Container yard digital twinborn model construction and monitoring method based on Internet of Things

Abstract

The invention belongs to the technical field of data processing, and particularly relates to a container yard digital twinborn model construction and monitoring method based on the Internet of Things, and the method comprises the steps: collecting the related information of a container yard in real time, and constructing a digital twinborn model of the container yard; dynamic data of the container yard is obtained in real time based on a sensor, the dynamic data can represent the operation state of the container yard, and the digital twinborn model is updated in real time based on the dynamic data; obtaining a monitoring result of the container yard based on the digital twinborn model, and generating a stockpiling scheme and a scheduling scheme based on the monitoring result; and simulating the stockpiling scheme and the scheduling scheme in the digital twinborn model to obtain a simulation result, optimizing the stockpiling scheme and the scheduling scheme based on the simulation result, and simulating the optimized schemes in the digital twinborn model again until an optimal stockpiling scheme and an optimal scheduling scheme are obtained. According to the method, the accurate digital twinborn model can be constructed, and the operation efficiency of the container yard can be improved based on the digital twinborn model.

Description

Technical Field

[0001] The present invention belongs to the technical field of data processing, and in particular relates to a container yard digital twin model construction and monitoring method based on the Internet of Things. Background Art

[0002] With the rapid development of global trade, container transportation has become the main mode of international trade. However, traditional container transshipment monitoring methods have problems such as untimely information updates and limited monitoring range, which are difficult to meet the requirements of modern logistics for real-time and accuracy. As an emerging technical means, digital twin technology can build a virtual model of a physical entity and realize real-time monitoring and management of the entity. A similar prior art Chinese patent application with publication number CN118069475A provides a digital twin monitoring system, including a digital twin management end, a monitoring terminal and a user end; the monitoring terminal generates monitoring operation data; the digital twin management end builds a digital image of the monitoring system and a digital image of the monitoring scene, and also drives the digital image of the monitoring system to perform monitoring operation simulation and change simulation of the digital image of the monitoring scene according to the monitoring operation data; the user end obtains the user's operating parameters and uploads the monitoring scene data and operating parameters to the digital twin management end; the digital twin management end also operates the digital image of the monitoring system and the digital image of the monitoring scene according to the operating parameters, and removes the digital images that do not meet the preset goals in the operation results, and renders the digital images that meet the preset goals. Similar prior art also has a Chinese patent application with publication number CN113886177A, which discloses a digital twin monitoring system, including a digital twin monitoring board, a digital twin management platform and a digital twin center, wherein the digital twin monitoring board includes: a monitoring module, a data storage unit, an analysis module and a transmission module, wherein the monitoring module is respectively connected to the server clock and the board clock located on the digital twin monitoring board, and is configured to obtain the clock information of the server clock and the board clock to perform clock information matching comparison, and send corresponding information to the transmission module according to the matching relationship of the clock information; the data storage unit is connected to the BMC on the server, and is configured to collect and store the log of the BMC; the analysis module is connected to the data storage unit and the transmission module, and is configured to analyze the log of the BMC collected and stored by the data storage unit, and send the analysis result to the digital twin management platform through the transmission module. However, the above two technical solutions do not consider the accuracy of the collected relevant data when creating the digital twin model, which may lead to inaccurate digital twin models created. Therefore, the present invention provides a digital twin model construction and monitoring method for container yards based on the Internet of Things. Summary of the invention

[0003] The present invention monitors the container yard by constructing a digital twin model of the container yard, and also generates storage plans and scheduling plans to improve the operating efficiency of the container yard.

[0004] In order to achieve the above-mentioned purpose of the invention, the present invention provides the following method for constructing and monitoring a digital twin model of a container yard based on the Internet of Things, which is implemented by executing the following steps: S1. Collect relevant information of the container yard in real time based on the Internet of Things technology, wherein the relevant information includes layout information and equipment information, wherein the equipment information includes container information and transshipment equipment information, wherein the container information includes the location, container number, container type, empty weight status and transportation task of the container, and wherein the transshipment equipment information includes reach stacker information and container truck information, and construct a digital twin model of the container yard based on the collected relevant information; S2. Acquire dynamic data of the container yard in real time based on sensors, where the dynamic data can represent the operating status of the container yard, including loading, unloading, handling and storage of containers, and send the dynamic data to the digital twin model, which processes and analyzes the dynamic data and updates the digital twin model in real time based on the dynamic data; S3. Acquire monitoring results of the container yard based on the digital twin model, analyze the container storage conditions of the container yard based on the monitoring results to generate a storage plan, and analyze the transshipment equipment based on the monitoring results to generate a scheduling plan; S4. Simulate the stockpiling plan and the scheduling plan in the digital twin model to obtain simulation results, optimize the stockpiling plan and the scheduling plan based on the simulation results, and re-simulate the optimized stockpiling plan and the scheduling plan in the digital twin model until the optimal stockpiling plan and the optimal scheduling plan are obtained.

[0005] As a preferred technical solution of the present invention, generating a stockpiling plan includes the following steps: S311, determine the time length, obtain the container entry volume of the container yard for each of the time lengths in the past, detect the container entry volume to obtain abnormal data therein, replace the abnormal data in the container entry volume with the average of the data before and after the abnormal data, and then normalize the processed data of the container entry volume, and generate time series data for the normalized container entry volume according to the chronological order of collection time; S312, obtaining first data within a first time interval from the time series data, using the first data as training data to train a first prediction model, predicting first prediction data of the next time length of the first time interval based on the first prediction model, obtaining first actual data corresponding to the first prediction data, and adjusting and optimizing the first prediction model based on the difference between the first prediction data and the first actual data; S313, deleting the first container entry quantity with the earliest collection time in the first data, and adding the first actual data input to the first data, using the updated first data as new training data to continue training the first prediction model, and repeatedly executing S312 until all the data in the time series data are used as training data to train the first prediction model, and using the first prediction model generated by the final training to predict the container entry quantity of the future time length; S314: Generate a storage plan based on the container entry quantity.

[0006] As a preferred technical solution of the present invention, generating a stockpiling plan based on the predicted entry volume includes the following steps: S3141, obtaining all storage areas of the container yard, obtaining the predicted container entry quantity, obtaining a first distance from the container to be entered to each storage area, randomly generating a first number of storage plans, and calculating a corresponding total moving distance for each storage plan; S3142: preset a first value, initialize the first value to zero, select a second number of candidate solutions from all the stacking solutions based on the total moving distance, generate a random number that is less than the predicted entry quantity, exchange the stacking methods of the containers after the random numbers in any two of the candidate solutions, generate a third number of new stacking solutions, add one to the first value, and calculate the total moving distance of each new stacking solution; S3143, obtaining the minimum total moving distance, if the minimum total moving distance is greater than a preset first threshold and the first value is less than a preset second threshold, returning to S3142, otherwise, executing the next step; S3144. Obtain the minimum value of all the calculated total moving distances and convert them into.

[0007] As a preferred technical solution of the present invention, selecting a second number of stockpiling solutions from all the stockpiling solutions based on the total moving distance comprises the following steps: A fourth number of stockpiling schemes are randomly selected from the stockpiling schemes as the first stockpiling schemes, and a fifth number of the first stockpiling schemes with the smallest total moving distances among the first stockpiling schemes are selected as candidate schemes, and this step is repeated until the number of the selected candidate schemes is equal to the second number.

[0008] As a preferred technical solution of the present invention, generating a scheduling plan includes the following steps: S321, obtaining a first storage area for containers that need to be transported out, obtaining a destination for transshipment, planning an optimal path from the first storage area to the corresponding destination for each transshipment device, simulating transshipment of the optimal path in the digital twin model, obtaining load data of each optimal path during transshipment, and determining a potential congested section based on the load data; S322: If the congested section exists in the optimal path, an idle section that can replace the congested section is obtained based on the digital twin model, and the congested section of the optimal path is modified into the idle section.

[0009] As a preferred technical solution of the present invention, planning the best path from the first storage area to the corresponding destination for each transfer device includes the following steps: S3211, initializing the first list, obtaining the position of each container in the first storage area as a starting point, adding the starting point to the first list, and taking the destination to which the container is to go as an end point; S3212: take the starting point as the first node, obtain four second nodes adjacent to the first node, calculate the second distances of four paths from the first node through each of the second nodes to the end point, obtain the minimum second distance therebetween, and add the second node corresponding to the minimum second distance to the first list; S3213. Take the second node as the first node, return to S3212, add one of the adjacent nodes of the second node to the first list, repeat this step until the end point is added to the first list, and take the path composed of all the nodes in the first list as the optimal path.

[0010] As a preferred technical solution of the present invention, before constructing the digital twin model of the container yard based on the collected relevant information, the following steps are further performed: A reference range for each data parameter is stored in advance, and each data parameter in the relevant information is compared with the corresponding reference range. If the data parameter is within the corresponding reference range, the data parameter is marked as normal; otherwise, the data parameter is marked as abnormal; for the data parameter marked as abnormal, a data error message is sent to the data sender; after receiving the data error message, the data sender obtains the correct data parameter and the corresponding correction code and sends them to the data receiver; the data receiver determines whether the received new data parameter is normal; if so, this step is terminated; otherwise, the data parameter is corrected based on the correction code.

[0011] Compared with the prior art, the beneficial effects of the present invention are at least as follows: In the present invention, firstly, the relevant information of the container yard is collected in real time based on the Internet of Things technology, the collected relevant information is verified, the wrong relevant information is corrected, and the digital twin model of the container yard is constructed based on the correct relevant information to ensure the construction of an accurate digital twin model; the dynamic data of the container yard is obtained in real time based on the sensor, and the dynamic data can represent the operating status of the container yard, and the digital twin model is updated in real time based on the dynamic data; the monitoring results of the container yard are obtained based on the digital twin model, and the storage plan and the scheduling plan are generated based on the monitoring results; the storage plan and the scheduling plan are simulated in the digital twin model to obtain the simulation results, the storage plan and the scheduling plan are optimized based on the simulation results, and the optimized storage plan and the scheduling plan are re-simulated in the digital twin model until the optimal storage plan and the optimal scheduling plan are obtained. The present invention can reasonably and effectively store containers in the container yard based on the optimal storage plan, and can also efficiently schedule the transfer equipment based on the optimal scheduling plan, thereby improving the overall operation efficiency of the container yard. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 This is a flowchart of the steps of the method for building and monitoring a digital twin model of a container yard based on the Internet of Things of the present invention. DETAILED DESCRIPTION

[0013] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0014] It is understood that the terms "first", "second", etc. used in this application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, without departing from the scope of this application, a first xx script may be referred to as a second xx script, and similarly, a second xx script may be referred to as a first xx script.

[0015] The present invention provides Figure 1 The method for constructing and monitoring a digital twin model of a container yard based on the Internet of Things is shown, which is implemented by executing the following steps: S1. Collect relevant information of the container yard in real time based on the Internet of Things technology. The relevant information includes layout information and equipment information. The equipment information includes container information and transshipment equipment information. The container information includes the location, container number, container type, empty weight status and transportation task of the container. The transshipment equipment information includes reach stacker information and container truck information. Build a digital twin model of the container yard based on the collected relevant information.

[0016] Specifically, in order to create a digital twin model of the container yard, the Internet of Things technology is used. For example, sensors are installed on the container yard, containers, reach stackers, container trucks and other transfer equipment to collect relevant data in real time. The relevant data includes the layout information and equipment information of the container yard. The layout information includes the container area, container position, number and other information of each storage area. The container information includes the location, container number, container type, empty weight status and transfer task information. The transfer task refers to the time when the container needs to be transported out of the yard and the transfer destination, which is the exit of the container yard. The transfer equipment includes reach stackers and container trucks. The reach stackers are used to load containers from the storage area into transport vehicles or unload containers from transport vehicles and put them into the storage area. The transfer equipment information includes the location, number, etc. of the reach stackers and container trucks. Based on the above relevant information, a digital twin model of the container yard can be accurately constructed. Based on the construction of the digital twin model, the optimal scheduling of container yard resources can be achieved, and the operating efficiency of the container yard can be improved.

[0017] S2. Real-time acquisition of dynamic data of the container yard based on sensors. The dynamic data can represent the operating status of the container yard, including the loading, unloading, handling and storage of containers. The dynamic data is sent to the digital twin model, which processes and analyzes the dynamic data and updates the digital twin model in real time based on the dynamic data.

[0018] Specifically, since the containers and transshipment equipment in the container yard are constantly running, in order to enable the digital twin model to actually interact with the actual container yard, the dynamic data of the container yard is acquired in real time based on sensors, and the dynamic data is sent to the digital twin model in real time. The digital twin model processes and analyzes the dynamic data to obtain the dynamic changes of the corresponding containers and transshipment equipment, and updates the dynamic changes of the containers and transshipment equipment to the digital twin model in real time, thereby realizing real-time monitoring of the overall operating status of the container yard. Through real-time monitoring of the container yard, the resource utilization data of the container yard can be obtained in real time, and resource scheduling can be optimized based on various resource utilization data of the container yard to improve resource utilization.

[0019] S3. Obtain the monitoring results of the container yard based on the digital twin model, analyze the container storage situation in the container yard based on the monitoring results to generate a storage plan, and analyze the transshipment equipment based on the monitoring results to generate a scheduling plan.

[0020] Specifically, the monitoring results of the container yard are obtained based on the digital twin model. The monitoring results include the distribution of containers in the current container yard, such as the number of containers stored in each storage area, the domain of containers that can be stored, the storage area where the containers that need to leave are located, the load information of each transfer path, the location and quantity of dispatchable transfer equipment, and other information. The monitoring results also include historical monitoring results. Based on the above information, the container storage situation of the container yard is analyzed to generate a storage plan. The specific method for generating the storage plan will be explained in detail later. The transfer equipment is also analyzed to generate a scheduling plan for the transfer equipment. The specific method for generating the scheduling plan will be explained in detail later.

[0021] S4. Simulate the stockpile plan and the scheduling plan in the digital twin model to obtain simulation results, optimize the stockpile plan and the scheduling plan based on the simulation results, and re-simulate the optimized stockpile plan and the scheduling plan in the digital twin model until the optimal stockpile plan and the optimal scheduling plan are obtained.

[0022] Specifically, after generating the stacking plan and the scheduling plan, the stacking plan and the scheduling plan can be simulated in the digital twin model to obtain simulation results, and the feasibility and effect of the plan can be evaluated based on the simulation results. The stacking plan and the scheduling plan are also optimized based on the simulation results. For example, in the process of storing containers according to the actual stacking plan, if the arrival time of the container is late, the idle front loaders or container trucks can be used to transfer the containers that arrived first. For example, in the simulation process, it is found that the containers in a certain storage area are stored very concentratedly. Since the queuing storage will cause the storage time to be longer, the stacking plan can be adjusted. Based on the above method, by continuously simulating the actual stacking operations and scheduling operations in the digital twin model, the stacking plan and the scheduling plan are continuously optimized until the optimal stacking plan and the optimal scheduling plan are obtained. Based on the optimal stacking plan, the containers in the container yard are reasonably and effectively stored, and based on the optimal scheduling plan, the transfer equipment is efficiently scheduled, thereby improving the operation efficiency of the container yard.

[0023] Furthermore, generating a stockpile plan includes the following steps: S311. Determine the time length, obtain the container entry volume of the container yard for each time length in the past, detect the container entry volume to obtain abnormal data, replace the abnormal data in the container entry volume with the mean of the data before and after the abnormal data, and then normalize the processed container entry volume data, and generate time series data for the normalized container entry volume in the chronological order of the collection time.

[0024] S312. Obtain first data within a first time interval from the time series data, use the first data as training data to train a first prediction model, predict first prediction data of the next time length of the first time interval based on the first prediction model, obtain first actual data corresponding to the first prediction data, and adjust and optimize the first prediction model based on the difference between the first prediction data and the first actual data.

[0025] S313, deleting the first container entry volume with the earliest collection time in the first data, and adding the first actual data input to the first data, using the updated first data as new training data to continue training the first prediction model, and repeating S312 until all the data in the time series data are used as training data to train the first prediction model, and using the first prediction model generated by the final training to predict the container entry volume of the future time length; S314, generating a storage plan based on the container entry quantity.

[0026] Specifically, under normal circumstances, it is necessary to pre-generate a stacking plan for containers that are about to enter the container yard, that is, which container is stored in which storage area faster, and the formulation of the stacking plan is highly correlated with the number of containers that are about to enter the container yard. Moreover, if the containers are stored too concentratedly or dispersedly during storage, the stacking efficiency of the containers will be affected. Therefore, it is necessary to accurately predict the number of containers entering the yard in order to generate an effective stacking plan. First, a time length is determined, such as one day, and the daily container entry volume of the container yard in the past is obtained. The container entry volume refers to the number of containers that enter the container yard within one day. The obtained container entry volume may be due to There are some abnormal values ​​due to transmission errors, such as negative values ​​or very large values. These abnormal data are impossible to appear in actual situations and have an impact on the subsequent prediction accuracy. Therefore, the abnormal data is replaced by the mean of the data before and after the abnormal data. For example, the container entry volume on May 20 is -12, which is obviously abnormal data. The container entry volume on May 19 and May 21 is obtained and the average of these two container entry volumes is taken to replace the abnormal data. Then, for the efficiency of subsequent predictions, all container entry data are normalized, such as converting all these values ​​into numbers from zero to one, and then these numbers are generated into time series data in the order of collection time.

[0027] Then, taking the time series data of 60 days as an example, the first data of the first 20 days are obtained from the 60 days of data as training data to train the first prediction model, and then the first prediction data of the 21st day is predicted based on the prediction model, that is, the predicted container entry volume, and the first prediction model is adjusted and optimized based on the difference between the first prediction data and the first actual data. Then, the container entry volume of the 1st day is deleted, and the container entry volume of the 21st day (the first actual data) is added to the first data to form new first data, and then the new first data is used to train the first prediction model, and S312 and S313 are executed repeatedly until the data in the time series data are used as training data to train the first prediction model. The above method generates the first prediction model by dynamically updating the training data, gradually advancing the prediction process to reduce errors, and can significantly improve the accuracy of the prediction results. Finally, the first prediction model generated by the final training is used to predict the container entry volume on the 61st day. The predicted container entry volume can provide more accurate data support for the subsequent generation of storage plans, which is helpful to formulate more reasonable storage plans, thereby improving the operating efficiency of the container yard.

[0028] Furthermore, generating a stockpiling plan based on the predicted inbound volume includes the following steps: S3141. Acquire all storage areas of the container yard, acquire the predicted container entry quantity, acquire a first distance from the container to be entered to each storage area, randomly generate a first number of storage plans, and calculate a corresponding total moving distance for each storage plan; S3142, preset a first value, initialize the first value to zero, select a second number of candidate solutions from all stacking solutions based on the total moving distance, generate a random number that is less than the predicted entry quantity, exchange the stacking methods of the containers after the random numbers in any two candidate solutions among the candidate solutions, generate a third number of new stacking solutions, add one to the first value, and calculate the total moving distance of each new stacking solution; S3143, obtaining the minimum total moving distance, if the minimum total moving distance is greater than a preset first threshold and the first value is less than a preset second threshold, returning to S3142, otherwise, executing the next step; S3144. Obtain the minimum value of all the calculated total moving distances, and use the stacking plan corresponding to the minimum value as the final stacking plan.

[0029] Specifically, in order to generate a storage plan, all storage areas of the container yard are obtained. Assume that there are three storage areas A, B and C in total, and the predicted number of containers entering the yard is 5. The first distance from each container to be entered to each storage area is obtained, and a first number of storage plans are randomly generated. For example, 5 storage plans are randomly generated, namely: [ABACB], [BCBAC], [AACBC], [CBACB], [CABAC] and [ABCAC]. The total moving distance of each storage plan is calculated, and a first value is preset and initialized to 0. The first value is used to control the number of iterations. Assume The second number is 3. Select 3 candidate solutions from the 6 stacking solutions. The specific selection method will be explained in detail in the candidate. Assuming that the three selected candidate solutions are the first 3 candidate solutions, and assuming that the generated random number is 2, then the stacking methods after the second digit of solution 1 and solution 2 are exchanged to generate [ABBAC] and [BCACB], the stacking methods after the second digit of solution 1 and solution 3 are exchanged to generate [ABCBC] and [AAACB], and the stacking methods after the second digit of solution 2 and solution 3 are exchanged to generate [BCCBC] and [AABAC]. After generating 6 new stacking solutions, , add 1 to the first value, then calculate the total moving distance of each new stacking scheme, and obtain the minimum total moving distance of all total moving distances. If the minimum total moving distance is greater than the preset first threshold, it means that the minimum total moving distance has not yet reached the preset requirement, that is, the total moving distance of the stacking scheme currently generated is greater than the preset first threshold. At this time, it is determined whether the first value is less than the preset second threshold. If so, it means that the current number of iterations has not reached the preset second threshold, so return to S3142 and continue to iterate. Otherwise, it means that the minimum total moving distance in the generated stacking scheme is less than or equal to the preset first threshold or the number of iterations has reached the preset second threshold. At this time, it means that there is no need to continue iterating, and the stacking scheme corresponding to the minimum value in the total moving distance is used as the final stacking scheme. In order to clearly explain the above method, the above embodiment only uses several stacking areas and randomly generates several stacking schemes. In fact, there are dozens or even hundreds of stacking areas in the container yard, and the number of stacking schemes generated for the first time is greater. Therefore, the above method can generate a new stacking scheme with new features by cross-combining multiple stacking schemes, increase the diversity of stacking schemes in the iterative process, and find a better stacking scheme.

[0030] Further, selecting a second number of stockpiling solutions from all stockpiling solutions based on the total moving distance comprises the following steps: A fourth number of stockpiling solutions are randomly selected from the stockpiling solutions as the first stockpiling solutions, and a fifth number of first stockpiling solutions with the smallest total moving distance among the first stockpiling solutions are selected as candidate solutions, and this step is repeated until the number of selected candidate solutions is equal to the second number.

[0031] Specifically, in order to make the selected candidate solutions random and diverse, thereby improving the global search capability of the method and avoiding premature convergence, and finally finding a better stockpiling solution, assuming that there are 100 stockpiling solutions, assuming that 50 stockpiling solutions are to be selected as candidate solutions, first randomly select 10 stockpiling solutions from the 100 stockpiling solutions as the first stockpiling solution, obtain the 5 stockpiling solutions with the smallest total moving distance among the 10 stockpiling solutions as candidate solutions, and then randomly select 10 stockpiling solutions from the remaining 90 stockpiling solutions as the first stockpiling solution, and then select the 5 stockpiling solutions with the smallest total moving distance from the 10 stockpiling solutions as candidate solutions, repeat this step until 50 candidate solutions are selected, and end this step. The above method can simply and quickly select stockpiling solutions with diversity and relatively small total moving distances. Using these stockpiling solutions as candidate solutions not only maintains the diversity of the stockpiling solutions but also ensures that the stockpiling solutions with a small total moving distance have a higher probability of being selected.

[0032] Furthermore, generating a scheduling plan includes the following steps: S321, obtaining a first storage area for containers that need to be transported out, obtaining a destination for transshipment, planning an optimal path from the first storage area to the corresponding destination for each transshipment device, simulating transshipment of the optimal path in the digital twin model, obtaining load data of each optimal path during transshipment, and determining a potential congested section based on the load data; S322. If there is a congested section on the optimal path, obtain an idle section that can replace the congested section based on the digital twin model, and modify the congested section on the optimal path into an idle section.

[0033] Specifically, in order to generate an efficient scheduling plan, first obtain the first storage area where the containers that need to leave are located, and then obtain the destination to which these containers are to be transferred. The destination refers to the exit of the container yard where the containers are to be transferred, and then transferred to the container truck or dedicated railway at the exit and then transferred to other places. Plan the best path from the first storage area to the corresponding destination for each transfer equipment. The specific best path planning method will be explained in detail later. After that, the best path will be simulated and transferred in the digital twin model. During the transfer process, the load data of the best path is obtained. The load data refers to the number of transfer equipment passing through each section of the best path at the same time. If the number exceeds the preset threshold, such as more than 3, this section is considered to be potentially congested. Road section. During the actual transfer process, if a certain transfer equipment on this potential congested road section is delayed, it may cause delays in other transfer equipment on the entire road section. Therefore, based on obtaining a relatively idle idle road section that can replace the congested road section, the congested road section in the optimal path is modified to the above idle road section. The modified optimal path may not be the fastest path in theory, but it is a relatively stable and relatively fast path that will not have relatively problems. The modified optimal path is used as the final optimal path. After obtaining the optimal path, the corresponding optimal path can be set as the transfer path for the transfer equipment allocated to the corresponding first storage area. The above method can improve the transfer efficiency of containers and the operating efficiency of the container yard.

[0034] Furthermore, based on the congestion information, an optimal path from the storage area to the corresponding destination is planned for each transfer device, including the following steps: S3211, initializing the first list, obtaining the position of each container in the first storage area as a starting point, adding the starting point to the first list, and taking the destination to which the container is to go as the end point; S3212, taking the starting point as the first node, obtaining four second nodes adjacent to the first node, calculating the second distances of four paths from the first node through each second node to the end point, obtaining the minimum second distance therein, and adding the second node corresponding to the minimum second distance to the first list; S3213. Take the second node as the first node, return to S3212, add one of the adjacent nodes of the second node to the first list, repeat this step until the end point is added to the first list, and take the path composed of all the nodes in the first list as the best path.

[0035] Specifically, there may be multiple containers that need to be shipped out of the container yard in the first storage area. Different containers may need to be transferred to different destinations. The location of the container is taken as the starting point, and the destination to which the container needs to be transferred is taken as the end point. The first distance between the starting point and the end point is obtained. The first distance is the total distance of any path that can be from the starting point to the end point. For each container, a first list is initialized. The starting point is first added to the first list, and the starting point is taken as the first node. Then the second node adjacent to the starting point is obtained. In the container yard, a node usually has four adjacent nodes above, below, left and right. Then the second distance of the four paths from the first node through each second node to the end point is calculated, and the minimum second distance is obtained. The second node corresponding to the minimum second distance is the next node from the first node to the end point, so the second node corresponding to the minimum second distance is added to the first list, and then the second node is taken as the first node. Return to execute S3212, obtain the next node of the second node, and repeat S3213 until the end point is added to the first list. Then the path composed of all nodes stored in the first list is the best path from the starting point to the end point.

[0036] Furthermore, before constructing the digital twin model of the container yard based on the collected relevant information, the following steps are performed: A reference range for each data parameter is stored in advance, and each data parameter in the relevant information is compared with the corresponding reference range. If the data parameter is within the corresponding reference range, the data parameter is marked as normal, otherwise, the data parameter is marked as abnormal. For the data parameter marked as abnormal, a data error message is sent to the data sender. After receiving the data error message, the data sender obtains the correct data parameter and the corresponding correction code and sends them to the data receiver. The data receiver determines whether the received new data parameter is normal. If it is normal, this step is terminated. Otherwise, the data parameter is corrected based on the correction code.

[0037] Specifically, before building a digital twin model, the data parameters in the relevant information obtained may be wrong or abnormal. If the digital twin model is built directly based on these data parameters, the constructed digital twin model may be inaccurate. Therefore, the reference range of each data parameter is stored in advance, and each data parameter of the obtained relevant parameters is compared with the corresponding reference range. If it is within the corresponding reference range, it means that the data parameter is correct and marked as normal. If the data parameter is not within the corresponding reference range, it means that the data parameter is wrong and it is marked as abnormal. For the data parameter marked as abnormal, data error information is sent to the data sender. The data sender may be a sensor or a data collection module. After receiving the data error information, the data sender sends the correct data parameter to the data receiver. In order to prevent errors during the transmission process, the correction code corresponding to the data parameter is also sent to the data receiver. The data receiver refers to the digital twin model construction module. After receiving the data parameter, the data receiver compares it with the corresponding reference range again to determine whether it is normal. If it is normal, this step is terminated. Otherwise, it means that the data is still abnormal. At this time, the received data parameter is corrected based on the correction code to obtain the correct data parameter, where the correction code can use a Chinese name code.

[0038] In summary, the present invention provides a method for constructing and monitoring a digital twin model of a container yard based on the Internet of Things, the method comprising real-time collection of relevant information of the container yard, construction of a digital twin model of the container yard; obtaining dynamic data of the container yard in real time based on sensors, the dynamic data being able to represent the operating status of the container yard, and updating the digital twin model in real time based on the dynamic data; obtaining monitoring results of the container yard based on the digital twin model, generating a stockpiling plan and a scheduling plan based on the monitoring results; simulating the stockpiling plan and the scheduling plan in the digital twin model to obtain simulation results, optimizing the stockpiling plan and the scheduling plan based on the simulation results, and re-simulating the optimized plan in the digital twin model until the optimal stockpiling plan and the optimal scheduling plan are obtained. The present invention can construct an accurate digital twin model, and can also improve the operating efficiency of the container yard based on the digital twin model.

[0039] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0040] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the above-mentioned program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0041] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0042] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0043] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for constructing and monitoring a digital twin model of a container yard based on the Internet of Things, characterized in that: The steps include: S1. Collect relevant information of the container yard in real time based on the Internet of Things technology, wherein the relevant information includes layout information and equipment information, wherein the equipment information includes container information and transshipment equipment information, wherein the container information includes the location, container number, container type, empty weight status and transportation task of the container, and wherein the transshipment equipment information includes reach stacker information and container truck information, and construct a digital twin model of the container yard based on the collected relevant information; S2. Acquire dynamic data of the container yard in real time based on sensors, where the dynamic data can represent the operating status of the container yard, including loading, unloading, handling and storage of containers, and send the dynamic data to the digital twin model, which processes and analyzes the dynamic data and updates the digital twin model in real time based on the dynamic data; S3. Acquire monitoring results of the container yard based on the digital twin model, analyze the container storage conditions of the container yard based on the monitoring results to generate a storage plan, and analyze the transshipment equipment based on the monitoring results to generate a scheduling plan; S4. Simulate the stockpiling plan and the scheduling plan in the digital twin model to obtain simulation results, optimize the stockpiling plan and the scheduling plan based on the simulation results, and re-simulate the optimized stockpiling plan and the scheduling plan in the digital twin model until the optimal stockpiling plan and the optimal scheduling plan are obtained.

2. The method according to claim 1, characterized in that Generate a stockpile plan, including the following steps: S311, determine the time length, obtain the container entry volume of the container yard for each of the time lengths in the past, detect the container entry volume to obtain abnormal data therein, replace the abnormal data in the container entry volume with the average of the data before and after the abnormal data, and then normalize the processed data of the container entry volume, and generate time series data for the normalized container entry volume according to the chronological order of collection time; S312, obtaining first data within a first time interval from the time series data, using the first data as training data to train a first prediction model, predicting first prediction data of the next time length of the first time interval based on the first prediction model, obtaining first actual data corresponding to the first prediction data, and adjusting and optimizing the first prediction model based on the difference between the first prediction data and the first actual data; S313, deleting the first container entry quantity with the earliest collection time in the first data, and adding the first actual data input to the first data, using the updated first data as new training data to continue training the first prediction model, and repeatedly executing S312 until all the data in the time series data are used as training data to train the first prediction model, and using the first prediction model generated by the final training to predict the container entry quantity of the future time length; S314: Generate a storage plan based on the container entry quantity.

3. The method according to claim 2, characterized in that Generating a stockpile plan based on the predicted inflow volume includes the following steps: S3141, obtaining all storage areas of the container yard, obtaining the predicted container entry quantity, obtaining a first distance from the container to be entered to each storage area, randomly generating a first number of storage plans, and calculating a corresponding total moving distance for each storage plan; S3142: preset a first value, initialize the first value to zero, select a second number of candidate solutions from all the stacking solutions based on the total moving distance, generate a random number that is less than the predicted entry quantity, exchange the stacking methods of the containers after the random numbers in any two of the candidate solutions, generate a third number of new stacking solutions, add one to the first value, and calculate the total moving distance of each new stacking solution; S3143, obtaining the minimum total moving distance, if the minimum total moving distance is greater than a preset first threshold and the first value is less than a preset second threshold, returning to S3142, otherwise, executing the next step; S3144. Obtain the minimum value of all the calculated total moving distances, and use the stacking scheme corresponding to the minimum value as the final stacking scheme.

4. The method according to claim 3, characterized in that Selecting a second number of stockpiling solutions from all the stockpiling solutions based on the total moving distance comprises the following steps: A fourth number of stockpiling schemes are randomly selected from the stockpiling schemes as the first stockpiling schemes, and a fifth number of the first stockpiling schemes with the smallest total moving distances among the first stockpiling schemes are selected as candidate schemes, and this step is repeated until the number of the selected candidate schemes is equal to the second number.

5. The method according to claim 1, characterized in that Generate a scheduling plan, including the following steps: S321, obtaining a first storage area for containers that need to be transported out, obtaining a destination for transshipment, planning an optimal path from the first storage area to the corresponding destination for each transshipment device, simulating transshipment of the optimal path in the digital twin model, obtaining load data of each optimal path during transshipment, and determining a potential congested section based on the load data; S322: If the congested section exists in the optimal path, an idle section that can replace the congested section is obtained based on the digital twin model, and the congested section of the optimal path is modified into the idle section.

6. The method according to claim 5, characterized in that Planning an optimal path from the first storage area to the corresponding destination for each transfer device includes the following steps: S3211, initializing the first list, obtaining the position of each container in the first storage area as a starting point, adding the starting point to the first list, and taking the destination to which the container is to go as an end point; S3212: take the starting point as the first node, obtain four second nodes adjacent to the first node, calculate the second distances of four paths from the first node through each of the second nodes to the end point, obtain the minimum second distance therebetween, and add the second node corresponding to the minimum second distance to the first list; S3213. Take the second node as the first node, return to S3212, add one of the adjacent nodes of the second node to the first list, repeat this step until the end point is added to the first list, and take the path composed of all the nodes in the first list as the optimal path.

7. The method according to claim 1, characterized in that Before constructing the digital twin model of the container yard based on the collected relevant information, the following steps are further performed: A reference range for each data parameter is stored in advance, and each data parameter in the relevant information is compared with the corresponding reference range. If the data parameter is within the corresponding reference range, the data parameter is marked as normal; otherwise, the data parameter is marked as abnormal; for the data parameter marked as abnormal, a data error message is sent to the data sender; after receiving the data error message, the data sender obtains the correct data parameter and the corresponding correction code and sends them to the data receiver; the data receiver determines whether the received new data parameter is normal; if so, this step is terminated; otherwise, the data parameter is corrected based on the correction code.

Images