Modeling and Simulation Method and Device for Production Process

Abstract

Embodiments of the present disclosure propose a method for modeling and simulating a production process, including: obtaining production input data of a current simulation case; determining a similarity between the current simulation case and each historical simulation case among at least one historical simulation case based on the production input data of the current simulation case and the production input data of at least one historical simulation case, wherein the production input data of at least one historical simulation case is stored in a database; and generating a simulation model of the current simulation case by using the simulation models of at least one historical simulation case based on the determined similarity to simulate the production process. The above method can use the simulation models of similar historical simulation cases to model the production process, thereby greatly improving the modeling efficiency, saving the time and effort of engineers to a great extent, and ensuring the consistency and accuracy of the simulation model.

Description

Technical Field

[0001] The present disclosure relates to the technical field of industrial manufacturing, and more particularly, to a method, apparatus, computing device, computer-readable storage medium, and program product for modeling and simulating a production process. Background Art

[0002] With the continuous improvement of the digitalization level of discrete manufacturing processes, the demand for simulating production processes is also increasing continuously, so as to identify production bottlenecks in the workshop, pursue higher production capacity and lower manufacturing costs. However, it is not easy to model the production process in discrete manufacturing. Currently, engineers usually manually build a simulation model of the production process, set the parameters of the simulation model, and write the internal logic of the simulation model in simulation software. Engineers can rely on experience to select the simulation models of historical simulation cases to assist in building the simulation model. In addition, when performing simulation benchmark analysis, engineers usually rely on experience to select some historical simulation cases and compare the simulation inputs and outputs of these historical simulation cases with those of the current simulation case. Summary of the Invention

[0003] The above-mentioned modeling method of the production process consumes a lot of time and energy of engineers for logic writing and parameter setting. Moreover, the accuracy of the model highly depends on the modeling experience of engineers and the domain knowledge of the manufacturing process. In addition, with the continuous increase of simulation cases, the amount of simulation data is getting larger and larger, and it is difficult to manage these simulation data. In addition, it is also difficult to effectively use historical simulation models and historical simulation data in a unified standard to assist in building a new simulation model and performing simulation benchmark analysis.

[0004] The first embodiment of the present disclosure provides a method for modeling and simulating a production process, including: obtaining production input data of a current simulation case; determining the similarity between the current simulation case and each of at least one historical simulation case based on the production input data of the current simulation case and the production input data of at least one historical simulation case, wherein the production input data of the at least one historical simulation case is stored in a database; and generating a simulation model of the current simulation case by using the simulation models of the at least one historical simulation case based on the determined similarity to simulate the production process.

[0005] In this embodiment, by comparing the similarity between the current simulation case and the historical simulation cases, the simulation models of similar historical simulation cases can be used to model the production process, thereby greatly improving the modeling efficiency, saving a great deal of time and energy of engineers to a large extent, and ensuring the consistency and accuracy of the simulation model. In addition, by recommending the simulation models of similar historical simulation cases to engineers, the requirements and dependence on personal experience and knowledge in the modeling process are reduced.

[0006] The second embodiment of the present disclosure proposes a modeling and simulation device for a production process, including: an input data acquisition unit configured to acquire production input data of a current simulation case; a simulation case comparison unit configured to determine the similarity between the current simulation case and each of at least one historical simulation case based on the production input data of the current simulation case and the production input data of at least one historical simulation case; and a simulation model generation unit configured to generate a simulation model of the current simulation case based on the determined similarity by using the simulation models of at least one historical simulation case to simulate the production process.

[0007] The third embodiment of the present disclosure proposes a computing device, which includes: a processor; and a memory for storing computer-executable instructions that cause the processor to execute the method in the first embodiment when the computer-executable instructions are executed.

[0008] The fourth embodiment of the present disclosure proposes a computer-readable storage medium having computer-executable instructions stored thereon, and the computer-executable instructions are used to execute the method in the first embodiment.

[0009] The fifth embodiment of the present disclosure proposes a computer program product, which is tangibly stored on a computer-readable storage medium and includes computer-executable instructions that cause at least one processor to execute the method in the first embodiment when the computer-executable instructions are executed. Description of the Drawings

[0010] In combination with the accompanying drawings and with reference to the following detailed description, the features, advantages and other aspects of the embodiments of the present disclosure will become more obvious. Several embodiments of the present disclosure are shown herein in an exemplary rather than restrictive manner. In the drawings:

[0011] Figure 1 A modeling and simulation method for a production process according to some embodiments of the present disclosure is shown;

[0012] Figure 2 A system architecture diagram for implementing the Figure 1 method in is shown;

[0013] Figure 3 Shown in Figure 2 is a flowchart of a method for modeling and simulating a production process in an embodiment;

[0014] Figure 4 Shown in Figure 3 are multiple sub-steps of step 302 in the method;

[0015] Figure 5 Shows Figure 2 a schematic diagram of a unified data template in an embodiment of

[0016] Figures 6(a) and 6(b) show Figure 2 a schematic diagram of the resource topology of the current simulation case and the historical simulation case in an embodiment of

[0017] Figure 7 Shows Figure 2 a schematic diagram of a deployment method of the system of

[0018] Figure 8 Shows a modeling and simulation device for a production process according to an embodiment of the present disclosure; and

[0019] Figure 9 Shows a block diagram of a computing device for modeling and simulation of a production process according to an embodiment of the present disclosure. Detailed Description of the Specific Embodiments

[0020] The following describes in detail various exemplary embodiments of the present disclosure with reference to the accompanying drawings. Although the following described exemplary methods and apparatuses include software and / or firmware executed on hardware among other components, it should be noted that these examples are merely illustrative and should not be considered restrictive. For example, it is contemplated that any or all of the hardware, software, and firmware components may be implemented exclusively in hardware, exclusively in software, or in any combination of hardware and software. Thus, although the following has described exemplary methods and apparatuses, those skilled in the art should readily understand that the provided examples are not used to limit the ways of implementing these methods and apparatuses.

[0021] In addition, the flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of the methods and systems according to various embodiments of the present disclosure. It should be noted that the functions marked in the blocks may also occur in an order different from that marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, or they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the flowchart and / or block diagram, and the combinations of blocks in the flowchart and / or block diagram, can be implemented using a dedicated hardware-based system that performs the specified functions or operations, or can be implemented using a combination of dedicated hardware and computer instructions.

[0022] As used herein, the terms "comprising", "including" and similar terms are open-ended terms, i.e., "including / including but not limited to", indicating that other content may also be included. The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment", and so on.

[0023] Figure 1 A method for modeling and simulating a production process according to some embodiments of the present disclosure is shown. Figure 1 The method can be executed by a server device communicating with a client device or by a device directly interacting with a user. Referring to Figure 1 , method 100 starts from step 101. In step 101, production input data for the current simulation case is obtained. The production input data, as simulation input, includes all data related to the production process in the simulation case, including but not limited to order order data (such as order order ID, order date, arrival date, raw material ID and quantity, etc.), sales order data (such as sales order ID, arrival date, due date, product ID and quantity, etc.), production order data (such as order ID, sales order ID, start time, end time, status, product ID and quantity, etc.), product data (such as product ID and product name, etc.), raw material data (such as raw material ID and raw material name, etc.), BOM data (bill of materials for product manufacturing) (such as record ID, product ID, raw material ID and quantity, etc.), BOP data (bill of processes for product) (such as record ID, product ID, production name, operation sequence, operation ID, operation name, processing time, incoming part ID, incoming part quantity, outgoing part ID and outgoing part quantity, etc.), resource data (such as resource ID, resource type, resource name, operation ID and operation name, etc.) and resource connection data (such as connection ID, previous resource ID, subsequent resource ID, connection type, flow type, etc.). The production input data can be from production files in the manufacturing-related systems of the factory.

[0024] In step 102, based on the production input data of the current simulation case and the production input data of at least one historical simulation case, the similarity between the current simulation case and each of the at least one historical simulation cases is determined respectively, wherein the production input data of the at least one historical simulation case is stored in a database. The database is used to store the production input data and production output data of all historical simulation cases. The similarity (or similarity score) is a value between 0 and 1, or can be expressed as a percentage. When the similarity between two simulation cases is high, their simulation models also have a high similarity. The similarity between two simulation cases is determined by the simulation input. The simulation input includes both the production input data and the resource topology. In some embodiments, all historical simulation cases can be traversed to determine the similarity between the current simulation case and each historical simulation case. In other embodiments, a part of the historical simulation cases can be screened for similarity comparison to improve the calculation speed. For example, different permissions can be set for users (such as engineers), so that a part of the historical simulation cases can be screened from all historical simulation cases for similarity comparison according to different permissions.

[0025] In some embodiments, a unified production data template is used to store the production input data and production output data of all simulation cases. The production output data is the simulation result of the simulation model. The unified data template can be a pre-established manufacturing ontology or a structured relational database. The unified data template should conform to certain data standards, such as the CMSD (Core Manufacturing Simulation Data) standard: SISO-STD-008-01-2012, so as to increase the generality of the data. By using the unified data template to store the production input data and production output data of all simulation cases, not only can the simulation data be effectively managed, but also the modeling and simulation method can be applied to various types of discrete manufacturing processes. In addition, the unified data template is also helpful for similarity comparison between two simulation cases (described below).

[0026] In some embodiments, step 102 further includes: determining the data similarity between the production input data of the current simulation case and the production input data of the historical simulation case; determining the structural similarity between the resource topology of the current simulation case and the resource topology of the historical simulation case, wherein the resource topology of the current simulation case is generated based on the production input data of the current simulation case, and the resource topology of the historical simulation case is generated based on the production input data of the historical simulation case; and obtaining the similarity between the current simulation case and the historical simulation case according to the data similarity and the structural similarity. As described above, the similarity between two simulation cases is determined by the simulation input. Therefore, it is necessary to separately determine the data similarity between the production input data of the two simulation cases and the structural similarity between the resource topologies, and then determine the similarity between the two simulation cases according to the data similarity and the structural similarity. The resource topology describes the resources (such as equipment, labor, etc.) used in the production process and the relationships between the resources, and can be regarded as a directed graph according to the material flow in the product production process. Each resource has multiple attributes, such as resource type (such as assembly, single processing, parallel processing, etc.), operation name (such as drilling, grinding, etc.), and so on. The connections between resources also have multiple attributes, such as the connection type of the material flow (such as manpower, conveyor belt, forklift, AGV, robotic arm, etc.), flow type (such as batch flow, single-piece flow, order flow, etc.), and so on. Since the production input data includes resource data and resource connection data, the resource topologies of the current simulation case and the historical simulation case can be generated respectively based on the production input data of the current simulation case and the historical simulation case.

[0027] In some embodiments, the production input data includes multiple types of data, and determining the data similarity between the production input data of the current simulation case and the production input data of the historical simulation case further includes: for each type of data, separately determining the data similarity between the data of this type in the current simulation case and the data of this type in the historical simulation case. As described above, the production input data includes various types of data such as order order data, sales order data, production order data, product data, raw material data, BOM data, BOP data, resource data, and resource connection data. Therefore, it is necessary to determine the similarity for each type of data separately.

[0028] In some embodiments, further determining the data similarity between the data of this type in the current simulation case and the data of this type in the historical simulation case further includes: based on a predetermined rule, respectively combining the data of this type in the current simulation case and the data of this type in the historical simulation case into a group of strings; and determining the similarity between the two groups of strings and using it as the data similarity between the data of this type in the current simulation case and the data of this type in the historical simulation case. In some embodiments, the unified production data template is a structured relational database. A number of attributes are preset for each type of data in the production input data, and each type of data is stored in a table. For example, attributes such as record ID, product ID, raw material ID, and quantity are set for BOM data, and attributes such as record ID, product ID, production name, operation sequence, operation ID, operation name, processing time, incoming part ID, incoming part quantity, outgoing part ID, and outgoing part quantity are set for BOP data. All production input data is converted and saved according to this unified production data template. When comparing the current simulation case with a certain historical simulation case, search for the production input data of the historical simulation case from the database storing the production data (including production input data and production output data) of all historical simulation cases according to the case name or case ID of the historical simulation case. Then, traverse each data type and compare the data of this type in the current simulation case with the data of this type in the historical simulation case.

[0029] Specifically, all records of all simulation cases are available in the table of each type of data. Each record has multiple attribute values, and a simulation case may have multiple related records. The attribute values of each record except the ID are combined in a predetermined order to generate a string. When the production input data of a certain type of a simulation case includes multiple records, a group of strings is generated. The data of the same type in the production input data of the current simulation case and the historical simulation case being compared are respectively combined into a group of strings, and the similarity between the two groups of strings is determined and used as the data similarity between the data of this type in the current simulation case and the data of this type in the historical simulation case.

[0030] For a certain type of data in the production input data, it is necessary to cross-compare two sets of strings, that is, each string in one set of strings is compared with each string in the other set of strings respectively, and then the overall similarity between the two sets of strings is calculated. The above process is repeated for each type of data respectively to obtain multiple data similarities, such as BOM data similarity, BOP data similarity, and so on. In some embodiments, the Levenshtein distance (also known as the edit distance) algorithm is used to calculate the similarity between two strings. In other embodiments, other methods can also be used to calculate the data similarity between the production input data of the current simulation case and the historical simulation case.

[0031] As described above, in addition to determining the data similarity between the production input data of the current simulation case and the historical simulation case, it is also necessary to determine the structural similarity between the resource topologies of the current simulation case and the historical simulation case. Two resource topologies can be regarded as two directed graphs. Each resource in the resource topology can be regarded as a node in the directed graph, and the connection between two resources can be regarded as an edge connecting these two nodes. In some embodiments, the Hyperlink-Induced Topic Search (HITS) algorithm can be used to calculate the structural similarity between the resource topologies of the current simulation case and the historical simulation case, that is, the similarity between two directed graphs. In other embodiments, other directed graph comparison algorithms (such as the Loopy Belief Propagation algorithm) can also be used to determine the structural similarity.

[0032] It should be noted that there is no specified execution order for the steps of determining the data similarity between the production input data of the current simulation case and the historical simulation case and the structural similarity between the resource topologies. They can be executed sequentially or in parallel.

[0033] After determining the data similarity between the production input data of the current simulation case and a certain historical simulation case and the structural similarity between the resource topologies, the similarity between the current simulation case and the historical simulation case, that is, the overall case similarity, can be calculated based on the data similarity and the structural similarity. In some embodiments, the average of the data similarity and the structural similarity is calculated and used as the similarity between the current simulation case and the historical simulation case. In other embodiments, other calculation methods can also be used to calculate the similarity between the current simulation case and the historical simulation case using the data similarity and the structural similarity. For example, weights are set for the data similarity and the structural similarity respectively, and their weighted average is used as the similarity between the current simulation case and the historical simulation case. Subsequently, the above process is repeated for other historical simulation cases to obtain the similarities between the current simulation case and other historical simulation cases respectively.

[0034] The process of calculating the similarity between the current simulation case and the historical simulation cases in a hierarchical manner is described above. First, the data similarity between each record of each type of production input data is calculated, then the data similarity between each type of production input data is calculated, and then the structural similarity between the resource topologies is calculated. Finally, the similarity between the current simulation case and the historical simulation cases is obtained. By calculating the similarity in a hierarchical manner, the accuracy of comparing the current simulation case with the historical simulation cases can be improved, thereby increasing the reliability of the simulation model recommendation.

[0035] Continuing to refer to Figure 1 , in step 103, based on the determined similarity, a simulation model of the current simulation case is generated using the simulation models of at least one historical simulation case to simulate the production process. If the similarity between a historical simulation case and the current simulation case is high, it means that their simulation models will also have a high similarity. Therefore, the simulation model of the historical simulation case with a high similarity to the current simulation case can be recommended to the user, thereby accelerating the modeling speed of the current simulation case.

[0036] In some embodiments, step 103 further includes: selecting a reference historical simulation case from at least one historical simulation case based on the determined similarity; providing the simulation model of the reference historical simulation case for adjustment; and using the adjusted simulation model as the simulation model of the current simulation case. The historical simulation cases can be sorted according to the similarity level and provided to the user via a user interface (such as a display interface). The user can manually select, according to his / her experience, a historical simulation case with a high similarity to the current simulation case as the reference historical simulation case. In response to the user's selection, the simulation model of the reference historical simulation case is provided via the user interface. It is also possible to automatically select the historical simulation case with the highest similarity to the current simulation case as the reference historical simulation case and provide the simulation model of the reference historical simulation case via the user interface. The user can adjust the provided simulation model of the reference historical simulation case according to the production input data of the current simulation case, such as modifying, deleting, or adding resources, changing the connections between resources, modifying the attribute values of resources and connections, etc. After the user determines the simulation model, the adjusted simulation model is used as the simulation model of the current simulation case.

[0037] In some embodiments, method 100 further includes ( Figure 1(not shown): simulate the production process via the simulation model of the current simulation case and generate production output data; re-determine the similarity between the current simulation case and each historical simulation case in at least one historical simulation case based on the production input data and production output data of the current simulation case and the production input data and production output data of at least one historical simulation case; and select historical simulation cases for simulation benchmarking analysis from at least one historical simulation case based on the re-determined similarity. After the simulation model of the current simulation case is generated, the simulation is run by an event-driven simulation engine. After the simulation is completed, production output data is obtained, for example, resource KPI data (including resource ID, working part, waiting part, stop part and blocking part, etc.) and order KPI data (including order ID, start time, end time and duration, etc.). Since the production output data is generated after the simulation, the similarity between the current simulation case and the historical simulation case may also be different from the previously determined similarity. Therefore, it is necessary to re-determine the similarity between the current simulation case and each historical simulation case. The process of re-determining the similarity between the current simulation case and the historical simulation case is similar to the previous one. First, the data similarity between the production input data of the current simulation case and the historical simulation case, the data similarity between the production output data, and the structural similarity between the resource topological structures are determined respectively, and then the overall case similarity between the current simulation case and the historical simulation case is calculated based on these similarities. The calculation method of the data similarity between the production output data is the same as the calculation method of the data similarity between the production input data. After determining the similarity, the historical simulation cases are also sorted according to the similarity. Several historical simulation cases (such as several historical simulation cases with high rankings) can be automatically selected or manually selected by the user for simulation benchmarking analysis, such as analyzing the differences between the production input data and the production output data of different simulation cases. By re-determining the similarity between the current simulation case and the historical simulation case after the simulation, similar historical simulation cases can be accurately selected, so simulation benchmarking analysis can be performed more easily and accurately.

[0038] In addition, after the simulation is completed, the production input data and production output data of the current simulation case are saved in the historical database. Therefore, with the continuous expansion and accumulation of historical simulation cases in the historical database, more intelligent and accurate simulation model recommendations can be achieved, reducing the reliance on personal experience.

[0039] In the above embodiments, by comparing the similarity between the current simulation case and the historical simulation cases, it is possible to use the simulation models of the similar historical simulation cases for the modeling of the production process, thereby greatly improving the modeling efficiency, saving the time and effort of engineers to a large extent, and ensuring the consistency and accuracy of the simulation models. In addition, by recommending the simulation models of the similar historical simulation cases to engineers, the requirements and dependence on personal experience and knowledge in the modeling process are reduced.

[0040] The following describes with reference to a specific embodiment Figure 1 the modeling and simulation method for the production process. Figure 2 FIG. shows a system architecture diagram for implementing Figure 1 the method in Figure 3 FIG. shows Figure 2 the flowchart of the method for modeling and simulating the production process in the embodiment of Figure 4 FIG. shows Figure 3 the multiple sub-steps of step 302 in the method of Figure 2 In , the system 200 includes a data loading module 220, a comparison module 221, a sorting module 222, a selection module 223, a modification module 224, a simulation module 225, a historical database 226, and a storage module 227.

[0041] Refer to Figure 2 - Figure 3 simultaneously to describe Figure 2 the embodiment in Figure 3 In the method 300 of , step 301 includes loading the collected production input data by the data loading module 220. The production input data includes all data related to the production process, including but not limited to order order data, sales order data, production order data, product data, BOM data, BOP data, resource data, and resource connection data, etc. The production input data and production output data of all simulation cases are saved in a unified data template. In this embodiment, a relational database is used to construct the data template, and this data template follows the CMSD (Core Manufacturing Simulation Data) standard: SISO-STD-008-01-2012. Figure 5 FIG. shows Figure 2 the schematic diagram of the unified data template in the embodiment of . In this data template, both the production input data and the production output data include various types of data (such as BOM data, BOP data, resource KPI data, etc.), and each type of data includes multiple attributes. Different types of data are associated through some identical attributes (such as ID, etc.) so that the search for associated data can be performed.

[0042] Next, in step 302, the comparison module 221 determines the similarity between the current simulation case and each historical simulation case, and the production input data and production output data of all historical simulation cases are stored in the historical database 226. In this embodiment, step 302 further includes a plurality of sub-steps. Refer to Figure 4 , in sub-step 3020, determine the data similarity between the production input data of the current simulation case and the production input data of the compared historical simulation case. For Figure 5 each type of production input data in, respectively determine the data similarity between the data of this type of the current simulation case and the data of this type of the historical simulation case. A certain type of data of a simulation case may have multiple related records. Combine the attribute values of each record except the ID in a predetermined order to generate a string. When a certain type of data of a simulation case includes multiple records, a group of strings is generated. Combine the data of the same type in the production input data of the current simulation case and the historical simulation case into a group of strings respectively, and determine the similarity between these two groups of strings. The method for calculating the similarity between two groups of strings is specifically described below.

[0043] The following formulas (1) and (2) are used to calculate the similarity sim between two strings s，t .

[0044]

[0045]

[0046] In the above formulas (1) and (2), s represents the source string, and t represents the target string. The source string can be the string corresponding to the current simulation case, and the target string can be the string corresponding to the compared historical simulation case, and vice versa. |s| is the length of the source string, |t| is the length of the target string, i is any value from 0 to |s|, and j is any value from 0 to |t|.

[0047] Compare each source string in the source string group with each target string in the target string group pairwise to obtain multiple similarities, and then calculate the average value of the multiple similarities according to the following formula (3) as the similarity sim between the two groups of strings table .

[0048]

[0049] In formula (3), sim paringstring is the similarity between the source string and the target string being compared. Num records represents the number of records of a certain type of data existing in the table. The obtained simtable That is, it is the data similarity between a certain type of data of the current simulation case and the same type of data of the historical simulation case being compared. The data similarity is determined for each type of data respectively, and sim is obtained table (BOM), sim table (BOP), etc. for multiple data similarities.

[0050] Subsequently, in step 3021, the structural similarity between the resource topologies of the current simulation case and the historical simulation case is determined. In this embodiment, the Hyperlink-Induced Topic Search (HITS) algorithm is used to calculate this structural similarity. FIGS. 6(a) and 6(b) show Figure 2 schematic diagrams of the resource topologies of the current simulation case and the historical simulation case in the embodiment of. In the source topology G in FIG. 6(a) A , M1 - M6 are resources, and L1 - L6 are connections between resources. M1 - M6 are regarded as nodes in a directed graph, and L1 - L6 are regarded as edges in a directed graph. Similarly, in the resource topology G in FIG. 6(b) B , M1 - M7 are resources, and L1 - L7 are connections between resources. M1 - M7 are regarded as nodes in a directed graph, and L1 - L7 are regarded as edges in a directed graph. The following specifically describes the process of calculating the similarity sim A between G B and G topology .

[0051] For two edges in G A and G B , if their source nodes and end nodes are similar, then these two edges are also similar. Let S A (i) represent the source node of edge i in G A , and T A (i) represent the end node of edge i. The cardinality |V A | of the node set in G A is represented by n A , and the cardinality |E A | of its edge set is represented by m A . In the following formula (4), through a pair of n A *m A matrices, namely the source - edge A S matrix and the end - edge matrix A T represent G A :

[0052]

[0053] Similarly, in the following formula (5), through a pair of n B *m BMatrix, source-edge B S Matrix and terminal-edge matrix B T Denote G B :

[0054]

[0055] Use x ij To denote the node similarity between node i in G B And node j in G A And use y pq To denote the edge similarity between edge p in G B And edge q in G A The following formulas (6)-(7) represent the update formulas for edge similarity and node similarity:

[0056] y pq (k) ← x s(p)s(q) (k - 1) + x t(p)t(q) (k - 1) (6)

[0057] x ij (k) ← Σ t(k)＝i，t(l)＝j y kl (k - 1) + Σ s(k)＝i，s(l)＝j y kl (k - 1) (7)

[0058] Denote all y pq (k) and x ij (k) by matrix Y k And X k In formulas (6)-(7), normalization is performed by the F-norm (Frobenius norm) at each stage, denoted by the ← symbol. Using the source-edge A A Matrix and terminal-edge matrix A S Of G T And the source-edge B B Matrix and terminal-edge matrix B S Of G T Transform formulas (6)-(7) into the following matrices (8) and (9):

[0059]

[0060]

[0061] Define y k = vec(Y k ) and x k = vec(X k ). Performing the vec(*) operation on a matrix means stacking the columns of the matrix into a vector. The above formulas (8) and (9) can be expressed as:

[0062]

[0063]

[0064] wherein, represents the Kronecker product of matrices, and aggregates y k and x k into the following update formula for a single matrix:

[0065]

[0066] In formula (12), s k is the graph similarity metric matrix. Under the initial conditions of arbitrarily selecting x0 and y0 = aGx0 (where α is any positive constant), the iterative process will converge to a unique set s final , which represents the non - negative similarity of each pair of nodes and edges in G A and G B . Then, according to the following formula (13), the average value of the highest similarity in the similarity of each pair of nodes and edges is calculated to obtain the similarity sim A between two resource topologies G B and G topology :

[0067]

[0068] where n and m are the sizes of the matrix .

[0069] Next, in sub - step 3022, the average value of the data similarity and the structure similarity is calculated and used as the similarity between the current simulation case and the historical simulation cases. The following shows formula (14) for calculating the average value of the data similarity and the structure similarity.

[0070] sim case = avg(sim topplogy , sim table (BOM),..., sim table (BOP))(14)

[0071] Return Figure 3, in step 303, the sorting module 222 sorts the historical simulation cases according to the similarity level from high to low and provides them to the user via a user interface (such as a display interface). The sorting result may include a ranking serial number, the name or ID of the historical simulation case, and a similarity between 0 and 1. In step 304, the selection module 223 automatically selects or the user manually selects a certain historical simulation case as a reference historical simulation case, and reads the simulation model of the reference historical simulation case from the storage module 227 and provides it to the user via the user interface. For example, the historical simulation case with the highest similarity to the current simulation case is automatically used as the reference historical simulation case.

[0072] In step 305, the user adjusts the simulation model of the provided reference historical simulation case through the modification module 224, and saves the adjusted simulation model in the storage module 227 for future reference. In step 306, the simulation module 225 simulates the production process via the generated simulation model and generates production output data. The simulation module 225 also saves the production input data and production output data of the current simulation case in the historical database 226 as historical simulation cases for future reference.

[0073] In addition, when simulation benchmarking analysis is required, the simulation module 225 can also provide the production input data and production output data to the comparison module 221. The comparison module 221 re-determines the similarity between the current simulation case and each historical simulation case in the historical database 226. Similarly, the similarity between the production input data and production output data of the current simulation case and the historical simulation cases is calculated using formulas (1)-(2), the similarity between the resource topologies is calculated using formulas (4)-(13), and the overall case similarity between the current simulation case and the historical simulation cases is calculated using formula (14).

[0074] After determining the similarity, the sorting module 222 re-sorts the historical simulation cases according to the recalculated similarity level from high to low. Then, the selection module 223 automatically selects or the user manually selects several historical simulation cases (such as several historical simulation cases ranked at the top) for simulation benchmarking analysis. The production input data, simulation model, and production output data of the selected historical simulation cases can all be provided to the user via the user interface. By re-determining the similarity between the current simulation case and the historical simulation cases after simulation, similar historical simulation cases can be accurately selected, so that simulation benchmarking analysis can be carried out more easily and accurately.

[0075] In this embodiment, by comparing the similarity between the current simulation case and historical simulation cases, the simulation model of a similar historical simulation case can be used to model the production process, thereby greatly improving the modeling efficiency, saving the time and effort of engineers to a great extent, and ensuring the consistency and accuracy of the simulation model. In addition, by recommending the simulation models of similar historical simulation cases to engineers, the requirements and dependence on personal experience and knowledge in the modeling process are reduced.

[0076] Figure 7 shows Figure 2 a schematic diagram of a deployment manner of the system of. In Figure 7 the example of, the data loading module 220, comparison module 221, sorting module 222, selection module 223, historical database 226 and storage module 227 in the system 200 are deployed at the cloud device. The modification module 224 and simulation module 225 are deployed at the client device. The client device can communicate with the cloud device. The client device may have a display interface for the user to perform operations such as viewing, adjusting the simulation model, and running the simulation.

[0077] Figure 8 shows a modeling and simulation device for a production process according to an embodiment of the present disclosure. Referring to Figure 8 , the device 800 includes an input data acquisition unit 801, a simulation case comparison unit 802, and a simulation model generation unit 804. The input data acquisition unit 801 is configured to acquire the production input data of the current simulation case. The simulation case comparison unit 802 is configured to determine the similarity between the current simulation case and each of at least one historical simulation case based on the production input data of the current simulation case and the production input data of at least one historical simulation case, wherein the production input data of at least one historical simulation case is stored in the database. The simulation model generation unit 803 is configured to generate the simulation model of the current simulation case by using the simulation models of at least one historical simulation case based on the determined similarity to simulate the production process. Figure 7 Each unit in can be implemented by using software, hardware (such as integrated circuits, FPGAs, etc.) or a combination of software and hardware.

[0078] In some embodiments, the simulation case comparison unit 802 further includes ( Figure 8(not shown in the figure): a data similarity determination unit, a structure similarity determination unit, and a case similarity determination unit. The data similarity determination unit is configured to determine the data similarity between the production input data of the current simulation case and the production input data of the historical simulation case. The structure similarity determination unit is configured to determine the structure similarity between the resource topology of the current simulation case and the resource topology of the historical simulation case, wherein the resource topology of the current simulation case is generated based on the production input data of the current simulation case, and the resource topology of the historical simulation case is generated based on the production input data of the historical simulation case. The case similarity determination unit is configured to obtain the similarity between the current simulation case and the historical simulation case according to the data similarity and the structure similarity.

[0079] In some embodiments, the production input data of the current simulation case and the production input data of the historical simulation case respectively include multiple types of data, and the data similarity determination unit is further configured to: for each type of data, respectively determine the data similarity between the data of this type of the current simulation case and the data of this type of the historical simulation case.

[0080] In some embodiments, the data similarity determination unit is further configured to: based on a predetermined rule, respectively combine the data of this type of the current simulation case and the data of this type of the historical simulation case into a group of strings; and determine the similarity between the two groups of strings, and use it as the data similarity between the data of this type of the current simulation case and the data of this type of the historical simulation case.

[0081] In some embodiments, the case similarity determination unit is further configured to: calculate the average value of the data similarity and the structure similarity, and use it as the similarity between the current simulation case and the historical simulation case.

[0082] In some embodiments, the simulation model generation unit 803 is further configured to: based on the determined similarity, select a reference historical simulation case from at least one historical simulation case; provide the simulation model of the reference historical simulation case for adjustment; and use the adjusted simulation model as the simulation model of the current simulation case.

[0083] In some embodiments, the production output data of at least one historical simulation case is also stored in the database, and the apparatus 800 further includes ( Figure 8An output data generation unit, a simulation case re-comparison unit, and a historical case selection unit (not shown in the figure). The output data generation unit is configured to simulate the production process via the simulation model of the current simulation case and generate production output data. The simulation case re-comparison unit is configured to re-determine the similarity between the current simulation case and each historical simulation case in at least one historical simulation case respectively based on the production input data and production output data of the current simulation case and the production input data and production output data of at least one historical simulation case. The historical case selection unit is configured to select a historical simulation case for simulation benchmark analysis from at least one historical simulation case based on the re-determined similarity.

[0084] In some embodiments, the production input data of the current simulation case and the production input data of at least one historical simulation case use a unified data template.

[0085] Figure 9 The block diagram of a computing device for modeling and simulation of a production process according to an embodiment of the present disclosure is shown. It can be seen from Figure 9 that the computing device 900 for modeling and simulation of a production process includes a processor 901 and a memory 902 coupled to the processor 901. The memory 902 is used to store computer-executable instructions, which cause the processor 901 to execute the methods in the above embodiments when the computer-executable instructions are executed.

[0086] In addition, alternatively, the above method can be implemented by a computer-readable storage medium. A computer-readable program instruction for executing various embodiments of the present disclosure is uploaded on the computer-readable storage medium. The computer-readable storage medium can be a tangible device that can hold and store instructions used by an instruction execution device. The computer-readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the above. More specific examples (non-exhaustive list) of the computer-readable storage medium include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disc (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punched card or raised structures in a groove storing instructions thereon, and any suitable combination of the above. The computer-readable storage medium used herein is not construed as an instantaneous signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagated through a waveguide or other transmission medium (e.g., an optical pulse through an optical fiber cable), or an electrical signal transmitted through a wire.

[0087] Thus, in another embodiment, the present disclosure provides a computer-readable storage medium having computer-executable instructions stored thereon for performing the methods in various embodiments of the present disclosure.

[0088] In another embodiment, the present disclosure provides a computer program product tangibly stored on a computer-readable storage medium and including computer-executable instructions that, when executed, cause at least one processor to perform the methods in various embodiments of the present disclosure.

[0089] In general, the various example embodiments of the present disclosure may be implemented in hardware or a special-purpose circuit, software, firmware, logic, or any combination thereof. Some aspects may be implemented in hardware, while other aspects may be implemented in firmware or software that may be executed by a controller, a microprocessor, or other computing device. When aspects of the embodiments of the present disclosure are illustrated or described as block diagrams, flowcharts, or using some other graphical representation, it will be understood that the blocks, devices, systems, techniques, or methods described herein may be implemented as non-limiting examples in hardware, software, firmware, a special-purpose circuit or logic, general hardware or a controller or other computing device, or some combination thereof.

[0090] The computer-readable program instructions or computer program product for performing the various embodiments of the present disclosure can also be stored in the cloud. When needed, a user can access the computer-readable program instructions stored in the cloud for performing an embodiment of the present disclosure through a mobile Internet, a fixed network, or other networks, so as to implement the technical solutions disclosed in the various embodiments of the present disclosure.

[0091] Although the embodiments of the present disclosure have been described with reference to several specific embodiments, it should be understood that the embodiments of the present disclosure are not limited to the specific embodiments disclosed. The embodiments of the present disclosure are intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims. The scope of the claims is to be accorded the broadest interpretation so as to encompass all such modifications and equivalent structures and functions.

Claims

1. A method for modeling and simulating a production process, comprising: Obtaining production input data of a current simulation case; Based on the production input data of the current simulation case and the production input data of at least one historical simulation case, respectively determining the similarity between the current simulation case and each historical simulation case in the at least one historical simulation case, wherein the production input data of the at least one historical simulation case is stored in a database; and Based on the determined similarity, generating a simulation model of the current simulation case by using the simulation models of the at least one historical simulation case to simulate the production process; Wherein, based on the production input data of the current simulation case and the production input data of at least one historical simulation case, respectively determining the similarity between the current simulation case and each historical simulation case in the at least one historical simulation case further includes: Determining the data similarity between the production input data of the current simulation case and the production input data of the historical simulation case; Determining the structural similarity between the resource topology structure of the current simulation case and the resource topology structure of the historical simulation case, wherein the resource topology structure of the current simulation case is generated based on the production input data of the current simulation case, and the resource topology structure of the historical simulation case is generated based on the production input data of the historical simulation case; and Based on the data similarity and the structural similarity, obtaining the similarity between the current simulation case and the historical simulation case.

2. The method according to claim 1, wherein, The production input data of the current simulation case and the production input data of the historical simulation case respectively include various types of data, and determining the data similarity between the production input data of the current simulation case and the production input data of the historical simulation case further includes: For each type of data, respectively determining the data similarity between the data of the type in the current simulation case and the data of the type in the historical simulation case.

3. The method according to claim 2, wherein, Respectively determining the data similarity between the data of the type in the current simulation case and the data of the type in the historical simulation case further includes: Based on a predetermined rule, respectively combining the data of the type in the current simulation case and the data of the type in the historical simulation case into a group of strings; and Determining the similarity between the two groups of strings and using it as the data similarity between the data of the type in the current simulation case and the data of the type in the historical simulation case.

4. The method according to claim 1, wherein Based on the data similarity and the structural similarity, obtaining the similarity between the current simulation case and the historical simulation case further includes: Calculating the average value of the data similarity and the structural similarity and using it as the similarity between the current simulation case and the historical simulation case.

5. The method according to claim 1, wherein Based on the determined similarity, generating a simulation model of the current simulation case by using the simulation models of the at least one historical simulation case further includes: Select a reference historical simulation case from the at least one historical simulation case based on the determined similarity; Provide the simulation model of the reference historical simulation case for adjustment; and Use the adjusted simulation model as the simulation model of the current simulation case.

6. The method according to claim 1, wherein The production output data of the at least one historical simulation case is also stored in the database, and the method further includes: Simulate the production process via the simulation model of the current simulation case and generate production output data; Based on the production input data and production output data of the current simulation case and the production input data and production output data of the at least one historical simulation case, respectively re-determine the similarity between the current simulation case and each historical simulation case in the at least one historical simulation case; and Based on the re-determined similarity, select a historical simulation case for simulation benchmark analysis from the at least one historical simulation case.

7. The method according to claim 1, wherein, The production input data of the current simulation case and the production input data of the at least one historical simulation case use a unified data template.

8. A modeling and simulation device for a production process, comprising: An input data acquisition unit configured to acquire production input data of a current simulation case; A simulation case comparison unit configured to respectively determine the similarity between the current simulation case and each historical simulation case in the at least one historical simulation case based on the production input data of the current simulation case and the production input data of the at least one historical simulation case, wherein the production input data of the at least one historical simulation case is stored in a database; and A simulation model generation unit configured to generate a simulation model of the current simulation case based on the determined similarity, using the simulation model of the at least one historical simulation case to simulate the production process; Wherein, the simulation case comparison unit further includes: A data similarity determination unit configured to determine the data similarity between the production input data of the current simulation case and the production input data of the historical simulation case; A structure similarity determination unit configured to determine the structure similarity between the resource topology structure of the current simulation case and the resource topology structure of the historical simulation case, wherein the resource topology structure of the current simulation case is generated based on the production input data of the current simulation case, and the resource topology structure of the historical simulation case is generated based on the production input data of the historical simulation case; and A case similarity determination unit configured to obtain the similarity between the current simulation case and the historical simulation case according to the data similarity and the structure similarity.

9. The device according to claim 8, wherein The production input data of the current simulation case and the production input data of the historical simulation case respectively include multiple types of data, and the data similarity determination unit is further configured to: For each type of data, respectively determine the data similarity between the data of the current simulation case of the type and the data of the historical simulation case of the type.

10. The device according to claim 9, wherein, The data similarity determination unit is further configured to: Based on a predetermined rule, respectively combine the data of the type of the current simulation case and the data of the type of the historical simulation case into a group of strings; And Determine the similarity between the two groups of strings and use it as the data similarity between the data of the type of the current simulation case and the data of the type of the historical simulation case.

11. The device according to claim 8, wherein, The case similarity determination unit is further configured to: Calculate the average value of the data similarity and the structure similarity and use it as the similarity between the current simulation case and the historical simulation case.

12. The apparatus according to claim 8, wherein, The simulation model generation unit is further configured to: Based on the determined similarity, select a reference historical simulation case from the at least one historical simulation case; Provide the simulation model of the reference historical simulation case for adjustment; And Use the adjusted simulation model as the simulation model of the current simulation case.

13. The apparatus according to claim 8, wherein The production output data of the at least one historical simulation case is also stored in the database, and the device further includes: An output data generation unit, which is configured to simulate the production process through the simulation model of the current simulation case and generate production output data; A simulation case re-comparison unit, which is configured to respectively re-determine the similarity between the current simulation case and each historical simulation case in the at least one historical simulation case based on the production input data and the production output data of the current simulation case and the production input data and the production output data of the at least one historical simulation case; and A historical case selection unit, which is configured to select a historical simulation case for simulation benchmark analysis from the at least one historical simulation case based on the re-determined similarity.

14. The apparatus according to claim 8, wherein, The production input data of the current simulation case and the production input data of the at least one historical simulation case use a unified data template.

15. A computing device, comprising: A processor; And A memory for storing computer-executable instructions that, when executed, cause the processor to execute the method according to any one of claims 1-7.

16. A computer-readable storage medium having computer-executable instructions stored thereon for executing the method according to any one of claims 1-7.

17. A computer program product tangibly stored on a computer-readable storage medium and including computer-executable instructions that, when executed, cause at least one processor to execute the method according to any one of claims 1-7.

Images