Intelligent production method, device, equipment and storage medium based on directed acyclic graph

Abstract

The present application provides an intelligent production method, device, equipment and storage medium based on a directed acyclic graph. By constructing a device model, the limitations of the performance of production equipment itself are determined. And by constructing an intelligent factory model to simulate the actual production process, and solving the intelligent factory model with the goal of minimizing cost, an optimal transmit power allocation ratio and an optimal offloading decision are obtained. Through the optimal transmit power allocation ratio and the optimal offloading decision, the energy cost and time cost of the intelligent factory can be minimized. Moreover, the intelligent factory model itself limits parameters such as the transmit power allocation ratio, the energy consumption of production equipment itself, and the production time, ensuring that the optimal production plan will not exceed the time limit of the production task and will not cause production stagnation due to insufficient energy. While minimizing the production cost, it is ensured that each production equipment has sufficient energy support and meets the time limit of production.

Description

Technical Field

[0001] The present application relates to the field of communication technologies, and in particular, to an intelligent production method, device, equipment, and storage medium based on a directed acyclic graph. Background Art

[0002] With the rise of intelligent manufacturing, the industrial Internet has developed rapidly, and the vision of digitization, intelligence, and networking has gradually become a reality. In the actual intelligent factory scenario, the research on the dependency relationship between tasks of production equipment mainly focuses on the data sharing ratio. However, in practice, the prerequisite for data sharing is the support of transmission power. In addition, in the research on production costs, related technical solutions often do not consider the energy of the production equipment itself and the time limit of each production task, which may lead to insufficient energy and production stagnation. Summary of the Invention

[0003] In view of this, the purpose of the present application is to propose an intelligent production method, device, equipment, and storage medium based on a directed acyclic graph, which can complete the intelligent production task at the lowest cost within a specified time while ensuring sufficient energy.

[0004] Based on the above purpose, the present application provides an intelligent production method based on a directed acyclic graph, including:

[0005] Determine the external dependency relationship between different production equipment and the internal dependency relationship of the same production equipment;

[0006] Construct a combined directed acyclic graph according to the external dependency relationship and the internal dependency relationship;

[0007] Construct an equipment model according to the equipment performance parameters of each production equipment and the server performance parameters of each edge server;

[0008] Construct an intelligent factory model according to the combined directed acyclic graph, the equipment model, and the task information of each production equipment;

[0009] Taking the minimum cost as the goal, solve the intelligent factory model through a linear goal programming algorithm to obtain the optimal transmission power allocation ratio and the optimal offloading decision;

[0010] Determine the optimal production plan according to the optimal transmission power allocation ratio and the optimal offloading decision, and perform intelligent production according to the optimal production plan.

[0011] Based on the same inventive concept, the present disclosure also provides an intelligent production device based on a directed acyclic graph, including:

[0012] A dependency determination module, configured to: determine the external dependencies between different production devices and the internal dependencies of the same production device;

[0013] A combined directed acyclic graph module, configured to: construct a combined directed acyclic graph according to the external dependencies and the internal dependencies;

[0014] A device model construction module, configured to: construct a device model according to the device performance parameters of each production device and the server performance parameters of each edge server;

[0015] A factory model construction module, configured to: construct an intelligent factory model according to the combined directed acyclic graph and the task information of each production device;

[0016] A model optimal solution module, configured to: aiming at minimizing the cost, solve the intelligent factory model through a linear objective programming algorithm to obtain the optimal transmission power allocation ratio and the optimal offloading decision;

[0017] A production plan generation module, configured to: determine an optimal production plan according to the optimal transmission power allocation ratio and the optimal offloading decision, and perform intelligent production according to the optimal production plan.

[0018] Based on the same inventive concept, the present disclosure also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable by the processor. When the processor executes the computer program, the method described above is implemented.

[0019] Based on the same inventive concept, the present disclosure also provides a non-transitory computer-readable storage medium. The non-transitory computer-readable storage medium stores computer instructions for causing a computer to execute the method described above.

[0020] As can be seen from the above, the intelligent production method, device, equipment, and storage medium based on a directed acyclic graph provided by this application determine the external dependencies between different production devices and the internal dependencies of the same production device; construct a combined directed acyclic graph according to the external and internal dependencies; construct a device model based on the device performance parameters of each production device and the server performance parameters of each edge server; construct an intelligent factory model according to the combined directed acyclic graph, the device model, and the task information of each production device; with the goal of minimizing cost, solve the intelligent factory model through a linear goal programming algorithm to obtain the optimal transmission power allocation ratio and the optimal offloading decision; determine the optimal production plan according to the optimal transmission power allocation ratio and the optimal offloading decision, and perform intelligent production according to the optimal production plan. By constructing a combined directed acyclic graph based on the internal and external dependencies, the dependencies between production tasks are uniformly managed, providing a dependency basis for the reasonable allocation of transmission power resources between sending data to edge servers and sending data to other production devices. And by constructing a device model to determine the limitations of the performance of the production device itself. And by constructing an intelligent factory model to simulate the actual production process, and solving the intelligent factory model with the goal of minimizing cost to obtain the optimal transmission power allocation ratio and the optimal offloading decision. Through the optimal transmission power allocation ratio and the optimal offloading decision, the energy cost and time cost of the intelligent factory can be minimized, and the intelligent factory model itself restricts parameters such as the transmission power allocation ratio, the energy consumption of the production device itself, and the production time, ensuring that the optimal production plan will not exceed the time limit of the production task and will not cause production stagnation due to insufficient energy, while minimizing the production cost and ensuring that each production device has sufficient energy support and meets the time limit of production. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in this application or related technologies, the following will briefly introduce the drawings required for use in the description of the embodiments or related technologies. Obviously, the drawings in the following description are only the embodiments of this application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0022] Figure 1 It is a flowchart of the intelligent production method based on a directed acyclic graph according to an embodiment of this application;

[0023] Figure 2 It is a schematic diagram of the intelligent factory model according to an embodiment of this application;

[0024] Figure 3 It is a schematic diagram of the combined directed five-ring graph according to an embodiment of this application;

[0025] Figure 4Flow chart for building an intelligent factory model in the embodiments of this application;

[0026] Figure 5 Flow chart for solving the intelligent factory model by the linear goal programming algorithm with the minimum cost as the goal in the embodiments of this application;

[0027] Figure 6 Structure diagram of an intelligent production device based on a directed acyclic graph in the embodiments of this application;

[0028] Figure 7 Structure diagram of an electronic device in the embodiments of this application. Detailed implementation manners

[0029] To make the objectives, technical solutions and advantages of this application clearer and more understandable, the following further elaborates on this application in detail with reference to specific embodiments and the accompanying drawings.

[0030] It should be noted that unless otherwise defined, the technical terms or scientific terms used in the embodiments of this application should have the ordinary meanings understood by those with ordinary skills in the field to which this application belongs. The "first", "second" and similar terms used in the embodiments of this application do not denote any order, quantity or importance, but are only used to distinguish different components. The terms such as "including" or "comprising" mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects. The terms such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The terms such as "upper", "lower", "left" and "right" are only used to represent relative positional relationships, and when the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0031] In this article, it should be understood that any number of elements in the drawings is for illustration rather than limitation, and any naming is only for distinction and does not have any limiting meaning.

[0032] Based on the above description of the background technology, the following situations also exist in the related technology:

[0033] 1) Intelligent manufacturing: Intelligent Manufacturing is a human-machine integrated intelligent system composed of intelligent machines and humans, aiming to achieve supervised intelligent activities in the manufacturing process through aspects such as analysis, reasoning, judgment and decision-making. Through the cooperation between humans and machines, it expands, extends and partially replaces the physical and mental activities of human experts in the manufacturing process, and extends the concept of manufacturing automation to flexibility, integration and overall intelligence.

[0034] 2) Industrial Internet: The Industrial Internet of Things (IIoT) is a new type of infrastructure, application model, and industrial ecosystem that deeply integrates the new generation of information and communication technologies with the industrial economy. By comprehensively connecting people, machines, things, systems, etc., it constructs a brand-new manufacturing and service system covering the entire industrial chain and value chain, providing a way to realize the digital, networked, and intelligent development of industry and even the entire industry.

[0035] 3) Mobile Edge Computing: Mobile edge computing (MEC) is an edge computing mode that uses network edge nodes to process and analyze data. As the heterogeneity and complexity of the communication network architecture gradually increase, the inherent defects of traditional cloud computing, such as limited computing resources and large processing delays, also become apparent. The introduction of MEC can effectively share the computing pressure of the central processor, thus effectively improving the performance of the network architecture.

[0036] 4) Task Offloading: Task offloading is a key technology in edge computing. Computing offloading in edge computing is to offload the computing tasks of mobile terminals to the edge cloud environment, solving the deficiencies of devices in terms of resource storage, computing performance, and energy efficiency.

[0037] 5) Directed Acyclic Graph: A directed acyclic graph (DAG) is a data structure in graph theory. It consists of vertices and directed edges, where each edge has a direction and there are no loops. DAGs are often used to describe the dependency relationships between a series of tasks or events, such as the task flow of an engineering project, code dependencies in an encoder, etc. Due to the absence of loops, DAGs have strong computability and reliability, and can perform operations such as topological sorting efficiently.

[0038] In the intelligent manufacturing factory in the related technologies, in the technical solution of task scheduling for production equipment in the intelligent factory under the industrial Internet scenario, either a single directed acyclic graph is considered, or only one edge server is considered. However, in a real intelligent factory, there are various production equipment, and their production tasks are naturally different, so different directed acyclic graphs are needed to represent them. In addition, as the number of production equipment increases, the number of production tasks also increases. Using only one edge server is not sufficient to support the execution of all tasks, which may lead to task queuing and congestion, increasing the time cost.

[0039] Problem 1: In the actual intelligent factory scenario, the external dependencies of production equipment will affect the distribution of transmission power. In an intelligent manufacturing factory, on the one hand, the production tasks of some production equipment require data on the execution results of the production tasks of other production equipment. For example, some production tasks require certain sensor equipment to provide data such as temperature and humidity. That is, there is data sharing between the tasks of different production equipment, and the premise of data sharing is to allocate the corresponding transmission power. On the other hand, because during the execution of tasks, tasks may be offloaded to edge servers, and the data transmission in this process also requires the corresponding transmission power.

[0040] Problem 2: In the intelligent factory scenario, the constraints of limited self-energy of production equipment and limited running time of production tasks are not considered. The problem of minimizing production costs in the intelligent manufacturing factory scenario often does not consider the limit of the self-energy of production equipment. However, in an actual intelligent factory, the energy of each production equipment itself is limited, and they all tend to complete their own production tasks first, and the remaining energy is used for data sharing with other production tasks. At the same time, if the running time of each production task is not restricted, due to the too long running time of some production tasks, production will stagnate, increasing the time cost.

[0041] The intelligent production method, device, equipment and storage medium based on a directed acyclic graph provided by the embodiments of the present application determine the external dependency relationships between different production devices and the internal dependency relationships of the same production device; construct a combined directed acyclic graph according to the external dependency relationships and internal dependency relationships; construct a device model according to the device performance parameters of each production device and the server performance parameters of each edge server; construct an intelligent factory model according to the combined directed acyclic graph, the device model and the task information of each production device; solve the intelligent factory model by a linear goal programming algorithm with the goal of minimizing cost to obtain the optimal transmission power allocation ratio and the optimal offloading decision; determine the optimal production plan according to the optimal transmission power allocation ratio and the optimal offloading decision, and perform intelligent production according to the optimal production plan. By constructing a combined directed acyclic graph according to the internal dependency relationships and external dependency relationships, the dependency relationships between production tasks are uniformly managed, providing a dependency relationship basis for the reasonable allocation of transmission power resources between sending data to edge servers and sending data to other production devices. And a device model is constructed to determine the limitations of the performance of the production device itself. And an intelligent factory model is constructed to simulate the actual production process, and the intelligent factory model is solved with the goal of minimizing cost to obtain the optimal transmission power allocation ratio and the optimal offloading decision. Through the optimal transmission power allocation ratio and the optimal offloading decision, the energy cost and time cost of the intelligent factory can be minimized, and the intelligent factory model itself limits parameters such as the transmission power allocation ratio, the energy consumption of the production device itself, and the production time, ensuring that the optimal production plan will not exceed the time limit of the production task and will not cause production stagnation due to insufficient energy, while minimizing the production cost and ensuring that each production device has sufficient energy support and meets the time limit of production.

[0042] The following will specifically describe the intelligent production method based on a directed acyclic graph provided by the embodiments of the present application with reference to the accompanying drawings.

[0043] In some embodiments, as Figure 1 shown, an intelligent production method based on a directed acyclic graph includes:

[0044] Step 101: Determine the external dependency relationships between different production devices and the internal dependency relationships of the same production device.

[0045] Specifically, for an intelligent factory including N manufacturing equipment (ME), the i-th production device of the intelligent factory can be represented by ME i , i ∈ {1, 2,..., N}.

[0046] The execution tasks of each production device are composed of production tasks with dependencies, that is, the execution tasks include at least one production task. The internal dependencies of the same production device are the sequence relationships of the execution order between different production tasks of the same production device. When the production device runs, the production tasks are executed in sequence according to the internal dependencies. For example, the production tasks are executed in sequence: the 1st production task, the 2nd production task, the 3rd production task, until the last output task of production device ME i is executed, and production device ME i ends its operation or executes the production tasks of the next cycle. i

[0047] The external dependencies between different production devices are the sequence relationships of execution between production tasks of different production devices. For example, after the 2nd production task of production device ME i is completed, it is necessary to continue to execute the 3rd production task of production device ME i while also executing the 5th production task of another production device ME i′ . Because when the smart factory is manufacturing products, different production devices need to cooperate with each other to complete the manufacturing of corresponding products, there are certain external dependencies between different production devices.

[0048] Step 102: Construct a combined directed acyclic graph according to the external dependencies and internal dependencies.

[0049] In specific implementation, for the first directed acyclic Figure 1 graph →2→3→4→5→6→7, it can be directly determined according to the internal dependencies. After the 3rd task of the first directed acyclic graph is executed, the 4th production task is continued to be executed according to the internal dependencies. However, according to the external dependencies, it can be seen that while continuing to execute the 4th production task of its own production device, it is also necessary to continue to execute the 4th production tasks of other connected production devices. Therefore, multiple single directed acyclic graphs are constructed according to the internal dependencies, and different single directed acyclic graphs are combined according to the external dependencies to obtain the corresponding combined directed acyclic graph as shown Figure 2 . The combined directed acyclic graph represents the execution order between all production tasks of N production devices.

[0050] In some embodiments, each production device needs to execute at least one production task; constructing a combined directed acyclic graph according to the external dependencies and internal dependencies includes:

[0051] ​Step 1011: Determine the internal task set and the internal dependency set according to the internal dependency relationship, define the start task and the exit task in the internal task set as virtual tasks, and define the intermediate tasks in the internal task set as real tasks.

[0052] Specifically, when implementing, determine the internal task set and the internal dependency set according to the internal dependency relationship. Among them, the internal task set corresponding to the production equipment ME i is Among them, the first production task q i,0 executed in the internal task set is determined as the entry task, and the last production task executed in the internal task set is determined as the exit task. Since the entry task and the exit task are locally processed on the production equipment ME i and the computational amount during task execution is zero, the entry task and the exit task are defined as virtual tasks. Then, all production tasks in the internal task set except the virtual tasks are real tasks. Among them, D i represents the number of production tasks of the production equipment ME i , and at the same time represents the number of real tasks in the internal task set. When defining the production task, for any production task q i,j , it is defined that i in the subscript represents the i-th production equipment, and j represents the j-th production task.

[0053] Then use ID i to represent the internal dependency set. Then, the elements in the internal dependency set corresponding to the production equipment ME i can be expressed as {id(q i,j , q i,k )|j ∈ {0, 1, 2,..., D i}, q i,k ∈ suc(q i,j )}, which represents a directed edge from the production task q i,j to the production task q i,k , and the size of the transmitted data is indicating that after the production task q i,j is completed, it is necessary to continue to execute the production task q i,k , and the amount of transmitted data between the two production tasks is

[0054] Step 1012: Determine the external dependency set according to the external dependency relationship.

[0055] Specifically, when implementing, determine the external dependency set according to the external dependency relationship. Among them, use ED i to represent the dependency set corresponding to the production equipment ME i , and the external dependency set ED iThe elements in can be expressed as {ed(q i,j ,q i′,j′ )|i≠i′,j∈{0,1,2,...,D i},j′∈{0,1,2,...,D i′}}, which represents a directed edge from production task q i,j to task q i′,j′ , and the size of the transmitted data is It means that after the production task q i,j is completed, it is necessary to continue to execute the production task q i′,j′ of other production equipment, and the amount of transmitted data between the two production tasks is Then the single directed acyclic graph of a single production equipment ME i can be defined as G i =(Q i ,ID i ,ED i ).

[0056] Step 1013: Construct a combined directed acyclic graph according to the internal dependency set, external dependency set and internal task set.

[0057] Specifically, when implemented, according to a single internal task set, the production tasks of a single production equipment can be determined, and the internal task sets of all production equipment include all production tasks that the intelligent factory needs to execute. The internal dependency set includes the dependency relationships between all production tasks within a single internal task set corresponding to the same production equipment, while the external dependency set includes the dependency relationships between the production tasks of different production equipment. Integrating the internal dependency set, external dependency set and internal task set can understand all production tasks and the dependency relationships between production tasks. Therefore, multiple single directed acyclic graphs are constructed according to the internal dependency relationships, and virtual tasks and real tasks in the single directed acyclic graph are defined according to the internal task set. Different single directed acyclic graphs are combined according to the external dependency relationships to obtain the corresponding combined directed acyclic graph as shown in Figure 2 . Among them, the numbers between different production tasks represent the amount of transmitted data for data transmission. Then, the combined directed acyclic graph is used to describe the dependency relationships between all production tasks in the intelligent factory. Since the combined directed acyclic graph has no loops, it has strong computability and reliability, and can efficiently perform operations such as topological sorting, providing a sufficient basis for dependency relationships for subsequent model construction processes.

[0058] By constructing a combined directed acyclic graph according to the internal dependency relationships and external dependency relationships, the unified management of the dependency relationships between production tasks is realized, providing a basis for the reasonable allocation of transmit power resources between sending data to the edge server and sending data to other production equipment.

[0059] Step 103: Construct an equipment model according to the equipment performance parameters of each production device and the server performance parameters of each edge server.

[0060] Specifically, according to the combined directed acyclic graph, all production tasks in the intelligent factory and the dependency relationships between production tasks can be understood. As the main bodies for executing production tasks, production devices and edge servers have their own constraints. Because in the actual use process, without replacing the devices or servers, the equipment performance parameters of production devices and the server performance parameters of edge servers are fixed. Therefore, it is first necessary to construct an equipment model according to the equipment performance parameters of each production device and the server performance parameters of each edge server to determine the performance of production devices and edge servers, providing basic model data on equipment performance for subsequent construction of the intelligent factory model.

[0061] Exemplarily, in an intelligent factory, N production devices and M edge servers (EdgeServer, ES) are randomly arranged. The production device can be represented as ME i , i ∈ {1, 2,..., N}. The production device ME i has a computing power of f i , a working power of a transmission power of a receiving power of a standby power of and its own available energy of The edge server can be represented as ES m , m ∈ {1, 2,..., M}. The edge server ES m has a computing power of f m . Among them, each edge server can execute multiple production tasks simultaneously, and the production operations of each production device start at the same time. Therefore, by constructing an equipment model to determine the limitations of the performance of production devices themselves. For example, the maximum energy that the production device ME i can use cannot exceed its own available energy of the energy limit, the maximum transmission power when sending production task data cannot exceed the transmission power the transmission power limit, the receiving power when receiving production task data cannot exceed the receiving power of the receiving power limit, and the operating power during standby cannot exceed the standby power of the standby power limit.

[0062] Step 104: Construct an intelligent factory model according to the combined directed acyclic graph, the equipment model, and the task information of each production device.

[0063] In specific implementation, considering the self - energy of production equipment, the running time of production tasks, and the computing power of the edge server comprehensively, the intelligent factory is abstracted into an intelligent factory model as shown in Figure 3 The intelligent factory model is a two - layer wireless communication network structure, namely the upper - layer edge layer and the lower - layer equipment layer. The intelligent factory model includes an equipment model, a task model, a communication model, a task execution model, and an external dependency evaluation model. Therefore, the process of constructing the intelligent factory model can be decomposed into the processes of constructing the equipment model, the task model, the communication model, the task execution model, and the external dependency evaluation model. Then, the constructed equipment model, task model, communication model, task execution model, and external dependency evaluation model are integrated to obtain the intelligent factory model.

[0064] Step 105: Taking the minimum cost as the goal, solve the intelligent factory model through the linear goal programming algorithm to obtain the optimal transmit power allocation ratio and the optimal offloading decision.

[0065] In specific implementation, first construct a cost objective function with the minimum cost as the goal, and then solve it through the linear goal programming algorithm based on the intelligent factory model. Since the intelligent factory model is a simulation of the actual intelligent factory, the intelligent factory model can determine various constraints for solving the cost objective function. Therefore, by solving the intelligent factory model through the linear goal programming algorithm, the obtained optimal transmit power allocation ratio and the optimal offloading decision are optimal control parameters that can be actually used. Because the solution process is completed under the constraints of the intelligent factory model, the optimal transmit power allocation ratio and the optimal offloading decision will not exceed the constraint limits of the intelligent factory model, ensuring the effectiveness of the optimal transmit power allocation ratio and the optimal offloading decision.

[0066] Step 106: Determine the optimal production plan according to the optimal transmit power allocation ratio and the optimal offloading decision, and perform intelligent production according to the optimal production plan.

[0067] During specific implementation, when determining the optimal production plan according to the optimal transmission power allocation ratio and the optimal offloading decision, and performing intelligent production according to the optimal production plan, reasonable energy allocation is performed for production tasks with different dependencies (internal task dependencies and external task dependencies). Moreover, the optimal transmission power allocation ratio and the optimal offloading decision can reasonably allocate the transmission power for both data sharing between tasks and task offloading to the edge, making full use of the transmission power of production equipment and effectively allocating the transmission power resources. The intelligent factory model realizes the constraints on the energy of production equipment itself and the running time of each production task, ensuring that each production task can have sufficient energy support and can be completed within the specified time when performing intelligent production according to the optimal production plan, ensuring that the equipment in the entire intelligent factory can complete a production operation and minimizing its production cost.

[0068] In summary, the intelligent production method based on a directed acyclic graph provided by the embodiments of the present application constructs a combined directed acyclic graph according to internal and external dependencies to realize unified management of the dependencies between production tasks, providing a dependency basis for reasonable allocation of transmission power resources between sending data to an edge server and sending data to other production equipment. And a device model is constructed to determine the limitations of the performance of production equipment itself. And an intelligent factory model is constructed to simulate the actual production process, and the intelligent factory model is solved with the goal of minimizing costs to obtain the optimal transmission power allocation ratio and the optimal offloading decision. Through the optimal transmission power allocation ratio and the optimal offloading decision, the energy cost and time cost of the intelligent factory can be minimized, and the intelligent factory model itself restricts parameters such as the transmission power allocation ratio, the energy consumption of production equipment itself, and the production time, ensuring that the optimal production plan will not exceed the time limit of production tasks and will not cause production stagnation due to insufficient energy, minimizing the production cost while ensuring that each production equipment has sufficient energy support and meets the time limit of production.

[0069] In some embodiments, as Figure 4 shown, constructing an intelligent factory model according to the combined directed acyclic graph, the device model, and the task information of each production equipment includes:

[0070] Step 401: Determine a task model according to the combined directed acyclic graph and the task information of each production equipment.

[0071] During specific implementation, determining a task model according to the combined directed acyclic graph and the task information of each production equipment includes:

[0072] Step 4011: Determine the previous generation tasks of each production task according to the combined directed acyclic graph to obtain a set of previous generation tasks.

[0073] Among them, according to the combined directed acyclic graph, all production tasks and the dependency relationships between production tasks in the intelligent factory can be understood. Then, the predecessor tasks of each production task can be determined based on the combined directed acyclic graph. Exemplarily, as Figure 2 shown, the predecessor tasks of production task q 2,7 include production task q 2,4 , production task q 2,5 and production task q 2,6 . And production task q 2,7 , production task q 2,4 , production task q 2,5 and production task q 2,6 all belong to the same production equipment and can be determined according to the internal dependency relationship in the combined directed acyclic graph; the predecessor tasks of production task q 1,6 include production task q 2,7 and production task q 1,5 . Among them, production task q 1,5 and production task q 1,6 belong to the same production equipment and can be determined according to the internal dependency relationship in the combined directed acyclic graph; production task q 1,6 and production task q 2,7 belong to different production equipment and can be determined according to the external dependency relationship in the combined directed acyclic graph.

[0074] It can be seen that the production tasks in the combined directed acyclic graph can serve as the input ends of multiple external dependencies and also as the output ends of multiple external dependencies. Then, the set of predecessor tasks of production task q i,j is defined as ψ i,j ={q l,s ∈q i′,j′ ∪q i,k |ed(q i′,j′ ,q i,j )∈ED i ,id(q i,k ,q i,j )∈ID i}, representing all the predecessor tasks of production task q i,j .

[0075] Step 4012: Determine the offloading decision of each production task in the combined directed acyclic graph according to the task information.

[0076] The task information of production task q i,j includes three elements, denoted as (x i,j ,o i,j ,z i,j ), where x i,j is the amount of computation for task execution, representing the amount of computation required when executing production task q i,j ; oi,j For production task q i,j The amount of data returned when the task is executed on the edge server. If the task is offloaded to the edge server for execution, the production task is considered completed only when the calculation result data is returned to the production device; i,j For production task q i,j The uninstall decision when z i,j = 0, the production task q i,j By production equipment ME i Self-execution, the unloading destination is the production equipment ME i itself. Otherwise, the production task q i,j The offload is executed to the edge server, and the offload destination is any edge server. i,j The uninstall decision is as follows:

[0077]

[0078] Step 4013: Determine the unloading destination of the production task in the combined directed acyclic graph in the edge server and the production equipment corresponding to the production task according to the unloading decision, integrate the previous generation task set, the combined directed acyclic graph and the unloading destination of each production task to obtain the task model.

[0079] In the specific implementation, it should be noted that the virtual task can only be executed by the production equipment itself, so the unloading decision of starting the task and exiting the task is 0, that is, By integrating the previous generation task set, combining the directed acyclic graph and the unloading destination of each production task, the relationship between production tasks, the execution order and the task unloading destination can be determined, all model data related to the production task can be determined, and the task model can be obtained.

[0080] Step 402: Determine a transmission power allocation decision for a target production device according to the task model, and determine a first transmission rate between production devices and a second transmission rate between an edge server and the production device according to the device model.

[0081] In specific implementation, when the production equipment ME i To production equipment ME i′ Send production task q i,j When the data is the transmit power Used to produce equipment ME i With production equipment ME i′ The first distribution ratio of data sent between Indicates the transmit power Used to produce equipment ME i The second distribution ratio of data between the edge server ES. Denote the production task as q i,j Without external dependencies, the production task q i,j is executed by the production equipment ME i itself. If the production task q i,j has external dependencies, the transmit power must be allocated. Then, is used as the transmit power allocation decision.

[0082] Through orthogonal frequency division multiplexing, each production equipment uses an orthogonal channel to communicate with the edge server, ensuring no interference between communication links. When the production task q i,j needs to be offloaded to the edge server ES m for execution, the production equipment ME i has to transfer the data required by the production task q i,j to ES m . According to Shannon's theory, the achievable uplink transmission rate is the second transmission rate:

[0083]

[0084] where w is the channel bandwidth between the production equipment and the edge server, δ 2 is the channel noise between the production equipment and the edge server, h i,m is the channel gain between the production equipment ME i and the edge server ES m , which depends on the distance d i between the production equipment ME m and the edge server ES i,m , that is where ρ is the path loss exponent. Assume that each wireless channel is symmetric, i.e., the achievable downlink transmission rate is the same as the uplink transmission rate.

[0085] Due to the existence of external dependencies, data transmission also occurs between production equipment, using channels in different frequency bands for transmission to avoid interference. When the production task q i′ in G i′,j′ needs the data of the production task q i in G i,j , the data of the production task q i,j needs to be transmitted to the production task q i′ ,j ′ , and the transmission rate at this time is the second transmission rate:

[0086]

[0087] where w0 is the channel bandwidth for communication between production equipment ME, is the channel noise for communication between production devices ME, h i,i′ is the production device ME i and the production device ME i′ The channel gain for communication between them depends on the production device ME i and the production device ME i′ The distance d between them i,i′ , that is where ρ is the path loss exponent.

[0088] Step 403: Construct a communication model based on the transmit power allocation decision, the first transmission rate, and the second transmission rate.

[0089] In specific implementation, constructing a communication model based on the transmit power allocation decision, the first transmission rate, and the second transmission rate includes:

[0090] Step 4031: Determine the task dependency relationship between the target task and the target previous-generation task of the target task according to the task model.

[0091] In specific implementation, let q i,j represent the target task, and let q l,s represent the target previous-generation task of the target task. If the directed edge between the target task and the target previous-generation task belongs to the internal dependency set ID i , determine that the task dependency relationship between the target task and the target previous-generation task of the target task is an internal dependency relationship. If the directed edge between the target task and the target previous-generation task belongs to the external dependency set ED i , determine that the task dependency relationship between the target task and the target previous-generation task of the target task is an external dependency relationship.

[0092] Step 4032: In response to the task dependency relationship being an internal dependency relationship, determine the edge transmission delay according to the first transmission rate and the amount of transmitted data of the target task, and determine the edge transmission delay as the task transmission delay.

[0093] Step 4033: In response to the task dependency relationship being an external dependency relationship, determine the edge transmission delay according to the first transmission rate and the amount of transmitted data of the target task, determine the external transmission delay according to the second transmission rate and the amount of transmitted data of the target task, and determine the sum of the external transmission delay and the edge transmission delay as the task transmission delay.

[0094] In specific implementation, because there are different situations in the task dependency relationship between the target task and the target previous-generation task of the target task, the task transmission delay for the data required by the target task q i,j to be transmitted from its target previous-generation task q l,s to its execution location can be calculated as:

[0095]

[0096] Among them, when the target previous task q l,s and the target task q i,j are in an internal dependency relationship, that is, id(q l,s , q i,j ) ∈ ID l at this time, then the data transmission delay only considers whether to offload the target task q i,j to the edge server ES, and only calculates the edge transmission delay That's it. If the target previous task q l,s and the target task q i,j are in an external dependency relationship, that is, ed(q l,s , q i,j ) ∈ ED l at this time, then there is an external transmission delay in the transmission between the target previous task q l,s and the target task q i,j At the same time, it also considers the edge transmission delay of whether to offload the target task q to the edge server ES i,j The edge transmission delay

[0097] Among them, 1 {#} is an indicator function, that is, if the event # is true, the function value is 1, and if the event # is false, the function value is 0. For the edge transmission delay and If the offloading decision is z i,j = 0, it means that the production task q i,j is executed by the production equipment itself, and there is no process of uploading task data to the edge server, and the event is false. If the offloading decision is z i,j = m, it means that the production task q i,j is executed by the edge device ES m , and it is necessary to upload the task data to the edge server, and the event is true.

[0098] Step 4034: Determine the task reception delay according to the return data volume of the target task and the second transmission rate.

[0099] Specifically, when the event # is true, when the target task q i,j is executed and completed in the edge server ES m , the result obtained needs to be returned to the production equipment ME i , then the transmission delay for receiving the execution result is:

[0100]

[0101] Step 4035: Determine the communication delay according to the task reception delay and the task transmission delay.

[0102] In specific implementation, the communication delay is 0 when the event is false. When the event is true, the sum of the task reception delay and the task transmission delay is determined as the communication delay, and the communication delay represents the delay generated during task data transmission when executing the production task.

[0103] Step 4036: Determine the transmission energy consumption of the target task according to the transmit power allocation decision, the transmit power of the target production device, the receive power of the target production device, and the amount of transmitted data.

[0104] In specific implementation, when transferring the data required by the target task q i,j to its execution location, the transmission energy consumption that can be consumed by the production device ME i can be calculated as:

[0105]

[0106] where represents the transmission energy consumption of transferring the target task q i,j to the edge server when there is no external dependency. represents the energy consumption of external dependency transmission and the energy consumption of transferring the task to the edge server when there is an external dependency.

[0107] Step 4037: Determine the reception energy consumption according to the amount of returned data, the second transmission rate, and the receive power of the target production device.

[0108] In specific implementation, when the target task q i,j is executed in the edge server ES m and the execution result needs to be returned to the production device ME i , the reception energy consumption for receiving the execution result is:

[0109]

[0110] Step 4038: Determine the communication energy consumption according to the reception energy consumption and the transmission energy consumption.

[0111] In specific implementation, according to different offloading decisions, the event may be true or false. When the event is false, only the transmission energy consumption needs to be considered, and the transmission energy consumption is determined as the communication energy consumption. When the event is true, the sum of the reception energy consumption and the transmission energy consumption is determined as the communication energy consumption.

[0112] Step 4039: Integrate the communication delay and the communication energy consumption to obtain a communication model.

[0113] In specific implementation, the communication model includes the communication delay and communication energy consumption for data transmission between production devices and between production devices and edge servers. Then, the communication model can be obtained by integrating the communication delay and communication energy consumption.

[0114] Step 404: Determine the first computing power, energy consumption parameters of the target production device, and the second computing power of the edge server according to the device model.

[0115] In specific implementation, the first computing power of the target production device is f i , and the second computing power of the edge server is f m . The energy consumption parameters of the target production device include the working power of the transmission power of the receiving power of the standby power of and its available energy of

[0116] Step 405: Construct a task execution model according to the first computing power, energy consumption parameters, and the second computing power.

[0117] In specific implementation, the energy consumption parameters include the working power and the standby power; constructing a task execution model according to the first computing power, energy consumption parameters, and the second computing power includes:

[0118] Step 4051: Determine the task completion time of the target previous task of the target task according to the task model, and determine the task transmission delay between the target task and the target previous task according to the communication model, and determine the sum of the task completion time and the task transmission delay as the calculation start time.

[0119] In specific implementation, due to the internal and external dependencies between tasks, the complexity of execution delay analysis is increased. For convenience, use to represent the task start time of executing the target task q i,j , to represent the running time of executing the target task q i,j , to represent the completion time of executing the target task q i,j .

[0120] The calculation start time of the target task q i,j is the completion time of its target previous task q l,s ∈ψ i,j and the task transmission delay of transmitting the data of the target previous task q l,s ∈ψ i,j to the execution location of the task q i,j , and

[0121] Step 4052: Determine the task start time of the target task as the maximum calculation start time among the calculation start times corresponding to different previous tasks.

[0122] In specific implementation, the task start time of the target task q i,j is the completion time of its target previous task q l,s ∈ψ i,j and the maximum value of the sum of the task transmission delays for transmitting the data of the target previous task q ∈ψ l,s to the execution location of task q i,j . Since only when the last previous task is transmitted and completed can it be ensured that all the data required for the execution of the target task has been obtained. Therefore, determine the maximum calculation start time among the calculation start times corresponding to different previous tasks as the task start time of the target task, and the task start time is expressed as: i,j the task transmission delay The sum. Because only when the last previous task is transmitted and completed can it be ensured that all the data required for the execution of the target task has been obtained. Therefore, determine the maximum calculation start time among the calculation start times corresponding to different previous tasks as the task start time of the target task, and the task start time is expressed as:

[0123]

[0124] Step 4053: Determine the target offloading destination of the target task according to the task model.

[0125] In specific implementation, when the offloading decision of the task model is z i,j = 0, the target task q i,j is executed by the production device ME i itself, and the target offloading destination is the target production device corresponding to the target task. When the offloading decision of the task model is z i,j = m, the target task q i,j is offloaded to the edge server ES m for execution, and the target offloading destination is the edge server.

[0126] Step 4054: In response to the target offloading destination being the target production device corresponding to the target task, determine the internal execution delay according to the task execution calculation amount of the target task and the first computing ability, determine the product of the internal execution delay and the working power as the internal computing energy consumption, determine the internal execution delay as the task execution delay, and determine the internal computing energy consumption as the task execution energy consumption.

[0127] Step 4055: In response to the target offloading destination being the edge server, determine the external execution delay according to the task execution calculation amount of the target task and the second computing ability, determine the product of the external execution delay and the standby power as the internal standby energy consumption, determine the sum of the external execution delay and the task reception delay as the task execution delay, and determine the sum of the internal standby energy consumption and the reception energy consumption as the task execution energy consumption.

[0128] In specific implementation, when the target task q i,j is executed by the target production device ME i itself, that is, when z i,j = 0, the internal execution delay can be determined according to the task execution computing amount of the target task and the first computing ability, which can be expressed as:

[0129]

[0130] Similarly, when the target task q i,j is offloaded to the edge server ES m for execution, that is, when z i,j = m, the external execution delay can be determined according to the task execution computing amount of the target task and the second computing ability, which can be expressed as:

[0131]

[0132] When the target task q i,j is executed by the target production device ME i itself, that is, when z i,j = 0, only the energy consumption of the production device is considered. Therefore, when the task q i,j is executed by the production device ME i itself, the product of the internal execution delay and the working power needs to be determined as the internal computing energy consumption, and the internal computing energy consumption can be expressed as:

[0133]

[0134] Although the computing energy consumption of the edge server ES is not considered when the target task is offloaded to the edge server, during the execution of the target task on the edge server, the target production device is in a standby state and is ready to receive the computing result from the edge server at any time. Therefore, the standby energy consumption of the target production device also needs to be considered, and the product of the external execution delay and the standby power is determined as the internal standby energy consumption, and the internal standby energy consumption can be expressed as:

[0135]

[0136] Therefore, when the target offloading destination is the target production device corresponding to the target task, the internal execution delay is determined as the task execution delay, and the internal computing energy consumption is determined as the task execution energy consumption. When the target offloading destination is the edge server, the sum of the external execution delay and the task reception delay is determined as the task execution delay, and the internal standby energy consumption and the data reception energy consumption are determined as the task execution energy consumption.

[0137] Then the task execution delay (also known as the running time) is expressed as:

[0138]

[0139] The task execution energy consumption (also known as the running energy consumption) is expressed as:

[0140]

[0141] Step 4056: Determine the target completion time of the target task according to the task start time and the task execution delay.

[0142] In specific implementation, determine the target completion time of the target task according to the task start time and the task execution delay:

[0143]

[0144] Step 4057: In response to the target task being the final production task, determine the target completion time as the equipment production completion time of the target production equipment, and determine the maximum value among the equipment production completion times corresponding to different production equipment as the total production time.

[0145] In specific implementation, when the target production task is the last production task of the target production equipment, that is, j = D i At this time, the target completion time Is equal to the equipment completion time of the target production equipment That is Since the total production time of a single production operation of the production equipment Is the equipment completion time of the last production task in the single directed acyclic graph corresponding to the production equipment Then there is For the entire intelligent factory, only when the last production equipment completes production can it be determined that a complete production operation has been completed. Then, the total production time for all production equipment in the intelligent factory to perform a production operation is the maximum value among the equipment production completion times corresponding to different production equipment. The total production time is expressed as:

[0146]

[0147] Step 4058: Determine the single - equipment energy consumption of the target production equipment according to the task execution energy consumption and the transmission energy consumption, and determine the sum of the single - equipment energy consumptions of all production equipment as the total energy consumption.

[0148] In specific implementation, the single - equipment energy consumption consumed by the production equipment ME i For a single production operation can be expressed as:

[0149]

[0150] The total energy consumption for the intelligent factory to perform a production operation is the sum of the single-device energy consumptions of all production devices, and the total energy consumption is expressed as:

[0151]

[0152] Step 4059: Integrate the total energy consumption and the total production time to obtain a task execution model.

[0153] In specific implementation, the task execution model includes the device production completion time and the single-device energy consumption of a single device, as well as the total energy consumption and the total production time of the entire intelligent factory. The device production completion time is a component of the total production time, and the single-device energy consumption is a component of the total energy consumption. Therefore, the task execution model can be obtained by integrating the total energy consumption and the total production time.

[0154] Step 406: Determine the transmission power of the production device according to the device model, and construct an external dependency evaluation model based on the transmission power and the transmission power allocation decision.

[0155] In specific implementation, the external dependency relationship between production devices can improve production efficiency and save some production costs. However, due to the limited resources of production devices, their available energy and transmission power will be consumed during the external dependency data sharing process. Therefore, it is necessary to reasonably evaluate the energy consumption and the allocation of transmission power of production devices, and propose an external dependency evaluation model for evaluation. The external dependency evaluation model can be expressed as:

[0156]

[0157] As the amount of shared data in the external dependency increases, the advantages of data sharing gradually weaken and even become negligible. Therefore, use to evaluate the impact of the amount of shared data, and the total external dependency evaluation is expressed as:

[0158]

[0159] Step 407: Integrate the device model, the task model, the communication model, the task execution model, and the external dependency evaluation model to obtain an intelligent factory model.

[0160] In specific implementation, by integrating the device model, the task model, the communication model, the task execution model, and the external dependency evaluation model, all functions of the intelligent factory can be fully executed, the simulation of the intelligent factory can be realized, and an intelligent factory model can be obtained.

[0161] In some embodiments, as Figure 5 shown, with the goal of minimizing cost, the intelligent factory model is solved by a linear goal programming algorithm to obtain the optimal transmission power allocation ratio and the optimal offloading decision, including:

[0162] Step 501: With the goal of minimizing cost, construct a cost objective function based on the total production time, total energy consumption, and total evaluation of external dependencies of the intelligent factory model.

[0163] Specifically, when implemented, the cost objective function with the goal of minimizing cost can be expressed as:

[0164]

[0165] Among them, E represents the total energy consumption, T represents the total production time, Eva represents the total evaluation of external dependencies, and α, β, and γ are weight parameters. Θ and Z are variables of the optimization problem, Θ represents the transmission power allocation ratio, and Z represents the offloading decision.

[0166] Step 502: Determine the task offloading constraints according to the intelligent factory model and the number of servers of the edge server.

[0167] Step 503: Determine the energy constraints according to the intelligent factory model and the available energy of the production equipment.

[0168] Step 504: Determine the time constraints according to the intelligent factory model and the preset maximum running time.

[0169] Step 505: Determine the constraint on the range of transmission power allocation ratio according to the preset ratio range.

[0170] Step 506: Solve the minimum value of the cost objective function through the linear goal programming algorithm under the constraint set to obtain the optimal transmission power allocation ratio and the optimal offloading decision; among them, the constraint set includes task offloading constraints, energy constraints, time constraints, and the constraint on the preset range of transmission power allocation ratio.

[0171] Specifically, when implemented, the constraint set is expressed as:

[0172]

[0173] Among them, the transmission power allocation ratio range constraint C1 represents the range of the transmission power allocation ratio for data sharing in external dependencies. Only when the allocation ratio is within the preset ratio range (0, 1) does it indicate that the transmission power has been allocated. The task offloading constraints include the offloading range constraint C2 and the virtual task offloading constraint C3. The intelligent factory model restricts the selection of task offloading destinations, that is, it can only be the production equipment itself and the edge server. The number of servers of the edge server restricts the selection range of the edge server, and thus the offloading range constraint C2 is obtained. In the intelligent factory model, the incoming tasks and outgoing tasks can only be executed on the production equipment itself, and thus the virtual task offloading constraint C3 is obtained. The energy constraint C4 is determined according to the intelligent factory model and the available energy of the production equipment, which is the self-energy constraint of each production equipment ME. It ensures that, on the premise that the available energy meets the single-device energy consumption of its own executed tasks, it provides data sharing energy for the allocated transmission power. The time constraint C5 restricts that the total production time of the intelligent factory for tasks cannot exceed the maximum running time.

[0174] Under the constraints of the constraint set, the minimum value of the cost objective function is solved through the linear goal programming algorithm to obtain the optimal transmission power allocation ratio and the optimal offloading decision. That is, when the calculation result of the cost objective function is the smallest, Θ is the optimal transmission power allocation ratio, and when the calculation result of the cost objective function is the smallest, Z is the optimal offloading decision. The optimal transmission power allocation ratio is obtained by changing the transmission power input. The optimal offloading decision determines whether each production equipment executes the production task itself or offloads the production task to the edge server, and when offloading to the edge server, it selects the most suitable edge server considering the comprehensive energy consumption and delay, while ensuring low energy consumption and low delay. The optimal transmission power allocation ratio can most efficiently utilize the available energy to avoid the situation of insufficient energy.

[0175] By constructing an intelligent factory model to simulate the actual production process and solving the intelligent factory model with the minimum cost as the goal, the optimal transmission power allocation ratio and the optimal offloading decision are obtained. Through the optimal transmission power allocation ratio and the optimal offloading decision, the energy cost and time cost of the intelligent factory can be minimized. Moreover, the intelligent factory model itself restricts parameters such as the transmission power allocation ratio, the self-energy consumption of production equipment, and the production time, ensuring that the optimal production plan will not exceed the time limit of the production task and will not lead to the situation of production stagnation due to insufficient energy, minimizing the production cost while ensuring that each production equipment has sufficient energy support and meets the time limit of production.

[0176] It should be noted that the method of the embodiment of the present application can be executed by a single device, such as a computer or a server. The method of this embodiment can also be applied to a distributed scenario and completed by multiple devices cooperating with each other. In such a distributed scenario, one of the multiple devices can only execute one or more steps of the method of the embodiment of the present application, and these multiple devices will interact with each other to complete the described method.

[0177] It should be noted that some embodiments of the present application have been described above. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the above embodiments and still achieve the desired result. Additionally, the processes depicted in the figures do not necessarily require the specific order or sequential order shown to achieve the desired result. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0178] In summary, for Problem 1, the embodiment of the present application proposes a determination model for the optimal transmission power allocation ratio, which first reasonably allocates power to task pairs with different dependencies (internal dependencies and external dependencies); then reasonably allocates transmission power for the two cases of data sharing between tasks and task offloading to the edge. It can make full use of the transmission power of production equipment and effectively allocate transmission power resources.

[0179] For Problem 2, the embodiment of the present application sets the available energy and maximum operating time of the production equipment itself to ensure that each production task can have sufficient energy support, and can ensure that the equipment in the entire intelligent factory can complete a production operation within the specified time and minimize its production cost.

[0180] Based on the same inventive concept, corresponding to the method of any of the above embodiments, the present application also provides an intelligent production device based on a directed acyclic graph.

[0181] Reference Figure 6 , the intelligent production device based on a directed acyclic graph includes:

[0182] Dependency relationship determination module 10, configured to: determine the external dependency relationship between different production devices and the internal dependency relationship of the same production device;

[0183] Combined directed acyclic graph module 20, configured to: construct a combined directed acyclic graph according to the external dependency relationship and the internal dependency relationship;

[0184] Device model construction module 30, configured to: construct a device model according to the device performance parameters of each production device and the server performance parameters of each edge server;

[0185] The factory model construction module 40 is configured to: construct an intelligent factory model based on the combined directed acyclic graph and the task information of each production device;

[0186] The model optimal solution module 50 is configured to: aiming at minimizing the cost, solve the intelligent factory model through the linear goal programming algorithm to obtain the optimal transmission power allocation ratio and the optimal offloading decision;

[0187] The production plan generation module 60 is configured to: determine the optimal production plan according to the optimal transmission power allocation ratio and the optimal offloading decision, and perform intelligent production according to the optimal production plan.

[0188] For the convenience of description, when describing the above device, various modules are described separately according to their functions. Of course, when implementing this application, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0189] The device in the above embodiment is used to implement the corresponding intelligent production method based on the directed acyclic graph in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be elaborated here.

[0190] Based on the same inventive concept, corresponding to the method in any of the above embodiments, the present application also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the intelligent production method based on the directed acyclic graph described in any of the above embodiments.

[0191] Figure 7 FIG. shows a more specific schematic diagram of the hardware structure of the electronic device provided in this embodiment. The device may include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. Among them, the processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040 are communicatively connected to each other inside the device through the bus 1050.

[0192] The processor 1010 can be implemented in a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, etc., and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0193] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage devices, dynamic storage devices, etc. The memory 1020 can store the operating system and other application programs. When implementing the technical solutions provided in the embodiments of this specification through software or firmware, the relevant program codes are stored in the memory 1020 and called and executed by the processor 1010.

[0194] The input / output interface 1030 is used to connect to the input / output module to implement information input and output. The input / output module can be configured as a component in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Among them, the input devices can include keyboards, mice, touchscreens, microphones, various sensors, etc., and the output devices can include displays, speakers, vibrators, indicator lights, etc.

[0195] The communication interface 1040 is used to connect to the communication module (not shown in the figure) to implement communication interaction between this device and other devices. Among them, the communication module can implement communication through wired means (such as USB, network cable, etc.) or through wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0196] The bus 1050 includes a path for transmitting information between various components of the device (such as the processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040).

[0197] It should be noted that although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040, and the bus 1050, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, those skilled in the art can understand that the above device may also only include the components necessary to implement the solutions of the embodiments of this specification, and do not have to include all the components shown in the figure.

[0198] The electronic device in the above embodiment is used to implement the corresponding intelligent production method based on a directed acyclic graph in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be elaborated here.

[0199] Based on the same inventive concept, corresponding to the method in any of the above embodiments, the present application also provides a non-transitory computer-readable storage medium. The non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to cause the computer to execute the intelligent production method based on a directed acyclic graph as described in any of the above embodiments.

[0200] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tapes, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible by a computing device.

[0201] The computer instructions stored in the storage medium of the above embodiment are used to cause the computer to execute the intelligent production method based on a directed acyclic graph as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be elaborated here.

[0202] Based on the same concept, corresponding to the method of any of the above embodiments, the present application also provides a computer program product, including computer program instructions, which when run on a computer, cause the computer to execute the method as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be elaborated here.

[0203] It can be understood that before using the technical solutions of the various embodiments of the present disclosure, the types, usage scopes, usage scenarios, etc. of the personal information involved will be informed to the user in an appropriate manner, and the user's authorization will be obtained.

[0204] For example, when responding to a user's active request, a prompt message is sent to the user to clearly prompt the user that the operation requested by the user will require obtaining and using the user's personal information. Thus, the user can autonomously choose whether to provide personal information to software or hardware such as an electronic device, application program, server, or storage medium that performs the operations of the technical solutions of the present disclosure according to the prompt message.

[0205] As an optional but non-limiting implementation manner, the way of sending a prompt message to the user in response to receiving the user's active request can be, for example, in the form of a pop-up window, and the prompt message can be presented in text in the pop-up window. In addition, the pop-up window can also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0206] It is understandable that the above notification and the process of obtaining user authorization are merely illustrative and do not limit the implementation manners of the present disclosure. Other manners that comply with relevant laws and regulations can also be applied to the implementation manners of the present disclosure.

[0207] Those of ordinary skill in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of the present application is limited to these examples; within the concept of the present application, the technical features in the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations in different aspects of the embodiments of the present application as described above. For the sake of brevity, they are not provided in detail.

[0208] In addition, for simplicity of explanation and discussion, and so as not to make the embodiments of the present application difficult to understand, the known power / ground connections to integrated circuit (IC) chips and other components may or may not be shown in the accompanying drawings. Further, the devices may be shown in block diagram form so as not to make the embodiments of the present application difficult to understand, and this also takes into account the fact that the details of the implementation manners of these block diagram devices are highly dependent on the platform on which the embodiments of the present application are to be implemented (i.e., these details should be fully within the understanding of those skilled in the art). In the case where specific details (such as circuits) are set forth to describe exemplary embodiments of the present application, it will be apparent to those skilled in the art that the embodiments of the present application can be implemented without these specific details or with variations of these specific details. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0209] Although the present application has been described in connection with specific embodiments of the present application, many alternatives, modifications, and variations of these embodiments will be apparent to those of ordinary skill in the art based on the foregoing description. For example, other memory architectures (such as dynamic RAM (DRAM)) can be used with the embodiments discussed.

[0210] The embodiments of the present application are intended to cover all such alternatives, modifications, and variations that fall within the broad scope of the claims of the present application. Therefore, any omissions, modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the embodiments of the present application shall be included within the protection scope of the present application.

Claims

1. An intelligent production method based on a directed acyclic graph, characterized in that, Including: Determine the external dependencies between different production devices and the internal dependencies of the same production device; Construct a combined directed acyclic graph according to the external dependencies and the internal dependencies; wherein, each production device needs to execute at least one production task; the constructing of the combined directed acyclic graph according to the external dependencies and the internal dependencies includes: determining an internal task set and an internal dependency set according to the internal dependencies, defining the start task and the exit task in the internal task set as virtual tasks, and defining the intermediate tasks in the internal task set as real tasks; determining an external dependency set according to the external dependencies; constructing the combined directed acyclic graph according to the internal dependency set, the external dependency set and the internal task set; Construct a device model according to the device performance parameters of each production device and the server performance parameters of each edge server; Construct an intelligent factory model according to the combined directed acyclic graph, the device model and the task information of each production device; Taking the minimum cost as the goal, solve the intelligent factory model through a linear goal programming algorithm to obtain the optimal transmission power allocation ratio and the optimal offloading decision; Determine an optimal production plan according to the optimal transmission power allocation ratio and the optimal offloading decision, and perform intelligent production according to the optimal production plan.

2. The method according to claim 1, characterized in that, The constructing of the intelligent factory model according to the combined directed acyclic graph, the device model and the task information of each production device includes: Determine a task model according to the combined directed acyclic graph and the task information of each production device; Determine the transmission power allocation decision of the target production device according to the task model, determine the first transmission rate between production devices according to the device model, and determine the second transmission rate between the edge server and the production device; Construct a communication model according to the transmission power allocation decision, the first transmission rate and the second transmission rate; Determine the first computing power of the target production device, the energy consumption parameter and the second computing power of the edge server according to the device model; Construct a task execution model according to the first computing power, the energy consumption parameter and the second computing power; Determine the transmission power of the production device according to the device model, and construct an external dependency evaluation model according to the transmission power and the transmission power allocation decision; Integrate the device model, the task model, the communication model, the task execution model and the external dependency evaluation model to obtain the intelligent factory model.

3. The method according to claim 2, characterized in that, The determining of the task model according to the combined directed acyclic graph and the task information of each production device includes: Determine the previous tasks of each production task according to the combined directed acyclic graph to obtain a previous task set; Determine the offloading decision of each production task in the combined directed acyclic graph according to the task information; Determine the offloading destination of the production tasks in the combined directed acyclic graph in the edge server and the production device corresponding to the production task according to the offloading decision, and integrate the previous task set, the combined directed acyclic graph and the offloading destination of each production task to obtain the task model.

4. The method according to claim 2, wherein Constructing a communication model based on the transmission power allocation decision, the first transmission rate, and the second transmission rate includes: Determining the task dependency relationship between the target task and the target previous task of the target task according to the task model; In response to the task dependency relationship being an internal dependency relationship, determining the edge transmission delay according to the first transmission rate and the amount of transmission data of the target task, and determining the edge transmission delay as the task transmission delay; In response to the task dependency relationship being an external dependency relationship, determining the edge transmission delay according to the first transmission rate and the amount of transmission data of the target task, determining the external transmission delay according to the second transmission rate and the amount of transmission data of the target task, and determining the sum of the external transmission delay and the edge transmission delay as the task transmission delay; Determining the task reception delay according to the amount of return data of the target task and the second transmission rate; Determining the communication delay according to the task reception delay and the task transmission delay; Determining the transmission energy consumption of the target task according to the transmission power allocation decision, the transmission power of the target production device, the reception power of the target production device, and the amount of transmission data; Determining the reception energy consumption according to the amount of return data, the second transmission rate, and the reception power of the target production device; Determining the communication energy consumption according to the reception energy consumption and the transmission energy consumption; Integrating the communication delay and the communication energy consumption to obtain the communication model.

5. The method according to claim 2, wherein The energy consumption parameters include the working power and the standby power; constructing a task execution model according to the first computing power, the energy consumption parameters, and the second computing power includes: Determining the task completion time of the target previous task of the target task according to the task model, determining the task transmission delay between the target task and the target previous task according to the communication model, and determining the sum of the task completion time and the task transmission delay as the calculation start time; Determining the maximum calculation start time among the calculation start times corresponding to different previous tasks as the task start time of the target task; Determining the target offloading destination of the target task according to the task model; In response to the target offloading destination being the target production device corresponding to the target task, determining the internal execution delay according to the task execution calculation amount of the target task and the first computing power, determining the product of the internal execution delay and the working power as the internal calculation energy consumption, determining the internal execution delay as the task execution delay, and determining the internal calculation energy consumption as the task execution energy consumption; In response to the target offloading destination being the edge server, determining the external execution delay according to the task execution calculation amount of the target task and the second computing power, determining the product of the external execution delay and the standby power as the internal standby energy consumption, determining the sum of the external execution delay and the task reception delay as the task execution delay, and determining the sum of the internal standby energy consumption and the reception energy consumption as the task execution energy consumption; Determining the target completion time of the target task according to the task start time and the task execution delay; In response to the target task being the final production task, determine the target completion time as the equipment production completion time of the target production equipment, and determine the maximum value among the equipment production completion times corresponding to different production equipment as the total production time; Determine the single-equipment energy consumption of the target production equipment according to the task execution energy consumption and transmission energy consumption, and determine the sum of the single-equipment energy consumptions of all production equipment as the total energy consumption; Integrate the total energy consumption and the total production time to obtain the task execution model.

6. The method according to claim 1, characterized in that With the goal of minimizing cost, solve the intelligent factory model through the linear goal programming algorithm to obtain the optimal transmit power allocation ratio and the optimal offloading decision, including: With the goal of minimizing cost, construct a cost objective function according to the total production time, total energy consumption of the intelligent factory model, and the total external dependence evaluation; Determine the task offloading constraints according to the intelligent factory model and the number of servers of the edge server; Determine the energy constraints according to the intelligent factory model and the available energy of the production equipment; Determine the time constraints according to the intelligent factory model and the preset maximum running time; Determine the transmit power allocation ratio range constraints according to the preset ratio range; Solve the minimum value of the cost objective function through the linear goal programming algorithm under the constraint set to obtain the optimal transmit power allocation ratio and the optimal offloading decision; where the constraint set includes the task offloading constraints, the energy constraints, the time constraints, and the preset transmit power allocation ratio range constraints.

7. An intelligent production device based on a directed acyclic graph, characterized in that, Including: A dependency determination module configured to determine the external dependencies between different production equipment and the internal dependencies of the same production equipment; A combined directed acyclic graph module configured to construct a combined directed acyclic graph according to the external dependencies and the internal dependencies; where each production equipment needs to execute at least one production task; constructing the combined directed acyclic graph according to the external dependencies and the internal dependencies includes: determining an internal task set and an internal dependency set according to the internal dependencies, defining the start task and the exit task in the internal task set as virtual tasks, defining the intermediate tasks in the internal task set as real tasks; determining an external dependency set according to the external dependencies; constructing the combined directed acyclic graph according to the internal dependency set, the external dependency set, and the internal task set; An equipment model construction module configured to construct an equipment model according to the equipment performance parameters of each production equipment and the server performance parameters of each edge server; A factory model construction module configured to construct an intelligent factory model according to the combined directed acyclic graph and the task information of each production equipment; A model optimal solution module configured to solve the intelligent factory model through the linear goal programming algorithm with the goal of minimizing cost to obtain the optimal transmit power allocation ratio and the optimal offloading decision; A production plan generation module configured to determine an optimal production plan according to the optimal transmit power allocation ratio and the optimal offloading decision, and perform intelligent production according to the optimal production plan.

8. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and running on the processor, characterized in that, When the processor executes the program, the method according to any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause a computer to execute the method according to any one of claims 1 to 6.

Images