Warehouse logistics scheduling system and method based on digital twin model

Abstract

The invention relates to the technical field of warehouse digital management, in particular to a warehouse logistics scheduling method and system based on a digital twinborn model. The system mainly comprises a data fusion module, a model establishment module, a cargo judgment module, a scheduling sequence generation module and a stacking module, parameter data are obtained through the modules, and multiple types of parameter data are fused and encrypted to generate twin data; mapping the decrypted twinborn data with buildings, goods shelves, goods and carrying equipment in a warehouse to generate a storage digital twinborn model; a temperature threshold value, a first shelving threshold value and a second shelving threshold value are set to obtain a first scheduling sequence and a second scheduling sequence, so that during scheduling, a certain scheduling sequence exists no matter whether each cargo is put in a warehouse or taken out of the warehouse, and the cargo with a higher priority level can be scheduled in advance.

Description

Technical Field

[0001] The present invention relates to the technical field of warehouse digital management, and in particular to a warehouse logistics scheduling system and method based on a digital twin model. Background Art

[0002] A digital twin model is a virtual copy of an entity or system created using digital data and technology. This model can be used to simulate, analyze and optimize physical entities or systems in the real world, such as buildings, equipment, supply chains, etc. Digital twin models are usually combined with technologies such as the Internet of Things, big data analysis, and artificial intelligence to help organizations better understand and manage their physical assets and improve efficiency and sustainability.

[0003] In the existing field of warehousing and logistics technology, digital twin technology can monitor and analyze the operating status, performance parameters and other information of logistics warehousing in real time, providing strong support for the intelligence of equipment. By mining and analyzing the operating data of equipment, data support can be provided for the optimization and improvement of equipment, further improving the intelligence level of equipment. However, there is still a lack of certain planning for the dispatch of goods. All goods are placed on the shelves in the order of storage. There is no reasonable dispatch system for some goods that need to be shipped out first, such as fresh goods. As a result, some goods that need to be shipped out first are backlogged or the shipping time is long, which delays time and causes losses. Summary of the invention

[0004] The purpose of the present invention is to provide a warehouse logistics scheduling system and method based on a digital twin model to solve the problem in the prior art that there is no reasonable scheduling system for goods that need to be shipped out first.

[0005] The embodiments of the present invention are implemented by the following technical solutions:

[0006] In a first aspect, the present invention provides a warehouse logistics scheduling system based on a digital twin model, comprising a data fusion module, a model building module, a cargo judgment module, a scheduling sequence generation module and a stacking module connected to the main control device;

[0007] A data fusion module is configured to obtain parameter data, classify the parameter data, and fuse and encrypt multiple types of parameter data to generate twin data;

[0008] A model building module is configured to map the decrypted twin data with buildings, shelves, goods, and handling equipment in the warehouse to generate a warehouse digital twin model;

[0009] The cargo judgment module is configured to obtain the type of current cargo through the warehouse digital twin model and judge whether the cargo is fresh cargo or non-fresh cargo;

[0010] The scheduling sequence generation module is configured to obtain the temperature and placement time of the current goods if the goods are fresh goods, and set a temperature threshold and a first holding threshold, and set a first scheduling sequence according to the temperature threshold and the first holding time; if the goods are judged to be non-fresh goods, set a second holding threshold, and set a second scheduling sequence according to the second holding threshold, and the priority of the first scheduling sequence is greater than that of the second scheduling sequence;

[0011] A stacking module, configured to perform sequential stacking based on a warehouse digital twin model through a first scheduling sequence and a second scheduling sequence;

[0012] The main control device is used to control the execution of the data fusion module, the model building module, the cargo judgment module, the scheduling sequence generation module and the stacking module.

[0013] Preferably, the classifying the parameter data and fusing and encrypting multiple types of parameter data to generate twin data includes:

[0014] The parameter data includes numerical data, text data and media data, and the data is cleaned;

[0015] Perform global initialization of parameter data and generate a key pair. Send the globally initialized parameter data and public key to the feature extraction module, which is used to extract data features from numerical data, text data, and media data, and fuse the extracted data features to output twin data.

[0016] The twin data is divided into several encrypted secret sharing fragments through secret sharing and homomorphic encryption, and the encrypted secret sharing fragments are summed up through the aggregation server, and the sum is sent to the model building module;

[0017] The model building module obtains secret shared fragments by decrypting the sum value with a private key, and restores the secret shared fragments through a secret reconstruction algorithm to obtain twin data.

[0018] Preferably, the data cleaning includes:

[0019] Establishing the original data sequence through parameter data and performing singular spectrum analysis on the original data sequence;

[0020] The original data sequence is mapped into a number of vectors, and the window length is set, and the original data sequence is lagged and arranged according to the window length to generate a first trajectory matrix;

[0021] Performing singular value decomposition on the first trajectory matrix to obtain a second trajectory matrix, and dividing the second trajectory matrix into m mutually non-overlapping subsets to obtain a third trajectory matrix;

[0022] Diagonally average the third trajectory matrix to transform the two-dimensional matrix into a one-dimensional sequence;

[0023] Set the cumulative contribution rate threshold, select the first t components greater than the cumulative contribution rate for reconstruction, and obtain the reconstructed data sequence;

[0024] Calculate the residuals between the reconstructed data sequence and the original data sequence to obtain the residual data sequence, and calculate the statistical mean and mean square error of the residual data sequence. Calculate the absolute value of the difference between the random data and the statistical mean. When the ratio of the absolute value to the mean square error is greater than three, identify the random data as an outlier and eliminate it.

[0025] Preferably, the fusing of the extracted data features includes:

[0026] Establishing a memory unit, wherein the memory unit is a high-order tensor with a heterogeneous data feature space, and each mode of the memory unit corresponds to a spatial mapping of a data feature;

[0027] By modularly multiplying the data features with the feature space corresponding to the memory unit, the fused twin data is obtained.

[0028] Preferably, the mapping to generate a warehouse digital twin model includes:

[0029] Map warehouse and parameter data using the order of data, events and logic;

[0030]

[0031] In the formula, To store equipment operating data, is the event set of the conveying system; DDM is the logic of the warehouse management system;

[0032] T pv ={S i ∪R i ∪C i}

[0033] Where, T pv It is the operation process of warehouse virtual-real mapping; S i is the virtual model set; R i is the twin data mapping set; C i is the data communication interface; i is the number of smart devices in the set.

[0034] Preferably, setting the first scheduling order by the temperature threshold and the first hold time includes:

[0035] When the temperature of the cargo is greater than the temperature threshold and the storage time is greater than the first holding time, it is classified as the highest priority scheduling level;

[0036] When the temperature of the goods is lower than the temperature threshold and the storage time is longer than the first holding time, it is classified as a strong priority scheduling level;

[0037] When the temperature of the cargo is greater than the temperature threshold and the storage time is less than the first holding time, it is classified as the second priority scheduling level;

[0038] When the temperature of the goods is lower than the temperature threshold and the storage time is lower than the first holding time, it is classified as a weak priority scheduling level.

[0039] Preferably, setting the second scheduling order by the second suspension threshold comprises:

[0040] When the current storage time of the goods is greater than the second hold threshold, it is classified as the strong severity level;

[0041] When the current storage time of the goods is less than the second holding threshold, it is classified as a weak scheduling level.

[0042] Preferably, it also includes a damage identification module;

[0043] The damage identification module is configured to identify whether the current goods are damaged. If damaged, the current goods are identified as a special scheduling level higher than the highest priority scheduling level.

[0044] Preferably, the sequential stacking based on the warehouse digital twin model includes:

[0045] Obtain the pick-up and placement points through the warehouse digital twin model, and generate the shortest walking path through the pick-up and placement points;

[0046] Obtain the dispatch order of goods, and set the placement points of goods from near to far from the warehouse exit according to the dispatch order of goods;

[0047] If the dispatch order of two or more goods is the same, they will be picked up in order from near to far according to the distance between the stacking module and the goods in the same dispatch order.

[0048] In a second aspect, the present invention also provides a warehouse logistics scheduling method based on a digital twin model, including using the above-mentioned warehouse logistics scheduling system based on a digital twin model to schedule warehouse logistics.

[0049] The technical solution of the embodiment of the present invention has at least the following advantages and beneficial effects:

[0050] The system provided by the present invention mainly includes a data fusion module, a model building module, a cargo judgment module, a scheduling sequence generation module and a stacking module. Through the above modules, parameter data is obtained, and multiple types of parameter data are fused and encrypted to generate twin data; the decrypted twin data is mapped with the buildings, shelves, cargo and handling equipment in the warehouse to generate a warehouse digital twin model; and by setting the temperature threshold, the first shelving threshold and the second shelving threshold, the first scheduling sequence and the second scheduling sequence are obtained. In this way, when scheduling, whether it is warehousing or out of the warehouse, each cargo has a certain scheduling sequence. Cargo with a higher priority level can be scheduled in advance, or stored in the appropriate location in advance, thus forming a more reasonable scheduling system to avoid the shelving of some cargo that needs priority scheduling and minimize losses as much as possible. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0052] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0054] The terms "first, second" and the like in the specification and claims of this application and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. The naming or numbering of steps in this application does not mean that the steps in the method flow must be executed in the time / logical sequence indicated by the naming or numbering. The named or numbered process steps can be changed in the execution order according to the technical purpose to be achieved, as long as the same or similar technical effects can be achieved.

[0055] The division of modules in this application is a logical division. There may be other division methods when implemented in actual applications. For example, multiple modules can be combined or integrated into another system, or some features can be ignored or not executed.

[0056] Please refer to Figure 1 As shown, the present invention provides a warehousing logistics scheduling system based on a digital twin model, including a data fusion module, a model building module, a cargo judgment module, a scheduling sequence generation module and a stacking module connected to the main control device;

[0057] A data fusion module is configured to obtain parameter data, classify the parameter data, and fuse and encrypt multiple types of parameter data to generate twin data;

[0058] Among them, parameter data includes numerical data, text data and media data. Numerical data is mainly collected from sensor devices, and its structure type is often JSON or XML, with a data accuracy of 0.1%;

[0059] Text data is collected from storage systems and device controllers, and its structure types include NoSQL and control instructions, with a data accuracy of 1%; media data is mainly collected from monitoring systems and visual systems, and its structure types include files and data streams.

[0060] A model building module is configured to map the decrypted twin data with buildings, shelves, goods, and handling equipment in the warehouse to generate a warehouse digital twin model;

[0061] The cargo judgment module is configured to obtain the type of current cargo through the warehouse digital twin model and judge whether the cargo is fresh cargo or non-fresh cargo;

[0062] The scheduling sequence generation module is configured to obtain the temperature and placement time of the current goods if the goods are fresh goods, and set a temperature threshold and a first holding threshold, and set a first scheduling sequence according to the temperature threshold and the first holding time; if the goods are judged to be non-fresh goods, set a second holding threshold, and set a second scheduling sequence according to the second holding threshold, and the priority of the first scheduling sequence is greater than that of the second scheduling sequence;

[0063] A stacking module, configured to perform sequential stacking based on a warehouse digital twin model through a first scheduling sequence and a second scheduling sequence;

[0064] Among them, the stacking module mainly includes: data integration function, which obtains various operation data in real time through sensor devices such as the Internet of Things, and uploads them to the data integration module in a centralized manner; virtual operation scene construction, the virtual operation scene construction module obtains the demand data from the data integration module, and constructs a virtual scene for the operation of the stacking equipment; intelligent stacking operation, which issues operation data and special instructions to the stacking equipment through the PLC control system to drive the movement and operation of the stacking equipment; visual auxiliary services, which feed back the collision detection results and other information to the physical space in real time, and the physical entity controls the stacking equipment accordingly based on the feedback information.

[0065] The main control device is used to control the execution of the data fusion module, the model building module, the cargo judgment module, the scheduling sequence generation module and the stacking module.

[0066] In one embodiment of the present invention, classifying the parameter data and fusing and encrypting multiple types of parameter data to generate twin data includes:

[0067] The parameter data includes numerical data, text data and media data, and the data is cleaned;

[0068] Perform global initialization of parameter data and generate a key pair. Send the globally initialized parameter data and public key to the feature extraction module, which is used to extract data features from numerical data, text data, and media data, and fuse the extracted data features to output twin data.

[0069] The twin data is divided into several encrypted secret sharing fragments through secret sharing and homomorphic encryption, and the encrypted secret sharing fragments are summed up through the aggregation server, and the sum is sent to the model building module;

[0070] Among them, homomorphic encryption preferably adopts Paillier homomorphic encryption.

[0071] Specifically, in the key generation, two large prime numbers are selected, and the values ​​of N and λ are calculated according to: N = pq, λ = lcm(p-1, q-1), where lcm(·, ·) represents the least common multiple, where q and p are two prime numbers, λ is a security parameter, and N is the product of pq;

[0072] Select a random number g, define the function L(x) = (x-1) / N, and calculate:

[0073] μ=L(g λ modN 2 ) -1 mod N

[0074] The public key is pk=(N, g), the private key is sk=(λ, μ), μ is the encryption factor of the plaintext message, k is a constant, x is the result value to be solved, and s is the secret value.

[0075] Get the plaintext to be encrypted as m, select a random number r, and calculate the ciphertext c=g λ r n mod N 2 , n is a constant.

[0076] After receiving the ciphertext, the plaintext can be calculated:

[0077] m=μL(c λmodN 2 )mod / N

[0078] Among them, secret sharing preferably adopts Shamir to split the plaintext into several equal parts. Specifically, t-1 random values ​​r1, r2, ..., r are randomly selected. t-1 , construct a polynomial:

[0079] f(x)=s+r1x1+…+r t-1 x t-1

[0080] Randomly select n random numbers and calculate the secret fragment, x1...x t-1 The plain text is split into several equal parts, and the results are sent to the aggregation server for addition, and the sum is sent to the model building module.

[0081] Through the secret reconstruction algorithm, the model building module can recover the secret fragments through Lagrangian interpolation.

[0082]

[0083] The model building module obtains secret shared fragments by decrypting the sum value with a private key, and restores the secret shared fragments through a secret reconstruction algorithm to obtain twin data.

[0084] In one embodiment of the present invention, the data cleaning includes:

[0085] Create the original data sequence X through parameter data data {x1, x2, ..., x n}, and perform singular spectrum analysis on the original data sequence;

[0086] The original data sequence is mapped into a vector sequence with a window length of L, forming a K vector with a length of L. That is, the data sequence is lagged and arranged according to the window length L, and the generated trajectory matrix X is:

[0087]

[0088] Wherein, the window length L (2≤L≤N) should not exceed N / 3 as much as possible and can be an integer multiple of the data period; 1≤i≤K, N is a natural number.

[0089] The first trajectory matrix is ​​subjected to singular value decomposition to obtain the second trajectory matrix.

[0090] Specifically, S = XX T , we get L eigenvalues ​​λ1≥λ2≥…≥λ L ≥0 and the corresponding orthogonal eigenvector, where T is the standard deviation multiple, which is used to determine the threshold of outliers. Let d = rank (X), the first trajectory matrix is ​​decomposed into:

[0091] X=X1+X2+…+X d

[0092] In the formula, λ i is the i-th eigenvalue, U i is the i-th eigenvalue λ i The corresponding orthogonal eigenvectors, V i is the independent variable, V i The calculation formula is

[0093] The second trajectory matrix is ​​divided into m mutually non-overlapping subsets to obtain a third trajectory matrix, and the third trajectory matrix is ​​diagonally averaged to convert the two-dimensional matrix into a one-dimensional sequence. At this time, the original data sequence can be expressed as the sum of several one-dimensional sequences.

[0094] Set the cumulative contribution rate threshold to 85%, select the first t components greater than the cumulative contribution rate for reconstruction, and obtain the reconstructed data sequence;

[0095] The contribution rate is calculated as:

[0096]

[0097] Among them, λ j is the jth eigenvalue, λ i is the i-th eigenvalue.

[0098] Calculate the residuals between the reconstructed data sequence and the original data sequence to obtain the residual data sequence, and calculate the statistical mean and mean square error of the residual data sequence. Calculate the absolute value of the difference between the random data and the statistical mean. When the ratio of the absolute value to the mean square error is greater than three, identify the random data as an outlier and eliminate it.

[0099] The calculation formula is:

[0100]

[0101] In the formula, x si is a random value, x sa Statistical mean, σ s is the mean square error.

[0102] Preferably, the fusing of the extracted data features includes:

[0103] Establishing a memory unit, wherein the memory unit is a high-order tensor with a heterogeneous data feature space, and each mode of the memory unit corresponds to a spatial mapping of a data feature;

[0104] By modularly multiplying the data features with the feature space corresponding to the memory unit, the fused twin data is obtained.

[0105] For example: the memory unit W is modularly multiplied with the data feature z1 of the numerical data along the first order to obtain a new memory unit W with z1 (1) , W (1) Perform modular multiplication along the second order with the data feature z2 of the text class data to obtain a new memory unit W with z2 (2) , W (2) Modular multiplication is performed along the third order with the data feature z3 of the media data to obtain a fused tensor Z having z1, z2 and z3.

[0106] Z=[(W×z1)×z2]×z3

[0107] Preferably, the mapping to generate a warehouse digital twin model includes:

[0108] Map warehouse and parameter data using the order of data, events and logic;

[0109]

[0110] In the formula, To store equipment operating data, is the event set of the conveying system; DDM is the logic of the warehouse management system;

[0111] T pv ={S i ∪R i ∪C i}

[0112] Where, T pv It is the operation process of warehouse virtual-real mapping; S i is the virtual model set; R i is the twin data mapping set; C i is the data communication interface; i is the number of smart devices in the set.

[0113] In one embodiment of the present invention, setting the first scheduling order by the temperature threshold and the first hold time includes:

[0114] When the temperature of the goods is greater than the temperature threshold and the storage time is greater than the first holding time, it is classified as the highest priority scheduling level; when the temperature of the goods is less than the temperature threshold and the storage time is greater than the first holding time, it is classified as the strong priority scheduling level; when the temperature of the goods is greater than the temperature threshold and the storage time is less than the first holding time, it is classified as the second priority scheduling level; when the temperature of the goods is less than the temperature threshold and the storage time is less than the first holding time, it is classified as the weak priority scheduling level.

[0115] Secondly, setting the second scheduling order by the second suspension threshold comprises:

[0116] When the current storage time of the goods is greater than the second hold threshold, it is classified as the strong severity level;

[0117] When the current storage time of the goods is less than the second holding threshold, it is classified as a weak scheduling level.

[0118] For example, to determine the type of goods, a QR code or other medium that can read information can be set on the goods, and the basic information of the goods can be saved on the medium, and then mapped to the warehouse digital twin model. When the current goods are judged to be fresh goods, if they are ready to be put into storage, the current storage time of the goods is obtained to determine whether it is greater than the first storage threshold, and then the corresponding priority level is obtained for priority storage.

[0119] If the goods are shipped out of the warehouse, it is necessary to obtain the current temperature and storage time for comprehensive judgment, whether it is greater than or less than the temperature threshold, whether it is greater than or less than the first storage threshold, and obtain the priority level of the current goods.

[0120] If it is determined that the current goods are non-fresh goods, the holding time is directly obtained to determine whether it is greater than the second holding threshold to obtain the scheduling level.

[0121] Among them, the scheduling urgency of the highest priority scheduling level, strong priority scheduling level, second priority scheduling level, weak priority scheduling level, strong urgency level and weak scheduling level decreases in sequence.

[0122] Secondly, it also includes a damage identification module;

[0123] The damage identification module is configured to identify whether the current goods are damaged. If damaged, the current goods are identified as a special scheduling level higher than the highest priority scheduling level.

[0124] In this embodiment, the sequential stacking based on the warehouse digital twin model includes:

[0125] The pick-up points and placement points are obtained through the warehouse digital twin model, and the shortest walking path is generated through the pick-up points and placement points; the scheduling order of the goods is obtained, and the placement points of the goods are set from near to far from the warehouse exit according to the scheduling order of the goods. In this way, when the goods are shipped out, the goods with high priority can be transferred out of the warehouse more quickly.

[0126] If the dispatch order of two or more goods is the same, they will be picked up in order from near to far according to the distance between the stacking module and the goods in the same dispatch order.

[0127] In a second aspect, the present invention also provides a warehouse logistics scheduling method based on a digital twin model, including using the above-mentioned warehouse logistics scheduling system based on a digital twin model to schedule warehouse logistics.

[0128] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0129] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A warehouse logistics scheduling system based on a digital twin model, characterized in that: It includes a data fusion module, a model building module, a cargo judgment module, a scheduling sequence generation module and a stacking module connected to the main control device; A data fusion module is configured to obtain parameter data, classify the parameter data, and fuse and encrypt multiple types of parameter data to generate twin data; A model building module is configured to map the decrypted twin data with buildings, shelves, goods, and handling equipment in the warehouse to generate a warehouse digital twin model; The cargo judgment module is configured to obtain the type of current cargo through the warehouse digital twin model and judge whether the cargo is fresh cargo or non-fresh cargo; The scheduling sequence generation module is configured to obtain the temperature and placement time of the current goods if the goods are fresh goods, and set a temperature threshold and a first holding threshold, and set a first scheduling sequence according to the temperature threshold and the first holding time; if the goods are judged to be non-fresh goods, set a second holding threshold, and set a second scheduling sequence according to the second holding threshold, and the priority of the first scheduling sequence is greater than that of the second scheduling sequence; A stacking module, configured to perform sequential stacking based on a warehouse digital twin model through a first scheduling sequence and a second scheduling sequence; The main control device is used to control the execution of the data fusion module, the model building module, the cargo judgment module, the scheduling sequence generation module and the stacking module.

2. According to the digital twin model-based warehouse logistics scheduling system of claim 1, it is characterized in that: The method of classifying the parameter data and fusing and encrypting multiple types of parameter data to generate twin data includes: The parameter data includes numerical data, text data and media data, and the data is cleaned; Perform global initialization of parameter data and generate a key pair. Send the globally initialized parameter data and public key to the feature extraction module, which is used to extract data features from numerical data, text data, and media data, and fuse the extracted data features to output twin data. The twin data is divided into several encrypted secret sharing fragments through secret sharing and homomorphic encryption, and the encrypted secret sharing fragments are summed up through the aggregation server, and the sum is sent to the model building module; The model building module obtains secret shared fragments by decrypting the sum value with a private key, and restores the secret shared fragments through a secret reconstruction algorithm to obtain twin data.

3. According to the warehousing logistics scheduling system based on the digital twin model according to claim 2, it is characterized in that: The data cleaning includes: Establishing the original data sequence through parameter data and performing singular spectrum analysis on the original data sequence; The original data sequence is mapped into a number of vectors, and the window length is set, and the original data sequence is lagged and arranged according to the window length to generate a first trajectory matrix; Performing singular value decomposition on the first trajectory matrix to obtain a second trajectory matrix, and dividing the second trajectory matrix into m mutually non-overlapping subsets to obtain a third trajectory matrix; Diagonally average the third trajectory matrix to transform the two-dimensional matrix into a one-dimensional sequence; Set the cumulative contribution rate threshold, select the first t components greater than the cumulative contribution rate for reconstruction, and obtain the reconstructed data sequence; Calculate the residuals between the reconstructed data sequence and the original data sequence to obtain the residual data sequence, and calculate the statistical mean and mean square error of the residual data sequence. Calculate the absolute value of the difference between the random data and the statistical mean. When the ratio of the absolute value to the mean square error is greater than three, identify the random data as an outlier and eliminate it.

4. According to the warehousing logistics scheduling system based on the digital twin model according to claim 2, it is characterized in that: The fusing of the extracted data features includes: Establishing a memory unit, wherein the memory unit is a high-order tensor with a heterogeneous data feature space, and each mode of the memory unit corresponds to a spatial mapping of a data feature; By modularly multiplying the data features with the feature space corresponding to the memory unit, the fused twin data is obtained.

5. According to the warehousing logistics scheduling system based on the digital twin model according to claim 1, it is characterized in that: The mapping to generate a warehouse digital twin model includes: Map warehouse and parameter data using the order of data, events and logic; In the formula, To store equipment operating data, is the event set of the conveying system; DDM is the logic of the warehouse management system; T pv ={S i ∪R i ∪C i } Where, T pv It is the operation process of warehouse virtual-real mapping; S i is the virtual model set; R i is the twin data mapping set; C i is the data communication interface; i is the number of smart devices in the set.

6. According to the digital twin model-based warehouse logistics scheduling system of claim 2, it is characterized in that: The setting of the first scheduling order by the temperature threshold and the first hold time comprises: When the temperature of the cargo is greater than the temperature threshold and the storage time is greater than the first holding time, it is classified as the highest priority scheduling level; When the temperature of the goods is lower than the temperature threshold and the storage time is longer than the first holding time, it is classified as a strong priority scheduling level; When the temperature of the cargo is greater than the temperature threshold and the storage time is less than the first holding time, it is classified as the second priority scheduling level; When the temperature of the goods is lower than the temperature threshold and the storage time is lower than the first holding time, it is classified as a weak priority scheduling level.

7. The warehousing logistics scheduling system based on the digital twin model according to claim 6 is characterized in that: The setting of the second scheduling order by the second suspension threshold comprises: When the current storage time of the goods is greater than the second hold threshold, it is classified as the strong severity level; When the current storage time of the goods is less than the second holding threshold, it is classified as a weak scheduling level.

8. The warehousing logistics scheduling system based on the digital twin model according to claim 7 is characterized in that: Also includes a damage recognition module; The damage identification module is configured to identify whether the current goods are damaged. If damaged, the current goods are identified as a special scheduling level higher than the highest priority scheduling level.

9. The warehousing logistics scheduling system based on the digital twin model according to claim 1 is characterized in that: The sequential stacking based on the warehouse digital twin model includes: Obtain the pick-up and placement points through the warehouse digital twin model, and generate the shortest walking path through the pick-up and placement points; Obtain the dispatch order of goods, and set the placement points of goods from near to far from the warehouse exit according to the dispatch order of goods; If the dispatch order of two or more goods is the same, they will be picked up in order from near to far according to the distance between the stacking module and the goods in the same dispatch order.

10. A warehouse logistics scheduling method based on a digital twin model, characterized in that: It includes using a warehouse logistics scheduling system based on a digital twin model as described in any one of claims 1 to 9 to schedule warehouse logistics.

Images