Digital twinborn simulation method and system for container terminal storage yard management and control and storage medium

Abstract

The invention discloses a container terminal storage yard management and control digital twinborn simulation method and system and a storage medium. The method comprises the steps of obtaining production data, related to storage yard management and control, of a current container terminal, and generating an initial state of a digital twin platform; according to the historical operation rule of the container wharf, container entry random event probability distribution is set in the digital twin platform; simulating an approach container position selection process based on a digital twin platform, and simulating a random event according to the set probability distribution; and selecting positions of entering containers through the container wharf storage yard management and control gravitational field model. According to the method, the defect that a container terminal storage yard management and control digital twinborn system is lack of future storage yard operation state mapping can be overcome through random event generation in cooperation with a storage yard management and control gravitational field model.

Description

Technical Field

[0001] The present invention relates to the technical field of container terminal management, and in particular to a digital twin simulation method, system and storage medium for container terminal yard management and control. Background Art

[0002] Digital twin technology is an innovative simulation method that simulates and analyzes the behavior of physical entities by building accurate virtual copies of physical entities. This technology has shown broad application potential in many fields such as industrial automation, intelligent manufacturing, and urban infrastructure management. The core advantage of the digital twin system is that it can accurately map physical entities in digital space, capture and reflect subtle changes in the state of entities in real time, and especially demonstrates its outstanding application value in the fast-paced manufacturing industry.

[0003] However, for complex systems such as container terminals, due to their discreteness and uncertainty, the current digital twin system for container terminal yard management can only achieve three-dimensional visualization rendering of the current state of physical entities, but for the random and discrete events in the production process, as well as the dynamic changes in yard management, these systems are often difficult to effectively map. This limitation leads to a key problem: the future state changes of the container terminal yard - which is crucial for production operations - cannot be accurately presented in the digital twin system. This is also the main defect of the current digital twin system for container terminal yards. Summary of the invention

[0004] In view of this, the purpose of the present invention is to propose a digital twin simulation method, system and storage medium for container terminal yard management and control, which can solve the shortcoming of the digital twin system for container terminal yard management and control that lacks mapping of future yard operation status through random event generation and coordination with the yard management and control gravitational field model.

[0005] According to one aspect of the present invention, a digital twin simulation method for container terminal yard management and control is provided, the method comprising:

[0006] Obtain production data related to current container terminal and yard management and control, and generate the initial state of the digital twin platform;

[0007] According to the historical operation rules of the container terminal, the probability distribution of random events of container entry is set in the digital twin platform;

[0008] The digital twin platform is used to simulate the process of container location selection, and random events are simulated according to the set probability distribution;

[0009] The incoming containers are positioned using the container terminal yard management and control gravity field model.

[0010] In the above technical solution, the changes in yard management and control in the future are simulated in the digital space to realize the mapping of the future state of the container terminal yard operation. The present invention can solve the defect that the traditional container terminal yard digital twin system cannot map the future operation state of the yard management and control.

[0011] In some embodiments, the probability distribution of random events includes: probability distribution of container arrival time, probability distribution of quay crane loading and unloading operation time, probability distribution of yard crane operation time, and probability distribution of container truck moving speed;

[0012] In the above technical solution, the purpose of selecting these probability distributions (arrival operation time of container entering the gate, loading and unloading operation time of quay crane, operation time of yard crane, moving speed of container truck) is to reflect the randomness and complexity in the system as realistically as possible when simulating the operation of container terminal yard. The use of these probability distributions provides an important basis for the prediction of future operation status. The selection and advantages of each probability distribution are as follows:

[0013] 1. Probability distribution of container arrival time

[0014] Reason for selection: The time when a container arrives at the terminal is random and is affected by multiple uncertain factors such as logistics arrangements and traffic conditions. Therefore, probability distribution can more realistically simulate the time pattern of container entry into the gate in actual operations.

[0015] Advantages: This distribution can help accurately predict the frequency and time period of container arrivals at different times, avoiding over- or underestimation of future yard loads.

[0016] Helps the overall solution: Enables the yard management system to predict future container arrivals and arrange yard resources (such as yard space and human resources) in advance, thereby improving operational efficiency.

[0017] 2. Probability distribution of quay crane loading and unloading operation duration

[0018] Reason for selection: The time it takes for a quay crane to load and unload containers is affected by factors such as the performance of the loading and unloading equipment and the loading conditions of the ship, and is uncertain, so it needs to be modeled through probability distribution.

[0019] Advantages: The probability distribution can be used to estimate the time variation range of quay crane loading and unloading operations, which helps to more accurately predict the time it takes to complete loading and unloading operations after the ship docks.

[0020] Help for the overall solution: Accurate prediction of loading and unloading operation time helps to formulate reasonable loading and unloading plans and ship scheduling, optimize resource allocation, and avoid long waiting times for ships.

[0021] 3. Probability distribution of crane operation time

[0022] Reason for selection: The yard crane is responsible for the transfer of containers from the quay crane to the yard. The operation time is affected by many factors such as the stacking location of the goods and the status of the yard crane equipment. Therefore, the use of probability distribution can better simulate these changes.

[0023] Advantages: Probability distribution can cover various uncertainties in yard crane operations and provide a relatively accurate operation duration range, thereby optimizing the overall container flow efficiency.

[0024] Help for the overall solution: Through reasonable operation time prediction, yard crane scheduling can be optimized, reducing waiting time and improving the operation smoothness and resource utilization of the entire terminal.

[0025] 4. Probability distribution of container truck moving speed

[0026] Reason for selection: The moving speed of container trucks in and outside the terminal is affected by multiple factors such as traffic flow, route, driver operation, etc., and is uncertain.

[0027] Advantages: By simulating the probability distribution of the moving speed of container trucks, the time required for a container to move from one location to another can be estimated, thus providing a reference for operation planning.

[0028] Help for the overall solution: It can provide more accurate vehicle dispatch time forecasts for job scheduling, avoiding problems such as yard congestion or untimely container stacking caused by improper vehicle scheduling.

[0029] These probability distributions provide a more accurate basis for predicting the future state of container terminals by simulating the randomness and uncertainty of different links. They help improve the accuracy of the system's mapping of future states, allowing managers to take optimization measures in advance based on the system's predictions, reasonably allocate terminal resources, and improve the overall operational efficiency of the terminal and its ability to cope with uncertainty. This solution makes up for the inability of traditional digital twin systems to predict future operating states, and truly realizes the intelligent and dynamic management and control of the yard.

[0030] In some embodiments, the probability model of the probability distribution includes: Poisson distribution, uniform distribution, normal distribution, exponential distribution.

[0031] In the above technical solution, different probability models such as Poisson distribution, uniform distribution, normal distribution and exponential distribution are used in order to more accurately and appropriately simulate the different characteristics and uncertainties of random events in various links of the storage yard. Each probability model has its applicable scenarios and advantages, as follows:

[0032] 1. Poisson Distribution

[0033] Reason for selection: Poisson distribution is usually used to describe the probability of rare events occurring per unit time or unit space. It is particularly suitable for modeling discrete events such as container arrival and job requests.

[0034] Advantages: Poisson distribution is very good at simulating the frequency of random events, such as the frequency of containers arriving at the terminal within a certain period of time. Therefore, it is very useful for dealing with events that occur relatively infrequently but cannot be ignored (such as container entry, ship arrival, etc.).

[0035] 2. Even distribution

[0036] Reason for selection: Uniform distribution is applicable to situations where random variables have the same probability of occurrence within a certain range. For some work processes, if there is no clear preference or historical data support, it can be assumed that the occurrence of events is uniformly distributed.

[0037] Advantages: Uniform distribution is suitable for scenarios where events occur relatively evenly over a period of time. It is characterized by unbiased prediction of random events and is suitable for use when accurate data is lacking.

[0038] 3. Normal Distribution

[0039] Reason for selection: Normal distribution is widely applicable to various continuous variables in nature and social systems. For example, loading and unloading operation time, transportation time, etc. usually fluctuate around a certain average value and show the characteristics of normal distribution.

[0040] Advantages: The two parameters of normal distribution (mean and standard deviation) can well describe the central tendency and fluctuation range of data. It is particularly suitable for scenarios with obvious "average values", such as the operating time of yard cranes or quay cranes, the travel time of container trucks, etc.

[0041] 4. Exponential Distribution

[0042] Reason for selection: Exponential distribution is suitable for describing the intervals between events, especially those independent and continuous random events. For example, the interval between container arrivals or the maintenance time of certain equipment usually conforms to the exponential distribution.

[0043] Advantages: The exponential distribution is characterized by its "memorylessness", that is, future events are not affected by past events, which is very suitable for simulating the time interval between container arrivals or the time interval between equipment failures.

[0044] The main purpose of selecting these probability models is to enable the system to more accurately simulate the randomness and complexity of various operation processes in the yard. By reasonably selecting models such as Poisson distribution, uniform distribution, normal distribution and exponential distribution, various random events in the yard operation and their time distribution can be more accurately described. This has the following specific help to the overall solution:

[0045] Improve forecast accuracy: With appropriate probabilistic models, the system can better predict the frequency and duration of future important operations such as container arrivals, loading and unloading operations, and equipment scheduling.

[0046] Enhance system robustness: By simulating the randomness of different events, the system can better cope with uncertainty, provide more stable operating performance, and avoid operational delays or yard congestion caused by unexpected situations.

[0047] Supporting intelligent decision-making: The data support provided by these models can help intelligent algorithms make more reasonable decisions and improve the overall operating efficiency and economic benefits of container terminals.

[0048] In some embodiments, the container terminal yard management and control gravity field model is used to select the location of the incoming container, specifically:

[0049] Generate a preset number of sub-goals, build a container terminal yard control gravity field model, and set simulation constraint rules;

[0050] Based on the constructed container terminal yard control gravity field model, the incoming containers are located, and the approximate optimal solution is obtained based on the set simulation constraint rules.

[0051] In the above technical solutions, the gravitational field model is chosen instead of other common prediction models (such as regression models, neural networks, time series analysis, etc.) because the gravitational field model has unique advantages in solving such complex, dynamic systems with multi-objective optimization problems. The following is a detailed analysis of why other prediction models are not used and their limitations:

[0052] 1. Linear regression and nonlinear regression models

[0053] limitation:

[0054] Linear assumptions are not suitable for complex systems: Linear regression is suitable for describing systems with linear relationships, while yard management involves many nonlinear factors (such as yard layout, equipment status, changes in operating load, etc.), and the relationship between these factors is difficult to accurately describe with linear equations. Even nonlinear regression, although it can handle some nonlinear relationships, is still limited in effect in a high-dimensional, multivariate environment.

[0055] Unable to capture dynamics: Regression models are usually used for static predictions, that is, given a set of inputs, predict the output value. However, the container yard is a constantly changing dynamic system, and the yard status and demand will change over time. Simple regression models cannot adapt to these dynamic changes in real time.

[0056] 2. Time series models (such as ARIMA, SARIMA)

[0057] limitation:

[0058] Single-objective forecasting: Time series models are good at dealing with time-dependent single-variable forecasts, or predicting the future trend of a specific variable in a multivariate environment. However, container location selection is a multi-objective optimization problem involving interactions in multiple dimensions such as space, time, and equipment use, and time series models cannot handle these complex interdependencies.

[0059] Difficult to adapt to multi-objective decision-making: Yard management is not only about predicting demand at a certain point in the future, but also involves weighing multiple objectives (such as minimizing waiting time, optimizing equipment utilization, etc.). Time series models cannot handle these complex optimization problems.

[0060] 3. Decision models such as decision trees and random forests

[0061] limitation:

[0062] Limited predictive power: Decision trees and random forests can make predictions and classifications based on existing data, but they rely on patterns in historical data and have difficulty handling the randomness and uncertainty in complex dynamic systems. Container terminal yard management involves a large number of real-time decisions, and these models cannot flexibly adapt to changes in system status.

[0063] Lack of spatial optimization capabilities: These models are more suitable for classification or regression tasks rather than spatial optimization decisions. The yard location problem is not just classification or prediction, but also requires optimizing the specific location of the container, minimizing the path, etc. The decision tree model cannot provide direct help in spatial optimization problems.

[0064] 4. Neural Network Model

[0065] limitation:

[0066] Black box problem: Although neural network models, especially deep neural networks, can handle complex nonlinear relationships, they are often "black box" models. It is difficult for yard managers to understand and explain the decisions made by the model, which may be detrimental to trust and adjust operations in practical applications.

[0067] High training and computational costs: Neural networks require a large amount of historical data for training and have high computational costs. Since the state of container yards changes rapidly, real-time training and optimization models are difficult to meet actual needs. In addition, neural networks have difficulty dynamically adapting to emergencies in the system (such as equipment failures, climate change, etc.).

[0068] Lack of mechanism for spatial optimization: Neural networks are suitable for prediction tasks, but the yard location problem is not just about prediction, but also requires optimizing the specific location of containers. The gravitational field model can more directly model the spatial layout and distribution of containers, which meets the essential needs of this problem.

[0069] The reason for choosing the gravitational field model for container location selection is that the model can effectively simulate complex dynamic environments and multi-objective optimization problems. The following are the specific reasons and the outstanding advantages of the model:

[0070] Strong spatial optimization capability: The gravity field model can directly model the spatial layout. By simulating the attraction and repulsion between containers and different areas and equipment, it ensures that the container location not only meets the yard layout requirements but also maximizes resource utilization. Other prediction models lack this direct modeling capability for spatial optimization.

[0071] Multi-objective optimization: The gravitational field model is naturally suitable for multi-objective optimization problems. Each goal (such as reducing waiting time, improving space utilization, and reducing stacking operations) can be regarded as a different source of gravity. By adjusting the weight of gravity, a balance can be found between different goals, while traditional prediction models perform poorly when dealing with multi-objective optimization.

[0072] Dynamic adaptability: The gravity field model can adapt to changes in the state of the yard in real time. For example, when a new container arrives or the yard in a certain area is close to saturation, the gravity field will automatically adjust to dynamically optimize the location of the container. Traditional prediction models need to be retrained or adjusted, and are difficult to flexibly adapt to rapidly changing systems.

[0073] Physical intuitiveness and explainability: The gravitational field model intuitively simulates various influencing factors in the yard through physical concepts (gravity and repulsion), making the model more transparent and easy to explain. In contrast, complex prediction models such as neural networks are difficult to provide explainability of decisions due to their "black box" characteristics.

[0074] The gravitational field model has outstanding advantages in handling spatial optimization, multi-objective optimization, dynamic adaptability and interpretability in container yard management, while traditional prediction models (such as regression, time series, neural networks, etc.) are not competent for this complex real-time system due to lack of spatial optimization capabilities, inability to handle complex dynamic environments, or requiring too high computing costs. Therefore, the gravitational field model was selected as the core model for container yard management and control.

[0075] In some embodiments, the sub-goals include: maximizing the gravity of container tonnage difference, the formula is as follows:

[0076]

[0077] In the formula, wi represents the weight of container i entering the port; w j represents the weight of container j present; X ijk Indicates that the inbound container i is attracted to the storage position next to the on-site container j in mode k; P k is the attraction factor, which indicates the preference degree of k attraction method.

[0078] In the above technical solution, for stacking containers in a yard, different suction methods can accept different weight differences between adjacent containers. The weight difference values ​​determined are different according to different container group classifications and different suction methods.

[0079] The stacking of containers in the yard needs to consider their physical stability. Stacking containers with large weight differences may lead to unstable stacking, increasing the risk of yard operations. For example, placing overweight containers on top of light containers may cause the yard structure to be unbalanced and there is a risk of tipping.

[0080] By maximizing the gravity of container tonnage differences, the system can guide containers of similar weight to attract each other, avoid stacking containers with large weight differences next to each other, ensure that the stacking of containers in the yard is more stable, reduce the risk of accidents, and ensure the safety of operations.

[0081] The stacking order of containers directly affects the operation efficiency. If the weight distribution of containers is unreasonable, the equipment (such as yard cranes, quay cranes, etc.) will need to be frequently moved or re-stacked during loading and unloading operations, increasing operation time and cost.

[0082] This goal can ensure that containers of similar weight are stacked together, thereby reducing the repeated movement of containers during loading, unloading and dispatching. Reasonable weight matching can optimize the frequency of equipment use, reduce unnecessary movement and improve operating efficiency.

[0083] Selecting "maximizing container tonnage differential gravity" as the goal can solve the stability and safety problems in the yard by optimizing weight differential gravity, while improving operational efficiency.

[0084] In some embodiments, the sub-goal includes: maximizing the gravity of the mechanical equipment on the site, the formula is as follows:

[0085]

[0086] In the formula, Z j Indicates whether there is loading and unloading operation in the container area where container j is located. If there is loading and unloading operation, it is 1, otherwise it is 0; L j Indicates whether there is a gantry crane in the container area where container j is located. If there is no gantry crane, it is 1, otherwise it is 0; X ijk It means that the incoming container i is attracted to the storage position next to the on-site container j in the way k.

[0087] In the above technical solution, in the process of allocating container storage space, there is another very important factor, which is whether the site and site tools are available. This is mainly reflected in whether the container area selected for placement is in operation and whether there are yard cranes. When there is no operation and there are idle yard cranes, it is more attractive.

[0088] The sub-goal of "maximizing the attraction of site machinery and equipment" is selected to optimize the utilization of sites and equipment during the container placement process and avoid idle or inefficient use of resources.

[0089] In container yards, the availability and frequency of use of sites and equipment (such as gantry cranes, yard cranes, etc.) directly affect operational efficiency. If the site or equipment is idle and not fully utilized, resource waste will increase; if the site is in operation, excessive load will lead to reduced equipment efficiency and even operational conflicts.

[0090] By aiming to "maximize the attraction of site machinery and equipment", containers can be preferentially arranged in areas where there are no ongoing operations and idle gantry cranes or yard bridges. This ensures the effective use of sites and equipment, avoids problems of idle or overloaded equipment, and improves resource utilization.

[0091] When multiple containers need to be loaded, unloaded or moved at the same time, equipment scheduling conflicts may occur in the yard. Especially when multiple equipment in the same area are operating at the same time, the complexity of resource allocation increases, operating efficiency decreases, and even delays may occur.

[0092] By considering whether the site is currently operating and the availability of the gantry crane, the system can avoid arranging new container stacking in areas with intensive operations. This can effectively reduce equipment scheduling conflicts, ensure that operations in different areas can be balanced, and improve operational efficiency.

[0093] Adopting the goal of "maximizing the attraction of site machinery and equipment" can effectively solve problems such as low site and equipment utilization, operation conflicts, high equipment scheduling costs, and poor operation continuity in container yards.

[0094] In some embodiments, the sub-goals include minimizing the repulsive force, as follows:

[0095]

[0096] In the formula, Q k The repulsion factor indicates the degree of repulsion of the k attraction mode; V ij Indicates whether the ship name and voyage of the incoming container i and the on-site container j are the same, if they are the same, it is 1, otherwise it is 0; POD ijIndicates whether the unloading port of incoming container i is the same as that of on-site container j. If they are the same, it is 1, otherwise it is 0; H ij Indicates whether the height of incoming container i is the same as that of on-site container j. If they are the same, it is 1, otherwise it is 0; W ij Indicates whether the weight level of incoming container i is the same as that of on-site container j, 1 if they are the same, otherwise 0.

[0097] In the above technical solution, if there is attraction, there will be repulsion. k The repulsion force is classified into four levels: weight, container height, unloading port and ship. The smallest repulsion force is when the weight, container height, unloading port and ship are the same, and the repulsion force is the smallest at this time.

[0098] Selecting the goal of "minimizing repulsive forces" can optimize the container stacking strategy and reduce yard management problems caused by inconsistent container properties.

[0099] Different properties of containers, such as weight, height, unloading port, and ship name and voyage, will complicate yard management. If containers with greatly different properties are stacked together, it may increase the difficulty of subsequent operations, such as container sorting and scheduling during transportation. In addition, unreasonable stacking may affect the stability of the container and increase operational risks.

[0100] By minimizing the repulsive force, the system can prioritize stacking containers with the same or similar properties together, making the entire yard layout more orderly and reasonable. Stacking containers of similar weight together helps maintain the stability of the stack, and containers with consistent heights reduce the risk of inconsistent stacking heights. At the same time, stacking containers with the same unloading port and voyage together can simplify subsequent sorting and scheduling.

[0101] If containers with different properties are mixed, subsequent operations (such as container extraction and sorting) will take more time and resources. This will lead to reduced operational efficiency, increased equipment operation time, and may delay the loading, unloading and transportation of goods.

[0102] By minimizing the repulsive force, containers with the same attributes are stacked together first, reducing the difficulty of container scheduling and operational interference. This can significantly improve the operating efficiency of the equipment and reduce the time for container extraction and movement. In particular, the containers with the same unloading port and voyage are gathered together, which can reduce the frequent movement of container trucks or lifting equipment between different yard areas, thereby speeding up the operation process.

[0103] Selecting this goal can effectively reduce the complexity of container yard operations and optimize yard space management; speed up loading and unloading and scheduling operations, and reduce the number of times equipment is moved within the yard, thereby improving overall operating efficiency and reducing operating costs.

[0104] In some embodiments, the simulation constraint rules include:

[0105] (1) The inbound container can only be attracted to one site location constraint:

[0106]

[0107] Where, X ijk Indicates that the inbound container i is attracted to the storage position next to the on-site container j in mode k;

[0108] (2) The tonnage difference between the incoming container and the on-site container that attracts it must be less than the maximum tonnage difference constraint:

[0109]

[0110] Where, X ijk Indicates that the incoming container i is attracted to the storage position next to the on-site container j in mode k; w i represents the weight of container i entering the port; w j represents the weight of container j present; M k is the maximum tonnage difference limit;

[0111] (3) The location of the site to be attracted cannot be occupied:

[0112]

[0113] Where, X ijk Indicates that the inbound container i is attracted to the storage position next to the on-site container j in mode k; O jk Indicates the occupancy of the attracted site location, occupied is 0, unoccupied is 1;

[0114] (4) The site location to be attracted cannot violate the floor height restriction constraints:

[0115]

[0116] Where, X ijk Indicates that the inbound container i is attracted to the storage position next to the on-site container j in mode k; T j represents the height of container j; T represents the floor height limit;

[0117] (5) The location of the site attracted cannot violate the box size restrictions:

[0118]

[0119] Where, X ijk Indicates that the inbound container i is attracted to the storage position next to the on-site container j by way k; S ijIndicates whether the size of incoming container i exceeds the container size limit, 1 if it does not exceed, and 0 if it does not.

[0120] In the above technical solution, the above constraints are selected to ensure safety, space optimization, and maximum operating efficiency during container stacking, while avoiding common problems in yard management. The following are the starting points, reasons, and advantages of these constraints:

[0121] 1. The inbound container can only be attracted to one site location

[0122] Starting point: Each incoming container can only have one clear storage location to ensure the uniqueness and rationality of the decision and avoid containers being incorrectly or repeatedly allocated to multiple locations, which would cause scheduling confusion.

[0123] Reason for selection: The location of the container must be unique to ensure the continuity and consistency of operations in the yard and reduce management complexity. Through this constraint, the system can find the optimal location for each container to avoid resource waste or operational errors caused by incorrect allocation.

[0124] advantage:

[0125] Ensure uniqueness of container stacking to avoid confusion or overlapping decisions.

[0126] 2. The tonnage difference between the incoming container and the on-site container that attracts it must be less than the maximum tonnage difference limit constraint

[0127] Starting point: This constraint ensures that the weight difference of containers is not too large, thereby avoiding the risk of unstable stacking and improving the safety and stability of the yard.

[0128] Reason for selection: Container weight is a key factor affecting stacking stability. Stacking containers with large weight differences may cause stacking imbalance and increase the risk of accidents. Setting a maximum tonnage difference limit can ensure the stability and safety of stacking.

[0129] advantage:

[0130] Improve yard safety and avoid stack collapse due to uneven weight distribution.

[0131] 3. The location of the site cannot be occupied

[0132] Starting point: This constraint ensures that containers can only be stacked in unoccupied spaces, avoiding stacking conflicts, and preventing space waste and misoperation during operations.

[0133] Reason for selection: The space in the container yard is limited, and the rational use of unoccupied space is the key to improving the utilization rate of the yard. Preventing containers from being stacked in occupied spaces can ensure smooth operations and avoid operational interruptions caused by space conflicts.

[0134] advantage:

[0135] Avoid stacking conflicts and ensure smooth operation.

[0136] 4. The location of the site to be attracted cannot violate the constraints of the floor height limit

[0137] Starting point: The yard has certain restrictions on the number of layers of containers that can be stacked. Stacking too high will increase safety hazards and affect the operational safety of the yard.

[0138] Reason for selection: Setting a height limit can prevent the instability caused by stacking containers too high, thereby ensuring the safe operation of the yard. The height limit can also prevent equipment operation difficulties caused by height restrictions (such as the inability to safely operate the gantry crane, etc.).

[0139] advantage:

[0140] Ensure stacking safety and prevent unstable stacking or inconvenience in equipment operation in the yard.

[0141] 5. The attracted site location cannot violate the box size limit constraints

[0142] Starting point: The dimensions of the container (such as length, width and height) need to match the yard space and equipment operating capacity to prevent inappropriate container stacking from causing equipment operation difficulties or wasting space.

[0143] Reason for selection: Containers of different types and specifications (such as 20 feet, 40 feet, etc.) need to comply with the standardized operation procedures of the yard when stacking. Setting size limit constraints can avoid operational difficulties and waste of space resources caused by mismatched container sizes.

[0144] advantage:

[0145] Avoid wasted space and prevent operational or space management issues due to sizing discrepancies.

[0146] According to another aspect of the present invention, a digital twin simulation system for container terminal yard management and control is provided, comprising a collection module, a setting module, a simulation module, and an analysis module connected in sequence; wherein:

[0147] Acquisition module: used to obtain production data related to the current container terminal and yard management and control, and generate the initial state of the digital twin platform;

[0148] Setting module: used to set the probability distribution of random events of container entry in the digital twin platform according to the historical operation rules of the container terminal;

[0149] Simulation module: simulates the process of container site selection based on the digital twin platform, and simulates random events according to the set probability distribution;

[0150] Analysis module: select the location of incoming containers through the container terminal yard management and control gravity field model.

[0151] In the above technical scheme, in order to better use the above method, the present application proposes a digital twin simulation system for container terminal yard management and control, and each module corresponds to each step of the above method. The specific principles have been described above and will not be repeated here.

[0152] According to another aspect of the present invention, there is provided a computer-readable storage medium storing a computer program, wherein the computer program implements the above method when executed by a processor.

[0153] In the above technical solution, in order to better run and use the method, the above method is stored in a computer-readable storage medium, and the above method is implemented by a processor. It should be noted that the principle and effect of each step have been described above and will not be described in detail here. BRIEF DESCRIPTION OF THE DRAWINGS

[0154] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in 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 described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0155] Figure 1 It is a flow chart of an embodiment of a digital twin simulation method for container terminal yard management and control according to the present invention;

[0156] Figure 2 It is a structural schematic diagram of an embodiment of a digital twin simulation system for container terminal yard management and control according to the present invention; DETAILED DESCRIPTION

[0157] The present invention will be further described in detail below in conjunction with the accompanying drawings and examples. It is particularly noted that the following examples are only used to illustrate the present invention, but are not intended to limit the scope of the present invention. Similarly, the following examples are only partial embodiments of the present invention rather than all embodiments, and all other embodiments obtained by those of ordinary skill in the art without creative work are within the scope of protection of the present invention.

[0158] The present invention provides a digital twin simulation method, system and storage medium for container terminal yard management and control, which can solve the shortcoming of the digital twin system for container terminal yard management and control that lacks mapping of future yard operation status through random event generation and in conjunction with a yard management and control gravitational field model.

[0159] Embodiment 1

[0160] See also Figure 1 , a digital twin simulation method for container terminal yard management and control, the method comprising:

[0161] S1. Obtain production data related to the current container terminal and yard management and control, and generate the initial state of the digital twin platform;

[0162] In this embodiment, the digital twin platform is built based on existing technologies. For details, please refer to existing technologies such as the use of large-scale LOD rendering technology and JS libraries such as Cesium, which will not be described in detail here.

[0163] S2. According to the historical operation rules of the container terminal, the probability distribution of random events of container entry is set in the digital twin platform;

[0164] In this embodiment, the probability distribution of random events includes: probability distribution of container arrival time, probability distribution of quay crane loading and unloading operation time, probability distribution of yard crane operation time, and probability distribution of container truck moving speed;

[0165] In this embodiment, the probability model of the probability distribution includes: Poisson distribution, uniform distribution, normal distribution, and exponential distribution.

[0166] S3. Simulate the process of container location selection based on the digital twin platform and simulate random events according to the set probability distribution;

[0167] S4. Select the location of incoming containers through the container terminal yard management and control gravity field model.

[0168] In this embodiment, S4, the incoming containers are positioned by using the container terminal yard management and control gravity field model, specifically:

[0169] S41, generating a preset number of sub-goals, constructing a container terminal yard control gravity field model, and setting simulation constraint rules;

[0170] In this embodiment, one of the preferred sub-goals includes:

[0171] Maximize the gravity of container tonnage difference, the formula is as follows:

[0172]

[0173] In the formula, w i represents the weight of container i entering the port; w j represents the weight of container j present; X ijk Indicates that the inbound container i is attracted to the storage position next to the on-site container j in mode k; P k is the attraction factor, which indicates the preference of the k attraction method. For the stacking of containers in the yard, the weight difference ranges of adjacent containers that can be accepted by different attraction methods are different. The weight difference values ​​determined are different according to the different classification of container groups and different attraction methods.

[0174] In this embodiment, one of the preferred sub-goals includes:

[0175] Maximize the attraction of the site's mechanical equipment. The formula is as follows:

[0176]

[0177] In the formula, Z j Indicates whether there is loading and unloading operation in the container area where container j is located. If there is loading and unloading operation, it is 1, otherwise it is 0; L j Indicates whether there is a gantry crane in the container area where container j is located. If there is no gantry crane, it is 1, otherwise it is 0; X ijk Indicates that the inbound container i is attracted to the storage location next to the container j in the yard by way k. In the process of allocating container storage space, another very important factor is whether the yard and yard tools are available. It is mainly reflected in whether the container area selected for the location is in operation and whether there is a yard crane. When there is no operation and there is an idle yard crane, the attraction is greater.

[0178] In this embodiment, one of the preferred sub-goals includes:

[0179] Minimize the repulsive force, the formula is as follows:

[0180]

[0181] In the formula, Q k The repulsion factor indicates the degree of repulsion of the k attraction mode; V ij Indicates whether the ship name and voyage of the incoming container i and the on-site container j are the same, if they are the same, it is 1, otherwise it is 0; POD ij Indicates whether the unloading port of incoming container i is the same as that of on-site container j. If they are the same, it is 1, otherwise it is 0; H ij Indicates whether the height of incoming container i is the same as that of on-site container j. If they are the same, it is 1, otherwise it is 0; W ij Indicates whether the weight level of incoming container i is the same as that of on-site container j. If they are the same, it is 1, otherwise it is 0. If there is attraction, there will be repulsion. kThe repulsion force is classified into four levels: weight, container height, unloading port and ship. The smallest repulsion force is when the weight, container height, unloading port and ship are the same, and the repulsion force is the smallest at this time.

[0182] In this embodiment, the simulation constraint rules include:

[0183] (1) The inbound container can only be attracted to one site location constraint:

[0184]

[0185] Where, X ijk Indicates that the inbound container i is attracted to the storage position next to the on-site container j in mode k;

[0186] (2) The tonnage difference between the incoming container and the on-site container that attracts it must be less than the maximum tonnage difference constraint:

[0187]

[0188] Where, X ijk Indicates that the incoming container i is attracted to the storage position next to the on-site container j in mode k; w i represents the weight of container i entering the port; w j represents the weight of container j present; M k is the maximum tonnage difference limit;

[0189] (3) The location of the site to be attracted cannot be occupied:

[0190]

[0191] Where, X ijk Indicates that the inbound container i is attracted to the storage position next to the on-site container j in mode k; O jk Indicates the occupancy of the attracted site location, occupied is 0, unoccupied is 1;

[0192] (4) The site location to be attracted cannot violate the floor height restriction constraints:

[0193]

[0194] Where, X ijk Indicates that the inbound container i is attracted to the storage position next to the on-site container j in mode k; T j represents the height of container j; T represents the floor height limit;

[0195] (5) The location of the site attracted cannot violate the box size restrictions:

[0196]

[0197] In the formula, Cijk Indicates that the inbound container i is attracted to the storage position next to the on-site container j by way k; S ij Indicates whether the size of incoming container i exceeds the container size limit, 1 if it does not exceed, and 0 if it does not.

[0198] S42. Based on the constructed container terminal yard control gravity field model, the incoming containers are located, and the approximate optimal solution is obtained based on the set simulation constraint rules.

[0199] In the above technical solution, the changes in yard control in the future are simulated in the digital space to realize the mapping of the future state of the container terminal yard operation. The present invention can solve the defect that the traditional digital twin system of the container terminal yard cannot map the future operation state of the yard control.

[0200] Based on one of the embodiments, it can be known that the present invention has the following advantages:

[0201] 1. Accurate job prediction and optimization

[0202] The present invention can simulate the yard operation status in the future in digital space by building a digital twin platform, combining random event generation and gravitational field model. Through accurate mapping of future operation scenarios, the system can detect potential problems in advance and make optimization adjustments. This capability solves the defect that traditional digital twin systems can only reflect the current status but cannot predict future operation status, helping terminal managers to better arrange future loading and unloading operations and resource scheduling.

[0203] 2. Improve resource utilization and operational efficiency

[0204] The gravity field model can optimize the stacking position of containers and equipment scheduling by maximizing the gravity of container tonnage difference and the gravity of mechanical equipment in the yard. It can ensure that containers are stacked according to the optimal weight difference and stacking order, reduce equipment scheduling conflicts, and reduce equipment idling time, thereby improving the utilization rate of the yard and equipment. Through refined management and reasonable allocation, the resource utilization efficiency of the yard has been significantly improved, and the operation process has become smoother and more efficient.

[0205] 3. Improve work safety

[0206] By setting reasonable constraint rules, such as weight difference limit, layer height limit, container size limit, etc., the present invention can ensure the safety of container stacking. In particular, the tonnage difference limit and layer height limit avoid the potential safety hazards caused by unreasonable stacking, and help prevent container collapse or unstable equipment operation. Through the implementation of these safety measures, yard management is not only more efficient, but also ensures safety during the operation and reduces the risk of accidents.

[0207] 4. Personalized stacking strategy and flexible management

[0208] By introducing the goal of minimizing repulsive force, the present invention realizes personalized management of container stacking. The system can classify containers according to their attributes (such as ship name and voyage, unloading port, container height, weight grade, etc.) and group containers with similar attributes together. This not only optimizes the stacking structure of containers, but also improves the efficiency of subsequent scheduling and loading and unloading operations. At the same time, based on the dynamic adjustment of the gravitational and repulsive force models, the system can flexibly adjust the stacking plan according to the real-time status of the site and equipment, realizing flexible management and decision optimization.

[0209] Embodiment 2

[0210] See also Figure 2 , a digital twin simulation system for container terminal yard management and control, including a collection module, a setting module, a simulation module, and an analysis module connected in sequence; wherein,

[0211] Acquisition module: used to obtain production data related to the current container terminal and yard management and control, and generate the initial state of the digital twin platform;

[0212] Setting module: used to set the probability distribution of random events of container entry in the digital twin platform according to the historical operation rules of the container terminal;

[0213] Simulation module: simulates the process of container site selection based on the digital twin platform, and simulates random events according to the set probability distribution;

[0214] Analysis module: select the location of incoming containers through the container terminal yard management and control gravity field model.

[0215] In this embodiment, in order to better use the method described in one of the embodiments, the present application proposes a digital twin simulation system for container terminal yard management and control, and each module corresponds to each step of the above method. The specific principles have been described above and will not be repeated here.

[0216] Embodiment 3

[0217] A computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the method described in one of the embodiments is implemented.

[0218] In the above technical solution, in order to better run and use the method described in one of the embodiments, the above method is stored in a computer-readable storage medium, and a processor is used to implement the method described in one of the embodiments. It should be noted that the principle and effect of each step have been described above and will not be described in detail here.

[0219] The above descriptions are only some embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Any equivalent device or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A digital twin simulation method for container terminal yard management and control, characterized in that: The method comprises: Obtain production data related to current container terminal and yard management and control, and generate the initial state of the digital twin platform; According to the historical operation rules of the container terminal, the probability distribution of random events of container entry is set in the digital twin platform; The digital twin platform is used to simulate the process of container location selection, and random events are simulated according to the set probability distribution; The incoming containers are positioned using the container terminal yard management and control gravity field model.

2. A digital twin simulation method for container terminal yard management and control according to claim 1, characterized in that: The probability distribution of random events includes: the probability distribution of container arrival time, the probability distribution of quay crane loading and unloading operation time, the probability distribution of yard crane operation time, and the probability distribution of container truck moving speed.

3. A digital twin simulation method for container terminal yard management and control as claimed in claim 2, characterized in that: The probability models of the probability distribution include: Poisson distribution, uniform distribution, normal distribution, and exponential distribution.

4. A digital twin simulation method for container terminal yard management and control as claimed in claim 1, characterized in that: The container terminal yard control gravity field model is used to select the location of incoming containers. Specifically: Generate a preset number of sub-goals, build a container terminal yard control gravity field model, and set simulation constraint rules; Based on the constructed container terminal yard control gravity field model, the incoming containers are located, and the approximate optimal solution is obtained based on the set simulation constraint rules.

5. A digital twin simulation method for container terminal yard management and control as claimed in claim 4, characterized in that: The sub-goals include: maximizing the gravity of container tonnage difference, the formula is as follows: In the formula, w i represents the weight of container i entering the port; w j represents the weight of container j present; X ijk Indicates that the inbound container i is attracted to the storage position next to the on-site container j in mode k; P k is the attraction factor, which indicates the preference degree of k attraction method.

6. A digital twin simulation method for container terminal yard management and control as claimed in claim 4, characterized in that: The sub-goals include: maximizing the gravity of the mechanical equipment on the site, the formula is as follows: In the formula, Z j Indicates whether there is loading and unloading operation in the container area where container j is located. If there is loading and unloading operation, it is 1, otherwise it is 0; L j Indicates whether there is a gantry crane in the container area where container j is located. If there is no gantry crane, it is 1, otherwise it is 0; X ijk It means that the incoming container i is attracted to the storage position next to the on-site container j in the way k.

7. A digital twin simulation method for container terminal yard management and control as claimed in claim 4, characterized in that: The sub-goals include: minimizing the repulsive force, the formula is as follows: In the formula, Q k The repulsion factor indicates the degree of repulsion of the k attraction mode; V ij Indicates whether the ship name and voyage of the incoming container i and the on-site container j are the same, if they are the same, it is 1, otherwise it is 0; POD ij Indicates whether the unloading port of incoming container i is the same as that of on-site container j. If they are the same, it is 1, otherwise it is 0; H ij Indicates whether the height of incoming container i is the same as that of on-site container j. If they are the same, it is 1, otherwise it is 0; W ij Indicates whether the weight level of incoming container i is the same as that of on-site container j, 1 if they are the same, otherwise 0.

8. A digital twin simulation method for container terminal yard management and control according to any one of claims 4 to 7, characterized in that: The simulation constraint rules include: (1) The inbound container can only be attracted to one site location constraint: Where, X ijk Indicates that the inbound container i is attracted to the storage position next to the on-site container j in mode k; (2) The tonnage difference between the incoming container and the on-site container that attracts it must be less than the maximum tonnage difference constraint: Where, X ijk Indicates that the incoming container i is attracted to the storage position next to the on-site container j in mode k; w i represents the weight of container i entering the port; w j represents the weight of container j present; M k is the maximum tonnage difference limit; (3) The location of the site to be attracted cannot be occupied: Where, X ijk Indicates that the inbound container i is attracted to the storage position next to the on-site container j in mode k; O jk Indicates the occupancy of the attracted site location, occupied is 0, unoccupied is 1; (4) The site location to be attracted cannot violate the floor height restriction constraints: Where, X ijk Indicates that the inbound container i is attracted to the storage position next to the on-site container j in mode k; T j represents the height of container j; T represents the floor height limit; (5) The location of the site attracted cannot violate the box size restrictions: Where, X ijk Indicates that the inbound container i is attracted to the storage position next to the on-site container j by way k; S ij Indicates whether the size of incoming container i exceeds the container size limit, 1 if it does not exceed, and 0 if it does not.

9. A digital twin simulation system for container terminal yard management and control, characterized in that: It includes a collection module, a setting module, a simulation module, and an analysis module connected in sequence; among which, Acquisition module: used to obtain production data related to the current container terminal and yard management and control, and generate the initial state of the digital twin platform; Setting module: used to set the probability distribution of random events of container entry in the digital twin platform according to the historical operation rules of the container terminal; Simulation module: simulates the process of container site selection based on the digital twin platform, and simulates random events according to the set probability distribution; Analysis module: select the location of incoming containers through the container terminal yard management and control gravity field model.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the digital twin simulation method for container terminal yard management and control described in any one of claims 1 to 8 is implemented.

Images