A human-machine collaborative digital twin model system based on deep learning function

Through the human-machine collaboration digital twin model system based on deep learning, the historical data is analyzed and the processing sequence is optimized, and the problem of low collaboration efficiency caused by one-way cooperation between human-machine is solved, and more efficient human-machine collaboration is achieved.

CN118747544BActive Publication Date: 2024-12-06JIANGSU ELECTRIC POWER INFORMATION TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411230334.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-04
Publication Date
2024-12-06
Estimated Expiration
2044-09-04

AI Technical Summary

Technical Problem

At this stage, when human-machine cooperation is collaborative, one-way cooperation between human-machine leads to low collaboration efficiency and the inability to reasonably allocate the processing process of processing workpieces, resulting in the problem of overdue completion of tasks.

Method used

Adopt a human-computer collaborative digital twin model system based on deep learning functions, including task generation module, task analysis module, database, encoding generation module, decoding analysis module and execution terminal. By analyzing historical data, abnormal signals or collaboration signals are generated and appropriately allocated. The collaborative machine and process optimizes the processing sequence and improves collaboration efficiency.

Benefits of technology

Through deep learning technology, analyze the staff and machines in the human-machine collaboration task, optimize the selection of processing machines, improve the efficiency of human-machine collaboration, and avoid overdue completion of tasks.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118747544B_ABST
    Figure CN118747544B_ABST
Patent Text Reader

Abstract

The present invention discloses a human-machine collaborative digital twin model system based on deep learning function, which relates to the field of human-machine collaboration and solves the problem that collaborative machines cannot be reasonably allocated to execute processing procedures during human-machine collaboration. The system includes a task generation module, a task analysis module, a coding generation module and a decoding analysis module. The task generation module is used to generate human-machine collaborative tasks, and the task analysis module is used to analyze human-machine collaborative tasks corresponding to one-way human-machine cooperation. The coding generation module is used to allocate corresponding collaborative machines to collaborative workpieces to obtain the machine code execution sequence and process code execution sequence corresponding to the collaborative workpieces. The decoding analysis module is used to obtain the processing execution sequence corresponding to the collaborative workpieces according to the machine code execution sequence and the process code execution sequence. The present invention is combined with deep learning technology, and analyzes the staff and processing machines in the human-machine collaborative tasks, so as to select suitable processing machines to perform human-machine collaborative tasks.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of human-computer interaction technology, and specifically is a human-computer collaborative digital twin model system based on deep learning function. Background Art

[0002] Deep learning is a subfield of machine learning that simulates the learning process of the human brain by building complex neural networks. In the fields of natural language processing and computer vision, deep learning technology enables machines to understand and generate natural language, recognize and understand images and videos, and provides strong technical support for human-machine collaboration; human-machine collaboration refers to the cooperative relationship between humans and computers or robots, aiming to combine the advantages of humans and computers or robots to achieve more efficient and accurate task completion. In the field of workpiece processing, human-machine collaboration specifically refers to the process in which workers and robots work together to complete workpiece processing;

[0003] However, at present, when human-machine collaboration is used to process workpieces, the human-machine collaboration is one-way, and since there is little intersection between the collaborative personnel and the collaborative machines, it is impossible to reasonably allocate the collaborative machines to execute the processing flow of the workpieces, resulting in poor efficiency of human-machine collaboration, and even overdue completion of human-machine collaboration tasks.

[0004] To this end, we propose a human-machine collaborative digital twin model system based on deep learning capabilities. Summary of the invention

[0005] The purpose of the present invention is to propose a human-machine collaborative digital twin model system based on deep learning function to solve the problems raised in the above background technology.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A human-machine collaborative digital twin model system based on deep learning function, comprising a task generation module, a task analysis module, a database, a coding generation module, a decoding analysis module, an execution terminal and a user terminal, wherein the task generation module is used to generate a human-machine collaborative task. If the collaboration mode of the human-machine collaborative task is bidirectional collaboration between humans and machines, the human-machine collaborative task is sent to the execution terminal; if the collaboration mode of the human-machine collaborative task is unidirectional collaboration between humans and machines, the human-machine collaborative task is sent to the task analysis module;

[0008] The database is used to store historical processing data corresponding to the collaborative personnel and the collaborative machines, and send the historical processing data to the task analysis module; the task analysis module is used to analyze the human-machine collaborative tasks corresponding to the one-way human-machine collaboration, and generate abnormal signals to send to the user terminal or generate collaborative signals to send to the code generation module;

[0009] The coding generation module is used to allocate the corresponding collaborative machine to the collaborative workpiece, obtain the machine coding execution order and the process coding execution order corresponding to the collaborative workpiece, and send them to the decoding analysis module; the decoding analysis module is used to obtain the processing execution order corresponding to the collaborative workpiece according to the machine coding execution order and the process coding execution order; the execution terminal is also used to receive the processing execution order and execute the human-machine collaborative task of the collaborative workpiece according to the processing execution order, and if the human-machine collaborative task is completed, a completion signal is generated and sent to the user terminal;

[0010] The user terminal is used to receive a completion signal or an abnormal signal corresponding to the human-machine collaborative task. If it is an abnormal signal, the collaborative personnel and the collaborative machine are reassigned to the human-machine collaborative task. If it is a completion signal, the user can understand the processing progress of the collaborative workpiece through the completion signal.

[0011] Furthermore, the human-machine collaboration task specifically includes the human-machine collaboration mode, the number of collaborative machines, the number of collaborative personnel, and the number of collaborative workpieces. The human-machine collaboration mode includes human-machine one-way collaboration and human-machine two-way collaboration.

[0012] The historical processing data includes the historical number of processed workpieces, the historical number of workpiece failures, and the historical working hours corresponding to the collaborative personnel and collaborative machines.

[0013] Furthermore, the analysis process of the task analysis module is as follows:

[0014] Obtain the generated human-machine collaboration task, and obtain the number of collaborating personnel and the number of collaborating machines corresponding to the human-machine collaboration task;

[0015] Obtain the historical number of processed workpieces LRSi, the historical number of workpiece failures LRGi, and the historical working time LRTi corresponding to the collaborative personnel, where i is the number corresponding to the collaborative personnel, i=1, 2, ..., z;

[0016] The personnel excellence value RYXi corresponding to the collaborative personnel in the human-machine collaborative task is calculated by the formula RYXi=LRSi×LRTi / LRGi;

[0017] The personnel excellence values ​​of all collaborators are added together to obtain the total personnel excellence value of the human-machine collaborative task;

[0018] If the total personnel excellence value is greater than the personnel standard excellence value, the personnel processing level of the human-machine collaborative task is recorded as first-level personnel processing;

[0019] If the total personnel excellence value is less than or equal to the standard excellence value, the personnel processing level of the human-machine collaborative task will be recorded as second-level personnel processing; among which, the processing capability of the collaborative personnel corresponding to the first-level personnel processing is better than the processing capability of the collaborative personnel corresponding to the second-level personnel processing.

[0020] Furthermore, the analysis process of the task analysis module also includes:

[0021] Obtain the historical number of processed workpieces LJSm, the historical number of workpiece failures LJGm, and the historical working time LJTm corresponding to the collaborative machine, where m is the number corresponding to the collaborative machine, m=1, 2, ..., y, and y is a positive integer;

[0022] The machine excellence value corresponding to the machine in the human-machine collaborative task is calculated by the formula JYXm=e / (LJSm+LJGm×LJTm); where e is a natural constant;

[0023] The machine excellence values ​​of all collaborative machines are added together to obtain the total machine excellence value of the human-machine collaborative task;

[0024] If the total machine excellence value is greater than the machine standard excellence value, the machine processing level of the human-machine collaborative task is recorded as first-level machine processing;

[0025] If the total machine excellence value is less than or equal to the standard excellence value, the machine processing level of the human-machine collaborative task is recorded as secondary machine processing, where the processing capability of the collaborative machine corresponding to the primary machine processing is better than the processing capability of the collaborative machine corresponding to the secondary machine processing;

[0026] If the human processing level of the human-machine collaborative task is level 2 human processing and the machine processing level is level 2 machine processing, an abnormal signal is generated;

[0027] If the human processing level of the human-machine collaborative task is first-level human processing or the machine processing level is first-level machine processing, a collaborative signal is generated.

[0028] Furthermore, the working process of the code generation module is as follows:

[0029] Obtain the collaborative workpiece number XGn corresponding to the collaborative workpiece, and record the process corresponding to the collaborative workpiece as GXnj; read the corresponding processing time according to the process, where n is the number corresponding to the collaborative workpiece and j is the number corresponding to the process;

[0030] Get the number of collaborative machines XJm of the collaborative machines;

[0031] constructing a processing schedule based on the number of collaborative workpieces of the collaborative workpiece and the number of collaborative machines of the collaborative machine;

[0032] At the same time, genetic algorithms are used to perform chromosome coding operations, and the machine coding execution order of collaborative workpieces is obtained through machine selection coding, and then combined with process sorting coding to obtain the process coding execution order of collaborative workpieces.

[0033] Furthermore, the machine selection code is specifically:

[0034] Each gene bit in the machine chromosome is represented by an integer, which is arranged in the order of the collaborative workpieces and the processes of the collaborative workpieces, and recorded as the machine code execution order;

[0035] Each integer represents the sequential number of the processing machine selected for the current operation in the set of optional machines.

[0036] Furthermore, the process sequence code is specifically as follows:

[0037] Each gene position in the process chromosome is directly encoded with the number corresponding to the collaborative workpiece. The order in which the numbers of the collaborative workpieces appear indicates the order in which the corresponding processes of the collaborative workpieces are processed.

[0038] The chromosomes are compiled from left to right and recorded as the order in which the process codes are executed.

[0039] Furthermore, the working process of the decoding analysis module is as follows:

[0040] Obtain the machine code execution sequence and process code execution sequence corresponding to the collaborative workpiece;

[0041] The execution order of the machine code is decoded and converted into the machine sequence matrix JJ and the time sequence matrix SJ from left to right, where JJ(n, j) represents the collaborative machine number corresponding to the process GXnj of the collaborative workpiece XGn, and SJ(n, j) represents the processing time corresponding to the process GHnj of the collaborative workpiece XGn. JJ(n, j) and SJ(n, j) are one-to-one corresponding.

[0042] Furthermore, the working process of the decoding analysis module also includes:

[0043] The process code execution order is decoded and read from left to right in sequence according to the process code execution order. The collaborative machines and processing time corresponding to each process in the process code execution order are read in sequence through the machine sequence matrix and time sequence matrix in the machine code execution order.

[0044] Furthermore, the reading process is as follows:

[0045] Step S1, read the first step GXnj in the execution order of the step code;

[0046] Step S2, obtain the collaborative machine XJm=JJ(n, j) and processing time SJ(n, j) corresponding to the process GXnj through the machine sequence matrix JJ and the time sequence matrix SJ;

[0047] Step S3: If the process GXnj is the first process of the collaborative machine XJm, then the process GXnj is the first process of the collaborative machine XJm. j-1 Processing end time JS j-1Start processing;

[0048] If the process GXnj is the first process of the collaborative workpiece XGn, the processing starts from the initial moment of the collaborative machine XJm;

[0049] Step S4: If step S3 is not satisfied, find the idle time period [KXk, KXj] corresponding to the cooperative machine XJm, where KXk is the start time of the idle time and KXj is the end time of the idle time;

[0050] The earliest processing time JG of the process GXnj is obtained by the following formula. The earliest processing time JG of the process GXnj satisfies the constraint condition: JG = max{JS j-1 , KXk};

[0051] Step S5, judging whether the idle time period of the collaborative machine XJm satisfies the insertion condition through constraint condition 5.1, if so, processing is performed directly according to the idle time period; if not, processing is performed on the collaborative machine XJm according to the time JGt shown in formula 5.2;

[0052] JG+SJ(n,j)≤KXj-KXk, constraint 5.1;

[0053] JGt=max{JS j-1 , LKXj}, Formula 5.2; where LKXj is the end time of the last process of the collaborative machine XJm;

[0054] Step S6, determining whether the processes corresponding to all the collaborative workpieces have been read, if not, executing step S1 for the next process;

[0055] If the reading is completed, the process ends and the corresponding processing execution sequence is generated and sent to the execution terminal.

[0056] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0057] The present invention generates a human-machine collaborative task through a task generation module. If the collaboration mode of the human-machine collaborative task is one-way collaboration between man and machine, the task analysis module is used to analyze the human-machine collaborative task corresponding to the one-way collaboration between man and machine, and the analysis generates an abnormal signal or a collaborative signal. If a collaborative signal is generated, the coding generation module is used to allocate a corresponding collaborative machine to the collaborative workpiece, and the machine coding execution sequence and process coding execution sequence corresponding to the collaborative workpiece are obtained. The decoding analysis module obtains the processing execution sequence corresponding to the collaborative workpiece according to the machine coding execution sequence and the process coding execution sequence. The execution terminal executes the human-machine collaborative task of the collaborative workpiece according to the processing execution sequence. The present invention is combined with deep learning technology, and the staff and processing machines in the human-machine collaborative task are analyzed, so as to select an adapted processing machine to perform the human-machine collaborative task. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0059] Figure 1 It is the overall system block diagram of the present invention.

[0060] Figure 2 A schematic diagram of the machine selection encoding in the present invention.

[0061] Figure 3 A schematic diagram of the process selection coding in the present invention. DETAILED DESCRIPTION

[0062] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0063] See also Figure 1-Figure 3 As shown, the technical solution provided by the present invention is: a human-machine collaborative digital twin model system based on deep learning function, including a task generation module, a task analysis module, a database, a coding generation module, a decoding analysis module, an execution terminal and a user terminal;

[0064] The task generation module is used to generate human-machine collaboration tasks, which specifically include human-machine collaboration mode, the number of collaborative machines, the number of collaborative personnel, and the number of collaborative workpieces; the human-machine collaboration mode includes human-machine unidirectional collaboration and human-machine bidirectional collaboration; human-machine unidirectional collaboration means that humans control machines to perform production work, and human-machine bidirectional collaboration means that both collaborative personnel and collaborative machines can select and optimize collaborative tasks;

[0065] If the collaboration mode of the human-machine collaborative task is two-way collaboration between humans and machines, the human-machine collaborative task will be sent to the execution terminal, and the collaborative personnel will directly adjust the processing flow of the collaborative workpiece in real time; if the collaboration mode of the human-machine collaborative task is one-way collaboration between humans and machines, the human-machine collaborative task will be sent to the task analysis module;

[0066] The database is used to store historical processing data corresponding to the collaborative personnel and the collaborative machines, and send the historical processing data to the task analysis module, wherein the historical processing data includes the number of historical processing workpieces, the number of historical workpiece failures and the historical working hours corresponding to the collaborative personnel and the collaborative machines;

[0067] The task analysis module is used to analyze the human-machine collaborative task corresponding to the one-way human-machine collaboration. The specific analysis process is as follows:

[0068] Obtain the generated human-machine collaboration task, and obtain the number of collaborating personnel and the number of collaborating machines corresponding to the human-machine collaboration task;

[0069] Read the historical number of processed workpieces LRSi, the historical number of workpiece failures LRGi and the historical working time LRTi corresponding to the collaborative personnel through the database;

[0070] The personnel excellence value RYXi corresponding to the collaborative personnel in the human-machine collaborative task is calculated by the formula, and the specific formula is as follows:

[0071] RYXi=LRSi×LRTi / LRGi; where i is the number corresponding to the collaborative personnel, i=1, 2, ..., z, and z is a positive integer. It can be understood that the longer the historical working time of the collaborative personnel and the more historical processed workpieces, the more rich the work experience of the corresponding collaborative personnel is, and the more positive effect it has on the processed workpieces;

[0072] The personnel excellence values ​​of all collaborative personnel are added together to obtain the total personnel excellence value of the human-machine collaborative task; if the total personnel excellence value is greater than the personnel standard excellence value, the personnel processing level of the human-machine collaborative task is recorded as first-level personnel processing; if the total personnel excellence value is less than or equal to the standard excellence value, the personnel processing level of the human-machine collaborative task is recorded as second-level personnel processing; the processing capability of the collaborative personnel corresponding to the first-level personnel processing is better than the processing capability of the collaborative personnel corresponding to the second-level personnel processing;

[0073] Then obtain the historical number of processed workpieces LJSm, the historical number of workpiece failures LJGm, and the historical working time LJTm corresponding to the collaborative machine;

[0074] Similarly, the machine excellence value corresponding to the machine in the human-machine collaborative task is calculated by JYXm=e / (LJSm+LJGm×LJTm); in the formula, e is a natural constant, m is the number corresponding to the collaborative machine, m=1, 2, ..., y, y is a positive integer. It can be understood that the longer the historical working time of the collaborative machine and the more the historical number of processed workpieces, the higher the wear degree of the corresponding collaborative machine is, which has a negative effect on the processed workpieces;

[0075] The machine excellence values ​​of all collaborative machines are added together to obtain the total machine excellence value of the human-machine collaborative task; if the total machine excellence value is greater than the machine standard excellence value, the machine processing level of the human-machine collaborative task is recorded as first-level machine processing; if the total machine excellence value is less than or equal to the standard excellence value, the machine processing level of the human-machine collaborative task is recorded as second-level machine processing; the processing capability of the collaborative machine corresponding to the first-level machine processing is better than the processing capability of the collaborative machine corresponding to the second-level machine processing;

[0076] If the human processing level of the human-machine collaborative task is level 2 human processing and the machine processing level is level 2 machine processing, an abnormal signal is generated and sent to the user terminal, indicating that the corresponding human-machine collaborative task cannot be completed as expected;

[0077] If the human processing level of the human-machine collaborative task is level one human processing or the machine processing level is level one machine processing, a collaborative signal is generated and sent to the code generation module;

[0078] The code generation module is used to allocate a corresponding collaborative machine to the collaborative workpiece after receiving the collaborative signal. The allocation process is as follows:

[0079] Obtain the number of collaborative workpieces XGn corresponding to the collaborative workpiece, and record the process corresponding to the collaborative workpiece as GXnj; read the corresponding processing time according to the process;

[0080] Where n is the number corresponding to the collaborative workpiece, and j is the number corresponding to the process; for example, GX11 represents the first process corresponding to the first collaborative workpiece;

[0081] Get the cooperative machine number XJm of the cooperative machine, where m is the number corresponding to the cooperative machine;

[0082] Generate a processing schedule corresponding to the collaborative workpiece. A processing schedule is shown in the following table:

[0083]

[0084]

[0085] It should be explained that: XG1 represents the first collaborative workpiece; GX11 represents the first process of the first collaborative workpiece; "-" represents that the collaborative machine is not selectable, that is, the collaborative workpiece cannot be processed on the collaborative machine; the Arabic numerals in the table represent the time required for processing the collaborative workpiece on the corresponding collaborative machine, and the time unit is minutes;

[0086] Use genetic algorithms to perform chromosome encoding operations. Genetic algorithms are a series of search algorithms inspired by natural evolution theory. By imitating the process of natural selection and reproduction, genetic algorithms can provide high-quality solutions for various problems involving search, optimization and learning. Chromosome encoding is to express the solution of the problem under study in the form of chromosomes, which is the key to genetic algorithms. A good encoding method can more easily generate feasible solutions in genetic operations and improve execution efficiency. Otherwise, infeasible solutions will be generated after genetic operations, which require certain repair measures, thus reducing execution efficiency. Chromosome encoding has a direct relationship with solution speed, calculation accuracy, etc., and has an important impact on the algorithm.

[0087] See also Figure 2-Figure 3 As shown, the encoding method selected in this embodiment is a combination of machine selection encoding and process sorting encoding;

[0088] Machine selection coding: Each gene bit in the machine chromosome is represented by an integer, which is arranged in the order of the collaborative workpieces and the processes of the collaborative workpieces, and recorded as the machine coding execution order; each integer represents the sequential number of the processing machine selected for the current process in the optional machine set;

[0089] like Figure 2 As shown, there are five cooperative machines to choose from in process GX11. Since the fourth machine is selected, that is, processing is performed on cooperative machine XJ4, the value of the first gene bit corresponding to the machine chromosome is "4"; there are two cooperative machines to choose from in process GX12 (XJ2 and XJ4). Since the first cooperative machine is selected, that is, processing is performed on cooperative machine XJ2, the value of the second gene bit corresponding to the machine chromosome is "1";

[0090] Process sequence coding: Each gene position in the process chromosome is directly coded with the number corresponding to the collaborative workpiece. The order of appearance of the numbers corresponding to the collaborative workpieces indicates the processing order of the corresponding processes of the collaborative workpieces. The chromosome is compiled from left to right and recorded as the process code execution order.

[0091] like Figure 3As shown, assuming that the process chromosome is [2, 2, 1, 1, 2], the first "2" represents the first process GX21 of the collaborative workpiece XG2, the second "2" represents the second process GX22 of the collaborative workpiece XG2, and so on, to obtain the processing order of the processes in each collaborative workpiece;

[0092] The code generation module sends the machine code execution sequence and the process code execution sequence corresponding to the collaborative workpiece to the decoding analysis module;

[0093] The decoding and analysis module is used to obtain the processing execution sequence corresponding to the collaborative workpiece according to the machine code execution sequence and the process code execution sequence corresponding to the collaborative workpiece. The working process of the decoding and analysis module is as follows:

[0094] Obtain the machine code execution sequence and process code execution sequence corresponding to the collaborative workpiece;

[0095] Decode the execution order of the machine code and convert it into the machine sequence matrix JJ and the time sequence matrix SJ from left to right, where JJ(n, j) represents the collaborative machine number corresponding to the process GXnj of the collaborative workpiece XGn, and SJ(n, j) represents the processing time corresponding to the process GHnj of the collaborative workpiece XGn. JJ(n, j) and SJ(n, j) are in one-to-one correspondence;

[0096] Such as processing schedule and Figure 2 The Arabic numerals shown in the figure are converted into a machine order matrix: The time sequence matrix is:

[0097] Decode the process code execution sequence, read it from left to right in sequence, and read the collaborative machine and processing time corresponding to each process in the process code execution sequence through the machine sequence matrix and time sequence matrix in the machine code execution sequence. The detailed process is as follows:

[0098] Step S1, read the first step GXnj in the execution order of the step code;

[0099] Step S2, obtain the collaborative machine XJm=JJ(n, j) and processing time SJ(n, j) corresponding to the process GXnj through the machine sequence matrix JJ and the time sequence matrix SJ;

[0100] Step S3: If process GXnj is the first processing process of collaborative machine XJm, the previous process GXnj is processed from process GXnj. j-1 Processing end time JS j-1 Start processing; if process GXnj is the first process of collaborative workpiece XGn, start processing directly from the initial moment of collaborative machine XJm;

[0101] Step S4: If step S3 is not satisfied, find the idle time period [KXk, KXj] corresponding to the collaborative machine XJm, where KXk is the start time of the idle time and KXj is the end time of the idle time; the earliest starting processing time JG of the process GXnj is obtained by the following formula, and the earliest starting processing time JG of the process GXnj satisfies the constraint condition: JG = max{JS j-1 , KXk};

[0102] Step S5, judging whether the idle time period of the collaborative machine XJm satisfies the insertion condition through constraint condition 5.1, if so, processing is performed directly according to the idle time period; if not, processing is performed on the collaborative machine XJm according to the time JGt shown in formula 5.2;

[0103] JG+SJ(n,j)≤KXj-KXk, constraint 5.1;

[0104] JGt=max{JS j-1 , LKXj}, Formula 5.2; where LKXj is the end time of the last process of the collaborative machine XJm;

[0105] Step S6, judging whether the processes corresponding to all the collaborative workpieces have been read, if not, executing step S1 for the next process; if reading is completed, ending and generating the corresponding processing execution sequence and sending it to the execution terminal;

[0106] The execution terminal is also used to receive the processing execution order and execute the human-machine collaboration task of the collaborative workpiece according to the processing execution order, and if the human-machine collaboration task is completed, a completion signal is generated and sent to the user terminal;

[0107] The user terminal is used to receive a completion signal or an abnormal signal corresponding to the human-machine collaborative task. If it is an abnormal signal, the collaborative personnel and the collaborative machine are reassigned to the human-machine collaborative task; if it is a completion signal, the user understands the processing progress of the collaborative workpiece through the completion signal;

[0108] In this application, if corresponding calculation formulas appear, the above calculation formulas are all dimensionless and take their numerical calculations. The weight coefficients, proportional coefficients and other coefficients in the formulas are set to a result value obtained by quantifying each parameter. The size of the weight coefficient and the proportional coefficient can be determined as long as it does not affect the proportional relationship between the parameter and the result value.

[0109] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A human-machine collaborative digital twin model system based on deep learning function, It is characterized in that It includes a task generation module, a task analysis module, a database, a coding generation module, a decoding analysis module, an execution terminal and a user terminal. The task generation module is used to generate a human-machine collaborative task. If the collaborative mode of the human-machine collaborative task is human-machine two-way collaboration, the human-machine collaborative task is sent to the execution terminal. If the collaborative mode of the human-machine collaborative task is human-machine unidirectional collaboration, the human-machine collaborative task is sent to the task analysis module. The human-machine collaborative task specifically includes the human-machine collaborative mode, the number of collaborative machines, the number of collaborative personnel and the number of collaborative workpieces. The human-machine collaborative mode includes human-machine unidirectional collaboration and human-machine two-way collaboration. The database is used to store historical processing data corresponding to the collaborative personnel and the collaborative machines, and send the historical processing data to the task analysis module, the historical processing data includes the historical processing workpiece number, historical workpiece failure number and historical working time corresponding to the collaborative personnel and the collaborative machines; the task analysis module is used to analyze the human-machine collaborative tasks corresponding to the one-way human-machine cooperation, and the analysis generates an abnormal signal and sends it to the user terminal or generates a collaborative signal and sends it to the code generation module; The analysis process of the task analysis module is as follows: Obtain the generated human-machine collaboration task, and obtain the number of collaborating personnel and the number of collaborating machines corresponding to the human-machine collaboration task; Obtain the historical number of processed workpieces LRSi, the historical number of workpiece failures LRGi, and the historical working time LRTi corresponding to the collaborative personnel, where i is the number corresponding to the collaborative personnel, i=1, 2, ..., z; The personnel excellence value RYXi corresponding to the collaborative personnel in the human-machine collaborative task is calculated by the formula RYXi=LRSi×LRTi / LRGi; The personnel excellence values ​​of all collaborators are added together to obtain the total personnel excellence value of the human-machine collaborative task; If the total personnel excellence value is greater than the personnel standard excellence value, the personnel processing level of the human-machine collaborative task is recorded as first-level personnel processing; If the total value of personnel excellence is less than or equal to the standard excellence value, the personnel processing level of the human-machine collaborative task is recorded as level 2 personnel processing; the processing capability of the collaborative personnel corresponding to level 1 personnel processing is better than the processing capability of the collaborative personnel corresponding to level 2 personnel processing; Then obtain the historical number of processed workpieces LJSm, the historical number of workpiece failures LJGm and the historical working time LJTm corresponding to the collaborative machine, where m is the number corresponding to the collaborative machine, m=1, 2, ..., y, and y is a positive integer; The machine excellence value corresponding to the machine in the human-machine collaborative task is calculated by the formula JYXm=e / (LJSm+LJGm×LJTm); where e is a natural constant; The machine excellence values ​​of all collaborative machines are added together to obtain the total machine excellence value of the human-machine collaborative task; If the total machine excellence value is greater than the machine standard excellence value, the machine processing level of the human-machine collaborative task is recorded as first-level machine processing; If the total machine excellence value is less than or equal to the standard excellence value, the machine processing level of the human-machine collaborative task is recorded as secondary machine processing, where the processing capability of the collaborative machine corresponding to the primary machine processing is better than the processing capability of the collaborative machine corresponding to the secondary machine processing; If the human processing level of the human-machine collaborative task is level 2 human processing and the machine processing level is level 2 machine processing, an abnormal signal is generated; If the human processing level of the human-machine collaborative task is level one human processing or the machine processing level is level one machine processing, a collaborative signal is generated; The coding generation module is used to allocate the corresponding collaborative machine to the collaborative workpiece, obtain the machine coding execution order and the process coding execution order corresponding to the collaborative workpiece, and send them to the decoding analysis module; the decoding analysis module is used to obtain the processing execution order corresponding to the collaborative workpiece according to the machine coding execution order and the process coding execution order; the execution terminal is also used to receive the processing execution order and execute the human-machine collaborative task of the collaborative workpiece according to the processing execution order, and if the human-machine collaborative task is completed, a completion signal is generated and sent to the user terminal; The working process of the code generation module is as follows: Obtain the collaborative workpiece number XGn corresponding to the collaborative workpiece, and record the process corresponding to the collaborative workpiece as GXnj; read the corresponding processing time according to the process, where n is the number corresponding to the collaborative workpiece and j is the number corresponding to the process; Get the number of collaborative machines XJm of the collaborative machines; constructing a processing schedule based on the number of collaborative workpieces of the collaborative workpiece and the number of collaborative machines of the collaborative machine; At the same time, genetic algorithms are used to perform chromosome coding operations, and the machine code execution order of the collaborative workpiece is obtained through machine selection coding, and then combined with process sorting coding to obtain the process code execution order of the collaborative workpiece; The working process of the decoding analysis module is as follows: Obtain the machine code execution sequence and process code execution sequence corresponding to the collaborative workpiece; Decode the execution order of the machine code and convert it into the machine sequence matrix JJ and the time sequence matrix SJ from left to right, where JJ(n, j) represents the collaborative machine number corresponding to the process GXnj of the collaborative workpiece XGn, and SJ(n, j) represents the processing time corresponding to the process GHnj of the collaborative workpiece XGn. JJ(n, j) and SJ(n, j) are in one-to-one correspondence; Decode the process code execution sequence, read it from left to right in sequence according to the process code execution sequence, and read the collaborative machine and processing time corresponding to each process in the process code execution sequence in sequence through the machine sequence matrix and time sequence matrix in the machine code execution sequence; The user terminal is used to receive a completion signal or an abnormal signal corresponding to the human-machine collaborative task. If it is an abnormal signal, the collaborative personnel and the collaborative machine are reassigned to the human-machine collaborative task. If it is a completion signal, the user can understand the processing progress of the collaborative workpiece through the completion signal.

2. According to claim 1, a human-machine collaborative digital twin model system based on deep learning function, It is characterized in that The machine selection code is as follows: Each gene bit in the machine chromosome is represented by an integer, which is arranged in the order of the collaborative workpieces and the processes of the collaborative workpieces, and recorded as the machine code execution order; Each integer represents the sequential number of the processing machine selected for the current operation in the set of optional machines.

3. According to the human-machine collaborative digital twin model system based on deep learning function according to claim 1, It is characterized in that The specific process sequence code is: Each gene position in the process chromosome is directly encoded with the number corresponding to the collaborative workpiece. The order in which the numbers of the collaborative workpieces appear indicates the order in which the corresponding processes of the collaborative workpieces are processed. The chromosomes are compiled from left to right and recorded as the order in which the process codes are executed.

4. According to claim 1, a human-machine collaborative digital twin model system based on deep learning function, It is characterized in that The reading process is as follows: Step S1, read the first step GXnj in the execution order of the step code; Step S2, obtain the collaborative machine XJm=JJ(n, j) and processing time SJ(n, j) corresponding to the process GXnj through the machine sequence matrix JJ and the time sequence matrix SJ; Step S3: If the process GXnj is the first process of the collaborative machine XJm, then the process GXnj is the first process of the collaborative machine XJm. j-1 Processing end time JS j-1 Start processing; If the process GXnj is the first process of the collaborative workpiece XGn, the processing starts from the initial moment of the collaborative machine XJm; Step S4: If step S3 is not satisfied, find the idle time period [KXk, KXj] corresponding to the cooperative machine XJm, where KXk is the start time of the idle time and KXj is the end time of the idle time; The earliest processing time JG of the process GXnj is obtained by the following formula. The earliest processing time JG of the process GXnj satisfies the constraint condition: JG = max{JS j-1 , KXk}; Step S5, judging whether the idle time period of the collaborative machine XJm satisfies the insertion condition through constraint condition 5.1, if so, processing is performed directly according to the idle time period; if not, processing is performed on the collaborative machine XJm according to the time JGt shown in formula 5.2; JG+SJ(n,j)≤KXj-KXk, constraint 5.1; JGt=max{JS j-1 , LKXj}, Formula 5.2; where LKXj is the end time of the last process of the collaborative machine XJm; Step S6, determining whether the processes corresponding to all the collaborative workpieces have been read, if not, executing step S1 for the next process; If the reading is completed, the process ends and the corresponding processing execution sequence is generated and sent to the execution terminal.