight distribution matrix of size n×m; the small intervals correspond one-to-one to the matrix cells in the cargo hold weight distribution matrix;

[0142] S307: Traverse each data point in the preprocessed data set, obtain the corresponding coordinates from the mapping table according to the sensor identifier of the data point; determine the matrix cell where the coordinates corresponding to the data point are located, and accumulate the weight value of the data point to the corresponding matrix cell;

[0143] In this embodiment, the number of weight sensors does not correspond one to one to the cell intervals. Therefore, if multiple sensors correspond to the same matrix unit, their weight values are added together. An interpolation algorithm (such as bilinear interpolation) can also be used to smooth the weight values in the matrix to make the weight distribution more continuous.

[0144] S308: After processing all data points in step S307, the obtained cargo hold weight distribution matrix is used as cargo hold weight distribution information; each element in the cargo hold weight distribution matrix represents the weight of the corresponding small area in the cargo hold.

[0145] 3. Real-time 3D model of cargo hold

[0146] The real-time 3D model of the cargo hold is obtained through the following steps:

[0147] S401: Real-time status information of the bulk carrier's cargo hold includes 3D point cloud data and photo images of loaded engineering equipment, as well as geometric dimension data of the cargo hold;

[0148] S402: performing point cloud preprocessing on the three-dimensional point cloud data of the loaded engineering equipment to obtain preprocessed three-dimensional point cloud data; performing point cloud feature extraction on the preprocessed three-dimensional point cloud data to obtain point cloud feature points;

[0149] In this embodiment, the point cloud data preprocessing includes denoising and downsampling the three-dimensional point cloud data of the loaded engineering equipment.

[0150] Point cloud feature extraction uses a normal vector-based feature extraction method to calculate the normal vector of each point and identify feature points based on the degree of change in the normal vector. For example, a normal vector change threshold is set. When the angle between the normal vector of a point and the points in its neighborhood exceeds the threshold, the point is identified as a feature point.

[0151] S403: performing image preprocessing on the photo image of the loaded engineering equipment to obtain preprocessed image data; performing image feature extraction on the preprocessed image data to obtain image feature points;

[0152] In this embodiment, image preprocessing includes grayscale conversion, image contrast enhancement, and edge information extraction.

[0153] Image feature extraction can use the SIFT (Scale Invariant Feature Transform) algorithm to extract image feature points and calculate the feature descriptor of each feature point.

[0154] S404: Matching the point cloud feature points with the image feature points to obtain successfully matched point cloud-image feature point pairs;

[0155] In this embodiment, a feature descriptor-based matching method is used to calculate the similarity between the point cloud feature descriptor and the image feature descriptor, and Euclidean distance is used as the similarity metric. When the similarity is less than a certain threshold, it is determined that the match is successful.

[0156] S405: Constructing a point cloud coordinate system based on the pre-processed 3D point cloud data; constructing a 3D coordinate system for the cargo hold with a corner point of the cargo hold as the origin, the length direction of the cargo hold as the X-axis, the width direction as the Y-axis, and the height direction as the Z-axis;

[0157] S406: Calculating a coordinate transformation matrix between the point cloud coordinate system and the cargo hold three-dimensional coordinate system based on the point cloud-image feature point pairs, the coordinates of the point cloud feature points in the point cloud coordinate system, and the coordinates of the image feature points in the cargo hold three-dimensional coordinate system;

[0158] S407: Generate a three-dimensional model framework of the cargo hold based on the geometric dimension data of the cargo hold;

[0159] S408: All point cloud data in the pre-processed three-dimensional point cloud data are converted to the cargo hold three-dimensional coordinate system through the coordinate conversion matrix; the converted point cloud data are integrated with the cargo hold three-dimensional model framework, and the engineering equipment shapes corresponding to the point cloud data are filled into the corresponding positions in the cargo hold three-dimensional model framework to obtain a preliminary cargo hold three-dimensional model that integrates the information of the loaded engineering equipment; the preliminary cargo hold three-dimensional model is reconstructed to obtain a real-time three-dimensional model of the cargo hold.

[0160] Specifically, the point cloud coordinate system is constructed through the following steps:

[0161] S4051: Calculate the centroid of preprocessed 3D point cloud data

[0162] S4052: Calculate the centroid of each point in the preprocessed 3D point cloud data distance, and filter out the distance centroid The farthest target point p f ;

[0163] S4053: Centroid As the starting point and the target point p f Get a preliminary direction vector v for the end point i ; through the preliminary direction vector v i The projection of the preprocessed 3D point cloud data is divided into two regions, and the centroid of the point cloud data in the two regions is calculated respectively. and Connect centroids and Get the auxiliary direction vector v a ;

[0164] In this embodiment, when dividing the area, the initial direction vector v from each point is calculated. i The projection length is used to divide the point cloud data into two areas according to the positive and negative projection lengths.

[0165] S4054: Calculate the covariance matrix of the preprocessed point cloud data and perform eigenvalue decomposition on the covariance matrix to obtain eigenvalues λ1≥λ2≥λ3 and their eigenvalue vectors v1, v2, and v3; the size of the eigenvalue represents the degree of discreteness of the point cloud data in the direction of the eigenvector, and the eigenvector represents the main direction of the point cloud data.

[0166] S4055: According to the preliminary direction vector v i and auxiliary direction vector v a Adjust the eigenvector to get the X coordinate of the point cloud system p 、Y p , Z p Axis direction vector;

[0167] In this embodiment, select v i The eigenvector with the smallest angle is taken as X p The candidate direction of the axis, combined with v a Further determine X p The final direction of the axis, ensure that X p The Y axis can better reflect the main extension direction of the point cloud. The same method can be used to determine the Y p and Zp Axis direction.

[0168] S4056: Based on centroid To X p 、Y p , Z p Fine-tune the X-axis direction and integrate p 、Y p , Z p The reference point in the axis direction determines the origin of the point cloud coordinate system O p ;

[0169] In this embodiment, along X p Axis direction, calculate the point cloud data in X p The boundary points on the positive and negative directions of the axis are taken as the midpoint of the two boundary points p The same method can be used to refer to the reference point in the Y axis direction. p and Z p Similar operations are performed in the axial direction.

[0170] S4057: Based on the origin of the point cloud coordinate system O p and X p 、Y p , Z p The point cloud coordinate system is constructed by the axis direction vector.

[0171] Specifically, the coordinate transformation matrix is constructed through the following steps:

[0172] S4061: screening and normalizing the point cloud-image feature point pairs to obtain pre-processed feature point pairs;

[0173] S4062: Estimate the initial rotation matrix R0 and translation vector T0 using the singular value decomposition algorithm;

[0174] In this embodiment, the existing means are used to estimate the initial rotation matrix and translation vector through the singular value decomposition algorithm.

[0175] S4063: Iteratively optimize the initial rotation matrix R0 and translation vector T0 using an iterative closest point algorithm to obtain an optimized rotation matrix R and translation vector T;

[0176] In this embodiment, let the current rotation matrix be R k , the translation vector is T k , in the K+1th iteration, for each point cloud feature point P i , calculate its corresponding point P in the image coordinate system i,v =R k P i +T k ;

[0177] Calculation error By solving E k The smallest R k+1 and T k+1 To update the transformation matrix. Quaternion methods or other optimization algorithms can be used to solve the optimal rotation matrix and translation vector. Set an iteration termination condition, such as when the error change is less than a certain threshold or when the maximum number of iterations is reached. The iteration stops when the termination condition is met.

[0178] S4064: Calculate the coordinate transformation matrix M = (R, T) based on the rotation matrix R and the translation vector T.

[0179] 4. Multi-objective optimization problem

[0180] In this embodiment, a multi-objective optimization problem is constructed through the following steps:

[0181] S501: unifying the engineering equipment 3D model of the current engineering equipment and the real-time 3D model of the cargo hold into the same coordinate system;

[0182] S502: Define the placement position coordinate variables of the current engineering equipment as (x, y, z) and the rotation angle variables as (α, β, γ);

[0183] S503: Calculate the remaining available volume V in the cargo hold based on the real-time three-dimensional model of the cargo hold a The cargo hold is divided into m = n × p three-dimensional grids; the space volume V occupied by the current engineering equipment in the cargo hold is calculated based on the three-dimensional model of the engineering equipment and the rotation angle variable. c ; Calculate the space utilization objective function f1;

[0184] The formula is:

[0185]

[0186] Where: v ijk Represents the volume of a single grid cell (i, j, k); s ijk Indicates the occupation status of the grid unit (i, j, k) by the current engineering equipment, 0 means unoccupied, 1 means occupied;

[0187] S504: Calculate the weight w of each grid cell (i, j, k) based on the cargo hold weight distribution information of the cargo hold ijk ; Update the weight w′ of each grid cell (i, j, k) based on the weight information of the current engineering equipment ijk , and calculate the average weight of all grid cells after update Calculate the weight distribution uniformity objective function f2;

[0188] In this embodiment, the cargo hold weight distribution information (a two-dimensional weight matrix) is used to calculate the weight of each grid cell using conventional methods. This involves layering along the z-axis of the cargo hold coordinate system and initializing the two-dimensional weight matrix as a layer of a three-dimensional matrix. When a new device is placed, the grid weight of the relevant layer in the three-dimensional weight matrix is dynamically updated based on the device's position, rotation angle, and height.

[0189] The formula is:

[0190]

[0191] Where: N represents the total number of grid cells;

[0192] S505: Obtain the equipment coordinates of other engineering equipment of the same project loaded in the cargo hold according to the project tag of the current engineering equipment Calculate the Euclidean distance d between the current engineering equipment and each other engineering equipment in the same project based on the placement coordinates (x, y, z) of the current engineering equipment j , calculate the objective function f3 of the degree of equipment aggregation in the same project;

[0193] The formula is:

[0194]

[0195] S505: Construct the objective function of the multi-objective optimization problem as follows:

[0196] minF(x,y,z,α,β,γ)=(1-f1,f2,f3).

[0197] The constraints of the multi-objective optimization problem include:

[0198] 1) Space constraints

[0199] The current engineering equipment does not extend beyond the cargo hold boundary;

[0200] The current engineering equipment does not collide with fixed obstacles or placed engineering equipment in the cargo hold;

[0201] 2) Rotation angle constraint

[0202] The rotation angle of the current engineering equipment does not exceed the allowable range;

[0203] 3) Weight constraints

[0204] The total weight of all engineering equipment loaded in the cargo hold shall not exceed the load-bearing capacity of the cargo hold.

[0205] When constructing a multi-objective optimization problem, the invention not only takes into account the irregular heterogeneous characteristics of the engineering equipment and the real-time status of the cargo hold, but also fully integrates the weight information of the current engineering equipment and the weight distribution information of the cargo hold, so that the optimal loading plan that makes the weight distribution uniform can be solved. That is, the placement of the engineering equipment can be reasonably arranged according to the weight of the engineering equipment and the weight distribution of the cargo hold, so that the center of gravity of the bulk carrier is kept within a reasonable range, thereby improving the uniformity of the weight distribution of the cargo hold when the engineering equipment is loaded. At the same time, the multi-objective optimization strategy of the present invention takes weight distribution uniformity as one of the important goals, and comprehensively considers it with other goals (space utilization and the distance to other engineering equipment in the same project). Through intelligent algorithms, it will seek a balance between multiple goals, thereby better improving the uniformity of the weight distribution of the cargo hold while ensuring space utilization and unloading efficiency.

[0206] When constructing a multi-objective optimization problem, the present invention fully considers the distance between engineering equipment of the same project based on the project label of the engineering equipment, avoiding the problem of increasing the cross-regional scheduling frequency of unloading machinery due to the dispersed storage of engineering equipment of the same project, reducing the complexity and time cost of unloading engineering equipment, thereby improving the efficiency of loading, unloading and transportation of smart port engineering equipment and reducing transportation costs.

[0207] 5. Genetic Algorithm

[0208] The genetic algorithm is used to solve the multi-objective optimization problem. The specific solution steps include:

[0209] S601: The placement position coordinates (x, y, z) and rotation angles (α, β, γ) of the current engineering equipment are encoded as genes to form an initial population. Each chromosome in the initial population corresponds to a loading plan, including the placement position coordinates (x, y, z) and rotation angles (α, β, γ).

[0210] S602: Define fitness function;

[0211] The formula of the fitness function is expressed as:

[0212] F=ω1f1+ω2f2-ω3f3;

[0213] Where: ω1, ω2, ω3 represent the set weights.

[0214] S603: Calculating the fitness function value of each chromosome using the fitness function;

[0215] S604: Selecting chromosomes to enter the next generation based on the fitness function value of each chromosome;

[0216] S605: Randomly select two chromosomes and exchange some of their genes to generate new chromosomes;

[0217] S606: randomly changing the value of a gene with a certain mutation probability, that is, randomly changing the loading position or direction of a certain engineering equipment or replacing the engineering equipment;

[0218] S607: Repeat steps S603 to S606 until the preset number of iterations is reached or the fitness function value changes less than the set threshold within several consecutive generations, then stop the iteration;

[0219] S608: The loading plan corresponding to the chromosome with the largest fitness function value is taken as the optimal loading plan.

[0220] The present invention solves the constructed multi-objective optimization problem through genetic algorithm. The population iteration characteristics of genetic algorithm can explore thousands of loading schemes at the same time, and automatically screen out the Pareto frontier solution set through non-dominated sorting and congestion calculation, which can ensure the optimal loading scheme with respect to unloading efficiency, space utilization and hull stability.

[0221] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the technical solutions. Those skilled in the art should understand that modifications or equivalent replacements of the technical solutions of the present invention that do not depart from the purpose and scope of the technical solutions of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A smart port engineering equipment intelligent loading scheduling optimization method, characterized in that: include: S1: Obtain equipment-related information, weight information, and project tags of the current engineering equipment; S2: Perform three-dimensional modeling based on the equipment-related information of the current engineering equipment to obtain a three-dimensional model of the engineering equipment; S3: Obtaining real-time status information and cargo hold weight distribution information of the bulk carrier; S4: Modeling is performed based on the real-time status information of the cargo hold to obtain a real-time three-dimensional model of the cargo hold; S5: Based on the real-time 3D model and weight distribution information of the cargo hold, as well as the 3D model, weight information, and project label of the current engineering equipment, a multi-objective optimization problem is constructed for the current engineering equipment, with the goals of maximizing cargo hold space utilization, maximizing weight distribution uniformity, and maximizing the degree of aggregation with other engineering equipment in the same project. S6: Solve the multi-objective optimization problem of the current engineering equipment through intelligent algorithms to obtain the optimal loading plan; the optimal loading plan includes the placement of the current engineering equipment in the cargo hold; S7: Loading the current engineering equipment to a corresponding position in the cargo hold based on the optimal loading plan for the current engineering equipment; S8: Repeat steps S1 to S7 until all engineering equipment is loaded or the cargo hold of the bulk carrier is fully loaded.

2. The intelligent loading scheduling optimization method for engineering equipment in a smart port according to claim 1 is characterized in that: In step S2, a three-dimensional model of the engineering equipment is obtained by modeling through the following steps: S201: Equipment related information includes point cloud data, multi-angle photos and external dimension data of engineering equipment; S202: performing point cloud denoising, point cloud registration, and point cloud simplification on the point cloud data of the engineering equipment to obtain a point cloud model of the engineering equipment; S203: extracting features from the multi-angle photos of the engineering equipment to obtain feature points of the photos at each angle; matching the feature points of the photos at each angle to obtain a photo feature point matching result; S204: Selecting a reference photo from the multi-angle photos based on the photo feature point matching results, and geometrically correcting the other photos using the reference photo; fusing the color information and texture information in the geometrically corrected angled photos with the engineering equipment point cloud model, and interpolating the color information and texture information of points in the engineering equipment point cloud model that do not directly correspond to pixels in the photo using an interpolation algorithm to obtain a color point cloud model that fuses color and texture; S205: Surface reconstruction is performed on the color point cloud model that integrates color and texture using a surface reconstruction algorithm to generate a three-dimensional mesh model of the engineering equipment; S206: Performing size calibration and physical property assignment on the three-dimensional mesh model of the engineering equipment according to the external dimension data and weight information of the engineering equipment to obtain the three-dimensional model of the engineering equipment.

3. The intelligent loading scheduling optimization method for engineering equipment in a smart port according to claim 1 is characterized in that: In step S3, cargo hold weight distribution information is obtained through the following steps: S301: Install several weight sensors at the bottom of the cargo hold, determine the installation location information of each weight sensor, and set a sensor identifier for each weight sensor; S302: Collect weight values at corresponding positions through each weight sensor to obtain a data set containing a number of data points; each data point includes a corresponding sensor identifier, weight value, and collection time; S303: Perform data cleaning, data alignment, and unit unification on the data set to obtain a preprocessed data set; S304: Establish a two-dimensional coordinate system for the cargo hold with a corner point of the cargo hold as the origin and the length and width directions as coordinate axes; S305: Determine the coordinates of each weight sensor in the cargo hold two-dimensional coordinate system based on the installation location information of each weight sensor, and construct a mapping table to associate the sensor identifier of each weight sensor with its coordinates in the cargo hold coordinate system; S306: Divide the cargo hold into n×m small intervals along the length and width directions of the cargo hold, and construct a two-dimensional cargo hold weight distribution matrix of size n×m; the small intervals correspond one-to-one to the matrix cells in the cargo hold weight distribution matrix; S307: Traverse each data point in the preprocessed data set, obtain the corresponding coordinates from the mapping table according to the sensor identifier of the data point; determine the matrix cell where the coordinates corresponding to the data point are located, and accumulate the weight value of the data point to the corresponding matrix cell; S308: After processing all data points in step S307, the obtained cargo hold weight distribution matrix is used as cargo hold weight distribution information; each element in the cargo hold weight distribution matrix represents the weight of the corresponding small area in the cargo hold.

4. The intelligent loading scheduling optimization method for engineering equipment in a smart port according to claim 1 is characterized in that: In step S4, a real-time three-dimensional model of the cargo hold is obtained by modeling through the following steps: S401: Real-time status information of the bulk carrier's cargo hold includes 3D point cloud data and photo images of loaded engineering equipment, as well as geometric dimension data of the cargo hold; S402: performing point cloud preprocessing on the three-dimensional point cloud data of the loaded engineering equipment to obtain preprocessed three-dimensional point cloud data; Perform point cloud feature extraction on the pre-processed three-dimensional point cloud data to obtain point cloud feature points; S403: performing image preprocessing on the photo image of the loaded engineering equipment to obtain preprocessed image data; Perform image feature extraction on the preprocessed image data to obtain image feature points; S404: Matching the point cloud feature points with the image feature points to obtain successfully matched point cloud-image feature point pairs; S405: Constructing a point cloud coordinate system based on the pre-processed 3D point cloud data; constructing a 3D coordinate system for the cargo hold with a corner point of the cargo hold as the origin, the length direction of the cargo hold as the X-axis, the width direction as the Y-axis, and the height direction as the Z-axis; S406: Calculating a coordinate transformation matrix between the point cloud coordinate system and the cargo hold three-dimensional coordinate system based on the point cloud-image feature point pairs, the coordinates of the point cloud feature points in the point cloud coordinate system, and the coordinates of the image feature points in the cargo hold three-dimensional coordinate system; S407: Generate a three-dimensional model framework of the cargo hold based on the geometric dimension data of the cargo hold; S408: All point cloud data in the pre-processed three-dimensional point cloud data are converted into the cargo hold three-dimensional coordinate system through the coordinate conversion matrix; the converted point cloud data are integrated with the cargo hold three-dimensional model framework to obtain a preliminary cargo hold three-dimensional model that integrates the information of the loaded engineering equipment; the surface of the preliminary cargo hold three-dimensional model is reconstructed to obtain a real-time three-dimensional model of the cargo hold.

5. The intelligent loading scheduling optimization method for engineering equipment in a smart port according to claim 4 is characterized in that: In step S405, a point cloud coordinate system is constructed by the following steps: S4051: Calculate the centroid of preprocessed 3D point cloud data ; S4052: Calculate the centroid of each point in the preprocessed 3D point cloud data distance, and filter out the distance centroid The farthest target point p f ; S4053: Centroid As the starting point and the target point p f Get a preliminary direction vector v for the end point i ; through the preliminary direction vector v i The projection of the preprocessed 3D point cloud data is divided into two regions, and the centroid of the point cloud data in the two regions is calculated respectively. and ; Connect centroids and Get the auxiliary direction vector v a ; S4054: Calculate the covariance matrix of the preprocessed point cloud data and perform eigenvalue decomposition on the covariance matrix to obtain eigenvalues λ1≥λ2≥λ3 and their eigenvalue vectors v1, v2, and v3; S4055: According to the preliminary direction vector v i and auxiliary direction vector v a Adjust the eigenvector to get the X coordinate of the point cloud system p 、Y p , Z p Axis direction vector; S4056: Comprehensive X p 、Y p , Z p The reference point in the axis direction determines the origin of the point cloud coordinate system O p ; S4057: Based on the origin of the point cloud coordinate system O p and X p 、Y p , Z p The point cloud coordinate system is constructed by the axis direction vector.

6. The intelligent loading scheduling optimization method for engineering equipment in a smart port according to claim 4 is characterized in that: In step S406, a coordinate transformation matrix is constructed by the following steps: S4061: screening and normalizing the point cloud-image feature point pairs to obtain pre-processed feature point pairs; S4062: Estimate the initial rotation matrix R0 and translation vector T0 using the singular value decomposition algorithm; S4063: Iteratively optimize the initial rotation matrix R0 and translation vector T0 using an iterative closest point algorithm to obtain an optimized rotation matrix R and translation vector T; S4064: Calculate the coordinate transformation matrix M = (R, T) based on the rotation matrix R and the translation vector T.

7. The intelligent loading scheduling optimization method for engineering equipment in a smart port according to claim 1 is characterized in that: In step S5, a multi-objective optimization problem is constructed by the following steps: S501: unifying the engineering equipment 3D model of the current engineering equipment and the real-time 3D model of the cargo hold into the same coordinate system; S502: Define the placement position coordinate variables of the current engineering equipment as (x, y, z) and the rotation angle variables as (α, β, γ); S503: Calculate the remaining available volume V in the cargo hold based on the real-time three-dimensional model of the cargo hold a The cargo hold is divided into m×n×p three-dimensional grids; the spatial volume V occupied by the current engineering equipment in the cargo hold is calculated based on the three-dimensional model of the engineering equipment and the rotation angle variable. c ; Calculate the space utilization objective function f1; The formula is: Where: v ijk Represents the volume of a single grid cell (i, j, k); s ijk Indicates the occupation status of the grid unit (i, j, k) by the current engineering equipment, 0 means unoccupied, 1 means occupied; S504: Calculate the weight w of each grid cell (i, j, k) based on the cargo hold weight distribution information of the cargo hold ijk ; Update the weight w′ of each grid cell (i, j, k) based on the weight information of the current engineering equipment ijk , and calculate the average weight of all grid cells after update Calculate the weight distribution uniformity objective function f2; The formula is: Where: N represents the total number of grid cells; S505: Obtain the equipment coordinates of other engineering equipment of the same project loaded in the cargo hold according to the project tag of the current engineering equipment Calculate the Euclidean distance d between the current engineering equipment and each other engineering equipment in the same project based on the placement coordinates (x, y, z) of the current engineering equipment j , calculate the objective function f3 of the degree of equipment aggregation in the same project; The formula is: S505: Construct the objective function of the multi-objective optimization problem as follows: min F(x,y,z,α,β,γ)=(1-f1,f2,f3).

8. The intelligent loading scheduling optimization method for engineering equipment in a smart port according to claim 7, characterized in that: In step S505, the constraints of the multi-objective optimization problem include: 1) Space constraints The current engineering equipment does not extend beyond the cargo hold boundary; The current engineering equipment does not collide with fixed obstacles or placed engineering equipment in the cargo hold; 2) Rotation angle constraint The rotation angle of the current engineering equipment does not exceed the allowable range; 3) Weight constraints The total weight of all engineering equipment loaded in the cargo hold shall not exceed the load-bearing capacity of the cargo hold.

9. The intelligent loading scheduling optimization method for engineering equipment in a smart port according to claim 8, characterized in that: In step S6, the multi-objective optimization problem is solved by a genetic algorithm. The specific solving steps include: S601: The placement position coordinates (x, y, z) and rotation angles (α, β, γ) of the current engineering equipment are encoded as genes to form an initial population. Each chromosome in the initial population corresponds to a loading plan, including the placement position coordinates (x, y, z) and rotation angles (α, β, γ). S602: Define fitness function; S603: Calculating the fitness function value of each chromosome using the fitness function; S604: Selecting chromosomes to enter the next generation based on the fitness function value of each chromosome; S605: Randomly select two chromosomes and exchange some of their genes to generate new chromosomes; S606: randomly changing the value of a gene with a certain mutation probability, that is, randomly changing the loading position or direction of a certain engineering equipment or replacing the engineering equipment; S607: Repeat steps S603 to S606 until the preset number of iterations is reached or the fitness function value changes less than the set threshold within several consecutive generations, then stop the iteration; S608: The loading plan corresponding to the chromosome with the largest fitness function value is taken as the optimal loading plan.

10. The intelligent loading scheduling optimization method for engineering equipment in a smart port according to claim 9, characterized in that: In step S602, the fitness function is expressed as: F=ω1f1+ω2f2-ω3f3; Where: ω1, ω2, ω3 represent the set weights.

Images