Intelligent factory management method and system based on Internet of Things and industrial big data

Abstract

The present invention provides a smart factory management method and system based on the Internet of Things and industrial big data, which relates to the technical field of the Internet of Things, including obtaining production information of the smart factory based on sensor equipment deployed in the smart factory; determining the production task sequence according to the task information, and constructing a first objective function in combination with the task start time and task deadline of each production task; constructing a second objective function according to raw material costs, labor costs and equipment depreciation costs, combined with pre-acquired historical production efficiency; transmitting the production information to a pre-constructed edge computing platform, jointly solving the first objective function and the second objective function through a multi-objective optimization model pre-deployed in the edge computing platform, and determining a production task optimization scheduling strategy.

Description

Technical Field

[0001] The present invention relates to Internet of Things technology, and in particular to an intelligent factory management method and system based on the Internet of Things and industrial big data. Background Art

[0002] Smart factories use various modern technologies to realize the automation of factory office, management and production, so as to strengthen and standardize enterprise management, reduce work errors, plug various loopholes, improve work efficiency, ensure safe production, provide decision-making references, strengthen external contacts and expand international markets.

[0003] CN201710627875.8, a smart factory system based on Internet of Things technology and industrial big data, including a lower-level SCADA process control system, a middle-level MES manufacturing execution system, and an upper-level ERP business planning system. The present invention solves the problems of difficulty in real-time analysis and calculation, difficulty in accident tracing and restoration, high frequency of unplanned parking, low worker production efficiency, difficulty in factory opaque decision-making, and difficulty in collaboration among personnel in various regions, thereby increasing product output, reducing energy consumption, and reducing costs, truly establishing a humanized smart factory with intelligent equipment, modern management, and information computerization, and improving the core competitiveness of enterprises.

[0004] CN202310483903.9, an intelligent factory management system based on the Internet of Things and industrial big data, relates to the field of equipment management technology, and solves the technical problem of not analyzing whether the equipment is normal and whether the equipment needs to be replaced based on the operating status of the equipment. Compared with the method of small to large, the input parameters of the secondary processing will appear too chaotic, and the output parameters recorded by the chaotic input parameters may be different from the output parameters corresponding to the original input parameters, so there is an energy conversion ratio obtained by the second group of tests; the input parameters belonging to the same energy conversion ratio are constructed into a year-on-year parameter interval, and then a parameter chaos test is performed based on the year-on-year parameter interval. It does not adopt a low-to-high input method, but a random input method, and then an energy conversion ratio test is performed. The tested energy conversion ratio has a wider coverage and its numerical value is more comprehensive, which makes the analysis more accurate.

[0005] In the existing technology, factory management often only focuses on one-sided needs, but in actual applications, the management of smart factories involves at least two aspects and needs to be considered comprehensively. Summary of the invention

[0006] The embodiments of the present invention provide a smart factory management method and system based on the Internet of Things and industrial big data, which can at least solve some problems in the prior art.

[0007] According to a first aspect of the embodiments of the present invention,

[0008] Provided is a smart factory management method based on the Internet of Things and industrial big data, including:

[0009] Acquiring production information of the smart factory based on sensor equipment deployed in the smart factory, wherein the production information includes task information and resource information, wherein the task information includes at least one of a task start time, a task priority, and a task deadline, and the resource information includes a production equipment capacity;

[0010] Determine the production task sequence according to the task information, and construct a first objective function with the goal of minimizing production time in combination with the task start time and task deadline of each production task, and set a first constraint condition corresponding to the first objective function; determine the raw material cost corresponding to the production task, the labor cost corresponding to the labor required to complete the production task, and the equipment depreciation cost corresponding to the production equipment required to complete the production task according to the resource information, and construct a second objective function with the goal of minimizing production cost in combination with the historical production efficiency obtained in advance, and set a second constraint condition corresponding to the second objective function;

[0011] The production information is transmitted to a pre-built edge computing platform, and the first objective function and the second objective function are jointly solved by a multi-objective optimization model pre-deployed in the edge computing platform to determine an optimal scheduling strategy for production tasks, wherein the multi-objective optimization model is constructed based on an improved genetic algorithm.

[0012] In an optional embodiment,

[0013] The method further includes constructing an edge computing platform, the edge computing platform comprising:

[0014] A device management module, used to assign a unique identity identifier to the sensor device registered on the edge computing platform, and assign device permissions matching the identity identifier based on the identity identifier, and after the identity and permissions of the sensor device are verified, transmit the production information collected by the sensor device to the application running module;

[0015] An application running module, used to receive deployment instructions from a cloud server through a pre-built container running environment, and control an application in the container running environment according to the deployment instructions, and upload the processing results of the application to the cloud server, wherein the application is used to determine production scheduling information according to the production information transmitted by the device management module in combination with a pre-deployed edge processing model;

[0016] The model deployment module is used to deploy the edge processing model trained on the cloud server on the edge computing platform after conversion and optimization.

[0017] In an optional embodiment,

[0018] Constructing a first objective function with the goal of minimizing production time and setting a first constraint condition corresponding to the first objective function includes:

[0019] Acquire task information and resource information of the smart factory, wherein the task information includes at least one of a task start time, a task priority, and a task deadline, and the resource information includes a production equipment capacity;

[0020] Determine the production task sequence according to the task priority in the task information, determine the corresponding predecessor task list for any production task, and determine the task start time and task deadline of each production task in the predecessor task list;

[0021] Introducing the urgency weight corresponding to the production task into the task priority, and combining the task start time and task deadline of each production task in the predecessor task list to construct the first objective function;

[0022] A first constraint condition corresponding to the first objective function is set, wherein the first constraint condition includes at least one of the following: the number of tasks of the production task does not exceed the capacity of the production equipment, the completion time of the production task is less than the task deadline, and the production tasks with higher priority are completed before the production tasks with lower priority.

[0023] In an optional embodiment,

[0024] The first objective function includes:

[0025]

[0026] Among them, Wtime i Indicates production time, ST i Indicates the production start time, PT i Indicates process time, proiority i Indicates priority, urgency i represents the urgency weight, and Cpre represents the task completion time of each production task in the predecessor task list of production project i.

[0027] In an optional embodiment,

[0028] Constructing a second objective function with the goal of minimizing production cost and setting a second constraint condition corresponding to the second objective function includes:

[0029] Determine the raw material costs corresponding to the production tasks of the smart factory, the labor costs corresponding to the labor required to complete the production tasks, and the equipment depreciation costs corresponding to the production equipment required to complete the production tasks;

[0030] Combining the previously acquired historical production efficiency of the production equipment and the time difference between the actual completion time of the production task and the target deadline, and assigning a corresponding penalty coefficient to the time difference, a second objective function with the goal of minimizing the production cost is constructed;

[0031] A second constraint condition corresponding to the second objective function is set, wherein the second constraint condition includes at least one of the following: the number of production task switching times is less than a preset switching threshold, the production equipment must not be overloaded, and the number of idle production equipment must not be less than a preset idle threshold.

[0032] In an optional embodiment,

[0033] The second objective function includes:

[0034]

[0035] Among them, C represents the production cost, n represents the number of production items, M i Indicates the cost of raw materials, L i Indicates labor cost, D i represents equipment depreciation expense, E i represents the historical production efficiency, P i represents the penalty term, T i represents the time difference, and i represents the production item.

[0036] In an optional embodiment,

[0037] The first objective function and the second objective function are jointly solved by the multi-objective optimization model pre-deployed in the edge computing platform, and the production task optimization scheduling strategy is determined to include:

[0038] Randomly initialize a population, wherein the initialized population includes a plurality of particles;

[0039] Determine an initial fitness value corresponding to each particle according to a preset Pareto constraint factor and in combination with the first objective function and the second objective function;

[0040] Dynamically setting a crossover rate and a mutation rate based on the initial fitness value, performing crossover and mutation operations on particles in the initialized population based on the crossover rate and the mutation rate, taking the fitness values ​​of the particles after the crossover and mutation operations as non-dominated solutions, performing Pareto sorting, and determining the crowding degree of each particle;

[0041] The congestion of particles is calculated iteratively in sequence, and each time the particle with the highest congestion replaces the particle with the lowest congestion until all particles are traversed and the particle update is completed. The particle with the highest congestion among the updated particles is used as the target particle to determine the optimal scheduling strategy for production tasks.

[0042] According to a second aspect of the embodiments of the present invention,

[0043] Provides an intelligent factory management system based on the Internet of Things and industrial big data, including:

[0044] A first unit is used to obtain production information of the smart factory based on sensor equipment deployed in the smart factory, wherein the production information includes task information and resource information, wherein the task information includes at least one of a task start time, a task priority, and a task deadline, and the resource information includes a production equipment capacity;

[0045] The second unit is used to determine the production task sequence according to the task information, and in combination with the task start time and task deadline of each production task, construct a first objective function with the goal of minimizing the production time and set a first constraint condition corresponding to the first objective function; determine the raw material cost corresponding to the production task, the labor cost corresponding to the labor required to complete the production task, and the equipment depreciation cost corresponding to the production equipment required to complete the production task according to the resource information, and in combination with the historical production efficiency obtained in advance, construct a second objective function with the goal of minimizing the production cost and set a second constraint condition corresponding to the second objective function;

[0046] The third unit is used to transmit the production information to a pre-built edge computing platform, and jointly solve the first objective function and the second objective function through a multi-objective optimization model pre-deployed in the edge computing platform to determine the production task optimization scheduling strategy, wherein the multi-objective optimization model is constructed based on an improved genetic algorithm.

[0047] According to a third aspect of the embodiments of the present invention,

[0048] An electronic device is provided, comprising:

[0049] processor;

[0050] a memory for storing processor-executable instructions;

[0051] The processor is configured to call the instructions stored in the memory to execute the aforementioned method.

[0052] A fourth aspect of the embodiments of the present invention is:

[0053] A computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the aforementioned method is implemented.

[0054] The present invention jointly solves the first objective function and the second objective function through a multi-objective optimization model to determine the optimal scheduling strategy for production tasks. By considering production time and cost at the same time, it allows to find the best trade-off between multiple objectives, which not only helps to reduce the total time of the task, but also helps to reduce production costs, thereby improving the overall performance of the smart factory. By transmitting production information to the edge computing platform and using the pre-deployed multi-objective optimization model, it is possible to optimize task scheduling at the edge computing level, reduce dependence on cloud computing resources, and improve real-time and responsiveness. In summary, the present invention improves production efficiency and better decision-making capabilities by comprehensively considering multiple objectives, resource constraints and historical data, and applying edge computing technology, which helps to improve the competitiveness and sustainability of the manufacturing industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a flow chart of a smart factory management method based on the Internet of Things and industrial big data according to an embodiment of the present invention;

[0056] Figure 2 This is a schematic diagram of the structure of an intelligent factory management system based on the Internet of Things and industrial big data according to an embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0059] Figure 1 FIG. 1 is a flow chart of a smart factory management method based on the Internet of Things and industrial big data according to an embodiment of the present invention. Figure 1 As shown, the method includes:

[0060] S101. Acquire production information of the smart factory based on sensor equipment deployed in the smart factory, wherein the production information includes task information and resource information, wherein the task information includes at least one of a task start time, a task priority, and a task deadline, and the resource information includes a production equipment capacity;

[0061] Exemplarily, task information may include: Task Start Time (TaskStartTime): the specific time point when the task starts to be executed, usually expressed in the form of date and time. For example, a production task on a machine starts to be executed at 8:00 AM on a certain day; Task Priority (TaskPriority): the urgency or importance of the task, usually expressed in numbers or levels. Tasks with higher priority are given priority in production scheduling. For example, a task for an urgent order may have a higher priority (for example, a priority of 1), while a regular order may have a lower priority (for example, a priority of 3); Task Deadline (TaskDeadline): the deadline by which a task must be completed. If the task is not completed before the deadline, it may cause production interruption or affect customer delivery. For example, the deadline for an order is 5:00 PM on a certain day.

[0062] Resource information may include production equipment capacity (ProductionEquipmentCapacity): the maximum number of tasks that the production equipment can handle or the output that can be processed per unit time, usually expressed as the number of products that can be produced or the number of workpieces processed per unit time. For example, a machine can produce 100 products per hour.

[0063] Sensor equipment can not only monitor the output and operating status of production equipment (resource information), but also record the start time, deadline and priority of each order (task information); this information can help smart factories perform production scheduling and task allocation to ensure that urgent orders are given priority before the deadline while maximizing the use of production equipment capacity.

[0064] S102. Determine the production task sequence according to the task information, and in combination with the task start time and task deadline of each production task, construct a first objective function with the goal of minimizing the production time and set a first constraint condition corresponding to the first objective function; determine the raw material cost corresponding to the production task, the labor cost corresponding to the labor required to complete the production task, and the equipment depreciation cost corresponding to the production equipment required to complete the production task according to the resource information, and in combination with the historical production efficiency obtained in advance, construct a second objective function with the goal of minimizing the production cost and set a second constraint condition corresponding to the second objective function;

[0065] The historical production efficiency refers to the relationship between the production results and input resources achieved in the past period of time, in order to measure the efficiency of its production activities. Historical production efficiency can be measured by different indicators and measurement methods, and is usually related to factors such as productivity, production costs, resource utilization and economic growth.

[0066] In an optional embodiment,

[0067] Constructing a first objective function with the goal of minimizing production time and setting a first constraint condition corresponding to the first objective function includes:

[0068] Acquire task information and resource information of the smart factory, wherein the task information includes at least one of a task start time, a task priority, and a task deadline, and the resource information includes a production equipment capacity;

[0069] Determine the production task sequence according to the task priority in the task information, determine the corresponding predecessor task list for any production task, and determine the task start time and task deadline of each production task in the predecessor task list;

[0070] Introducing the urgency weight corresponding to the production task into the task priority, and combining the task start time and task deadline of each production task in the predecessor task list to construct the first objective function;

[0071] A first constraint condition corresponding to the first objective function is set, wherein the first constraint condition includes at least one of the following: the number of tasks of the production task does not exceed the capacity of the production equipment, the completion time of the production task is less than the task deadline, and the production tasks with higher priority are completed before the production tasks with lower priority.

[0072] First, collect all relevant task information, including the task name, description, process, task duration, task priority and possible predecessor task relationship, and collect resource information, specifically the capacity of production equipment and the number of production equipment;

[0073] Sort the tasks according to their priorities. Generally, the higher the priority of a task, the earlier it is executed. For each task, check its possible predecessor tasks. Based on the dependencies between tasks, for each task, check its possible predecessor tasks and create a list of predecessor tasks. If there are no predecessor tasks, you can start from the planned start time (for example, the start of the production day) and calculate the deadline based on the duration of the task. If there are predecessor tasks, the start time of the task will depend on the completion time of the predecessor task. The deadline of the task will be calculated based on the start time and duration of the task. Check whether there are conflicts in the task time calculation results. If there are multiple tasks that need to be executed in the same time period or the task time conflicts with other constraints, reorder the tasks or adjust the time to avoid conflicts.

[0074] Collect information on all production tasks, including task name, task duration, task priority, and predecessor task relationship, introduce emergency weight or urgency for each task, calculate the comprehensive priority of each task based on the task priority and emergency weight, and construct the first objective function based on the emergency weight, priority, start time, and deadline of the task.

[0075] In this embodiment, production tasks are sorted according to their priorities, which helps ensure that high-priority tasks are given priority in the plan. By introducing emergency weights, the importance of tasks can be better reflected, ensuring that more urgent tasks are given higher priority. For each task, determining its predecessor task list based on the predecessor task relationship helps ensure that tasks are executed in the correct order. In summary, by combining task information, resource information and priority, this embodiment enables smart factories to plan and manage production tasks more effectively.

[0076] In an optional embodiment,

[0077] The first objective function includes:

[0078]

[0079] Among them, Wtime i Indicates production time, ST i Indicates the production start time, PT i Indicates process time, proiority i Indicates priority, urgency i represents the urgency weight, and Cpre represents the task completion time of each production task in the predecessor task list of production project i.

[0080] In an optional embodiment,

[0081] Constructing a second objective function with the goal of minimizing production cost and setting a second constraint condition corresponding to the second objective function includes:

[0082] Determine the raw material costs corresponding to the production tasks of the smart factory, the labor costs corresponding to the labor required to complete the production tasks, and the equipment depreciation costs corresponding to the production equipment required to complete the production tasks;

[0083] Combining the previously acquired historical production efficiency of the production equipment and the time difference between the actual completion time of the production task and the target deadline, and assigning a corresponding penalty coefficient to the time difference, a second objective function with the goal of minimizing the production cost is constructed;

[0084] A second constraint condition corresponding to the second objective function is set, wherein the second constraint condition includes at least one of the following: the number of production task switching times is less than a preset switching threshold, the production equipment must not be overloaded, and the number of idle production equipment must not be less than a preset idle threshold.

[0085] The penalty coefficient is a technique for introducing constraints into optimization problems so as to comprehensively consider objectives and constraints. Appropriate adjustment of the penalty coefficient can balance objectives and constraints according to the nature of the problem and optimization requirements.

[0086] In an optional embodiment,

[0087] The second objective function includes:

[0088]

[0089] Among them, C represents the production cost, n represents the number of production items, M i Indicates the cost of raw materials, L i Indicates labor cost, D i represents equipment depreciation expense, E i represents the historical production efficiency, P i represents the penalty term, T i represents the time difference, and i represents the production item.

[0090] S103. Transmit the production information to a pre-built edge computing platform, and jointly solve the first objective function and the second objective function through a multi-objective optimization model pre-deployed in the edge computing platform to determine the production task optimization scheduling strategy, wherein the multi-objective optimization model is constructed based on an improved genetic algorithm.

[0091] In an optional embodiment,

[0092] The method further includes constructing an edge computing platform, the edge computing platform comprising:

[0093] A device management module, used to assign a unique identity identifier to the sensor device registered on the edge computing platform, and assign device permissions matching the identity identifier based on the identity identifier, and after the identity and permissions of the sensor device are verified, transmit the production information collected by the sensor device to the application running module;

[0094] An application running module, used to receive deployment instructions from a cloud server through a pre-built container running environment, and control an application in the container running environment according to the deployment instructions, and upload the processing results of the application to the cloud server, wherein the application is used to determine production scheduling information according to the production information transmitted by the device management module in combination with a pre-deployed edge processing model;

[0095] The model deployment module is used to deploy the edge processing model trained on the cloud server on the edge computing platform after conversion and optimization.

[0096] For example,

[0097] The device management module can be EdgeIoTCore: EdgeIoTCore ensures that only legitimate devices can communicate with EdgeIoTEdgeRuntime by providing device authentication and access control. It is also responsible for device registration and configuration management to ensure that edge devices can be correctly identified and managed.

[0098] For example, in EdgeIoTCore, devices are assigned a unique identifier (usually a device ID), which is used for device authentication. When communicating with EdgeIoTCore, the device uses the corresponding certificate or token for authentication to ensure the legitimacy of both parties in communication. Edge IoTCore is responsible for managing the registration information of the device. It maintains a device registry that records the unique identifier, authentication information, authorization information, etc. of each device. When a device joins the network, its registration information will be added to the registry.

[0099] EdgeIoTCore is responsible for the device registration process. When a new device joins the network, it needs to provide relevant identity information, EdgeIoTCore will verify the legitimacy of the information, and then assign a unique identifier to the device. The registered device can communicate securely with EdgeIoTCore. EdgeIoT Core allows remote configuration of devices. The configuration information of the device, such as connection parameters, upgrade policies, data transmission protocols, etc., can be managed by EdgeIoTCore. In this way, the configuration of the device can be dynamically adjusted as needed without manual intervention.

[0100] EdgeIoTCore implements strict access control policies to ensure that only authorized devices and users can access specific resources. This access control is usually based on role-based permission allocation to ensure that only users and devices with specific roles can perform specific operations. EdgeIoTCore can monitor the status of devices, including online status, connection status, active status, etc. It can identify inactive or abnormal devices and trigger corresponding alarms or actions to solve potential problems in a timely manner.

[0101] The application runtime module can provide a lightweight container runtime environment for EdgeIoTEdgeRuntime: EdgeIoTEdgeRuntime allows containerized applications to run on edge devices. These applications can be containerized applications for tasks such as data analysis, data preprocessing, and machine learning inference. It is also responsible for communicating with EdgeIoTCore to receive commands and configuration information from the cloud.

[0102] EdgeIoTEdgeRuntime uses container technology (such as Docker) to encapsulate and isolate applications and their dependencies. In this way, applications can run in an independent container without being affected by other applications on the edge device. In order to adapt to the resource constraints of edge devices, EdgeIoTEdge Runtime provides a lightweight container runtime environment, which is lighter than traditional container runtimes and occupies less memory and computing resources.

[0103] Through EdgeIoTCore, cloud users can remotely deploy applications to edge devices. After receiving the deployment instructions from the cloud, EdgeIoTEdgeRuntime will download, install and run the corresponding containerized application. EdgeIoTEdgeRuntime is responsible for monitoring the life cycle of the application. It can start, stop, and restart containerized applications, and can detect application crashes and take appropriate measures to ensure that the application is always available.

[0104] EdgeIoTEdgeRuntime receives commands and configuration information from the cloud through a secure communication channel with EdgeIoTCore. These commands and configurations can include operations such as starting, stopping, and adjusting parameters of the application, as well as data information that the application needs to process. If the application needs to upload processing results or other real-time data to the cloud, EdgeIoTEdgeRuntime is also responsible for establishing a secure communication channel and transmitting the data to EdgeIoTCore for further cloud processing and analysis.

[0105] In addition, EdgeIoTEdgeRuntime ensures that the running container application is secure and takes measures to prevent container escape and attack. It can also limit the resource usage of the container to prevent the abnormal behavior of a certain application from affecting other applications or systems. The communication between EdgeIoTEdgeRuntime and EdgeIoTCore is encrypted, and an authentication mechanism is used to ensure the two-way security of communication. Data is encrypted during transmission to ensure data privacy and integrity. EdgeIoT EdgeRuntime records the logs and running status of the application. This information can be used to monitor the operation of the application, detect problems in time and take measures. In addition, it can also send monitoring data to the cloud for centralized monitoring and analysis.

[0106] The model deployment module can be EdgeIoTEdgeML: EdgeIoTEdgeML is responsible for deploying machine learning models to edge devices and performing inference locally. It can accept model deployment commands from Edge IoTCore, load the trained TensorFlow model or AutoMLTables model to the device, and perform inference on the input data locally to generate output results.

[0107] In an optional embodiment,

[0108] The first objective function and the second objective function are jointly solved by the multi-objective optimization model pre-deployed in the edge computing platform, and the production task optimization scheduling strategy is determined to include:

[0109] Randomly initialize a population, wherein the initialized population includes a plurality of particles;

[0110] Determine the initial attribute corresponding to each particle according to a preset Pareto constraint factor and in combination with the first objective function and the second objective function;

[0111] Based on the initial attributes, combined with the preset Pareto constraint conditions, determine whether each particle is a non-dominated particle, sort the particles in the population by non-dominated order, and divide them into different Pareto fronts. For each particle, dynamically update the particle attributes according to the attribute update equation to obtain the offspring particles;

[0112] According to the offspring particles, in combination with the Pareto front, the offspring particles are traversed, the dominated particles in the Pareto front are removed, the non-dominated particles are retained, and the retained non-dominated particles are recorded as candidate particles, the candidate particles are traversed, and for each candidate particle, the individual optimal position and the global optimal position are determined, and the corresponding weights are assigned to the individual optimal position and the global optimal position in combination with the weight allocation algorithm, and the particle score is determined, and the particle with the highest particle score value is selected as the target particle to determine the production task optimization scheduling strategy.

[0113] Exemplarily, the Pareto constraint factor is a method of adjusting the fitness value of a particle to balance multiple objective functions, which can help guide the particle in a more ideal direction while considering the trade-offs between objectives. The Pareto front represents a set of optimal solutions that can no longer improve one objective without compromising other objectives when there is a trade-off relationship between multiple objectives. Solution A dominates solution B if, for all optimization objectives, solution A is not inferior to solution B in at least one objective and is better in at least one objective. This means that solution A is better than solution B in at least some objectives and not worse than solution B in other objectives. Solution A is non-dominated if no other solution dominates solution A. This means that solution A is either better than other solutions or at least as good as other solutions in all optimization objectives. The initial attributes may be the position and velocity of the particle.

[0114] Set the population size, then randomly generate a group of particles, and determine the position and speed of each particle according to the preset Pareto constraint factor.

[0115] For each particle, use its position to calculate the value of the first objective function and the second objective function. For each particle, compare its objective function value with the Pareto constraint. If the particle is not dominated by other particles, that is, no other particle is better under multiple objectives, then it is marked as a non-dominated particle. Perform non-dominated sorting on all non-dominated particles and sort them according to the number of non-dominated particles and the number of dominated particles. This will form different levels of Pareto fronts, where the first level is the optimal non-dominated solution. According to the results of non-dominated sorting, the particles are divided into different Pareto fronts. For each particle, according to the historical behavior, neighbor information and Pareto front of its corresponding particle, use a multi-objective optimization algorithm to update its speed and position. Repeat the above steps until the maximum number of iterations or the preset stop condition is reached, and finally the offspring particles are obtained.

[0116] Get the generated offspring particles, traverse the solutions in the Pareto front, for each offspring particle, check whether there is a solution in the Pareto front that dominates it, if there is a dominating relationship, remove it from consideration, and retain the undominated offspring particles as candidate particles, for each candidate particle, calculate the individual optimal position and the global optimal position according to its historical performance and current attributes, use the weight allocation algorithm to assign weights to the individual optimal position and the global optimal position of each candidate particle, for each candidate particle, calculate the score according to its individual optimal position, global optimal position and weight allocation, and select the particle with the highest score from the candidate particles as the target particle. This target particle represents the best production task optimization scheduling strategy.

[0117] For example, if there is a particle whose individual optimal position is 5 and whose global optimal position is 6, and whose corresponding weights are 0.4 and 0.6, then the score of the particle is 5.6.

[0118] In this embodiment, by jointly solving the first objective function and the second objective function, the multi-objective optimization model can comprehensively consider multiple objectives such as task completion time and production cost, which helps to achieve the best trade-off between multiple objectives. By identifying non-dominated particles, the system can find the best solution that is not dominated by other solutions, which helps to determine the best strategy among alternative solutions, and divide the particles into different Pareto front levels, which allows the system to generate a series of possible optimization strategies, each with different performance under multiple objectives. In summary, this embodiment uses a multi-objective optimization algorithm and Pareto front analysis, combining multiple optimization objectives and constraints to find the best production task scheduling strategy, thereby improving the production efficiency and resource utilization of smart factories.

[0119] Figure 2 FIG. 1 is a schematic diagram of the structure of an intelligent factory management system based on the Internet of Things and industrial big data according to an embodiment of the present invention. Figure 2 As shown, the system comprises:

[0120] A first unit is used to obtain production information of the smart factory based on sensor equipment deployed in the smart factory, wherein the production information includes task information and resource information, wherein the task information includes at least one of a task start time, a task priority, and a task deadline, and the resource information includes a production equipment capacity;

[0121] The second unit is used to determine the production task sequence according to the task information, and in combination with the task start time and task deadline of each production task, construct a first objective function with the goal of minimizing the production time and set a first constraint condition corresponding to the first objective function; determine the raw material cost corresponding to the production task, the labor cost corresponding to the labor required to complete the production task, and the equipment depreciation cost corresponding to the production equipment required to complete the production task according to the resource information, and in combination with the historical production efficiency obtained in advance, construct a second objective function with the goal of minimizing the production cost and set a second constraint condition corresponding to the second objective function;

[0122] The third unit is used to transmit the production information to a pre-built edge computing platform, and jointly solve the first objective function and the second objective function through a multi-objective optimization model pre-deployed in the edge computing platform to determine the production task optimization scheduling strategy, wherein the multi-objective optimization model is constructed based on an improved genetic algorithm.

[0123] According to a third aspect of the embodiments of the present invention,

[0124] An electronic device is provided, comprising:

[0125] processor;

[0126] a memory for storing processor-executable instructions;

[0127] The processor is configured to call the instructions stored in the memory to execute the aforementioned method.

[0128] A fourth aspect of the embodiments of the present invention is:

[0129] A computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the aforementioned method is implemented.

[0130] The present invention may be a method, an apparatus, a system and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present invention.

[0131] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A smart factory management method based on the Internet of Things and industrial big data, characterized in that: include: Acquiring production information of the smart factory based on sensor equipment deployed in the smart factory, wherein the production information includes task information and resource information, wherein the task information includes at least one of a task start time, a task priority, and a task deadline, and the resource information includes a production equipment capacity; Determine the production task sequence according to the task information, and construct a first objective function with the goal of minimizing production time in combination with the task start time and task deadline of each production task, and set a first constraint condition corresponding to the first objective function; determine the raw material cost corresponding to the production task, the labor cost corresponding to the labor required to complete the production task, and the equipment depreciation cost corresponding to the production equipment required to complete the production task according to the resource information, and construct a second objective function with the goal of minimizing production cost in combination with the historical production efficiency obtained in advance, and set a second constraint condition corresponding to the second objective function; The production information is transmitted to a pre-built edge computing platform, and the first objective function and the second objective function are jointly solved by a multi-objective optimization model pre-deployed in the edge computing platform to determine an optimal scheduling strategy for production tasks, wherein the multi-objective optimization model is constructed based on an improved genetic algorithm; Constructing a first objective function with the goal of minimizing production time and setting a first constraint condition corresponding to the first objective function includes: Acquire task information and resource information of the smart factory, wherein the task information includes at least one of a task start time, a task priority, and a task deadline, and the resource information includes a production equipment capacity; Determine the production task sequence according to the task priority in the task information, determine the corresponding predecessor task list for any production task, and determine the task start time and task deadline of each production task in the predecessor task list; Introducing the urgency weight corresponding to the production task into the task priority, and combining the task start time and task deadline of each production task in the predecessor task list to construct the first objective function; Setting a first constraint condition corresponding to the first objective function, wherein the first constraint condition includes at least one of the following: the number of tasks of the production task does not exceed the capacity of the production equipment, the completion time of the production task is less than the task deadline, and the production task with a high priority is completed before the production task with a low priority; Constructing a second objective function with the goal of minimizing production cost and setting a second constraint condition corresponding to the second objective function includes: Determine the raw material costs corresponding to the production tasks of the smart factory, the labor costs corresponding to the labor required to complete the production tasks, and the equipment depreciation costs corresponding to the production equipment required to complete the production tasks; Combining the previously acquired historical production efficiency of the production equipment and the time difference between the actual completion time of the production task and the target deadline, and assigning a corresponding penalty coefficient to the time difference, a second objective function with the goal of minimizing the production cost is constructed; Setting a second constraint condition corresponding to the second objective function, wherein the second constraint condition includes at least one of the following: the number of production task switching times is less than a preset switching threshold, the production equipment is not overloaded, and the number of idle production equipment is not less than a preset idle threshold; The method further includes constructing an edge computing platform, the edge computing platform comprising: A device management module, used to assign a unique identity identifier to the sensor device registered on the edge computing platform, and assign device permissions matching the identity identifier based on the identity identifier, and after the identity and permissions of the sensor device are verified, transmit the production information collected by the sensor device to the application running module; An application running module, used to receive deployment instructions from a cloud server through a pre-built container running environment, and control an application in the container running environment according to the deployment instructions, and upload the processing results of the application to the cloud server, wherein the application is used to determine production scheduling information according to the production information transmitted by the device management module in combination with a pre-deployed edge processing model; The model deployment module is used to deploy the edge processing model trained on the cloud server on the edge computing platform after conversion and optimization; The first objective function includes: Among them, Wtime i Indicates the production time corresponding to the i-th production item, ST i Indicates the production start time, PT i Indicates process time, proiority i Indicates priority, urgency i represents the urgency weight, Cpre represents the task completion time of each production task in the predecessor task list of production project i; The second objective function includes: Among them, C represents the production cost, n represents the number of production items, M i Indicates the cost of raw materials, L i Indicates labor cost, D i represents equipment depreciation expense, E i represents the historical production efficiency, P i represents the penalty term, T i represents the time difference, and i represents the production item.

2. The method according to claim 1, characterized in that The first objective function and the second objective function are jointly solved by the multi-objective optimization model pre-deployed in the edge computing platform, and the production task optimization scheduling strategy is determined to include: Randomly initialize a population, wherein the initialized population includes a plurality of particles; Determine the initial attribute corresponding to each particle according to a preset Pareto constraint factor and in combination with the first objective function and the second objective function; Based on the initial attributes, combined with the preset Pareto constraint conditions, determine whether each particle is a non-dominated particle, sort the particles in the population by non-dominated order, and divide them into different Pareto fronts. For each particle, dynamically update the particle attributes according to the attribute update equation to obtain the offspring particles; According to the offspring particles, in combination with the Pareto front, the offspring particles are traversed, the dominated particles in the Pareto front are removed, the non-dominated particles are retained, and the retained non-dominated particles are recorded as candidate particles, the candidate particles are traversed, and for each candidate particle, the individual optimal position and the global optimal position are determined, and the corresponding weights are assigned to the individual optimal position and the global optimal position in combination with the weight allocation algorithm, and the particle score is determined, and the particle with the highest particle score value is selected as the target particle to determine the production task optimization scheduling strategy.

3. An intelligent factory management system based on the Internet of Things and industrial big data, characterized in that: include: A first unit is used to obtain production information of the smart factory based on sensor equipment deployed in the smart factory, wherein the production information includes task information and resource information, wherein the task information includes at least one of a task start time, a task priority, and a task deadline, and the resource information includes a production equipment capacity; The second unit is used to determine the production task sequence according to the task information, and in combination with the task start time and task deadline of each production task, construct a first objective function with the goal of minimizing the production time and set a first constraint condition corresponding to the first objective function; determine the raw material cost corresponding to the production task, the labor cost corresponding to the labor required to complete the production task, and the equipment depreciation cost corresponding to the production equipment required to complete the production task according to the resource information, and in combination with the historical production efficiency obtained in advance, construct a second objective function with the goal of minimizing the production cost and set a second constraint condition corresponding to the second objective function; A third unit is used to transmit the production information to a pre-built edge computing platform, and jointly solve the first objective function and the second objective function through a multi-objective optimization model pre-deployed in the edge computing platform to determine an optimization scheduling strategy for production tasks, wherein the multi-objective optimization model is constructed based on an improved genetic algorithm; Constructing a first objective function with the goal of minimizing production time and setting a first constraint condition corresponding to the first objective function includes: Acquire task information and resource information of the smart factory, wherein the task information includes at least one of a task start time, a task priority, and a task deadline, and the resource information includes a production equipment capacity; Determine the production task sequence according to the task priority in the task information, determine the corresponding predecessor task list for any production task, and determine the task start time and task deadline of each production task in the predecessor task list; Introducing the urgency weight corresponding to the production task into the task priority, and combining the task start time and task deadline of each production task in the predecessor task list to construct the first objective function; Setting a first constraint condition corresponding to the first objective function, wherein the first constraint condition includes at least one of the following: the number of tasks of the production task does not exceed the capacity of the production equipment, the completion time of the production task is less than the task deadline, and the production task with a high priority is completed before the production task with a low priority; Constructing a second objective function with the goal of minimizing production cost and setting a second constraint condition corresponding to the second objective function includes: Determine the raw material costs corresponding to the production tasks of the smart factory, the labor costs corresponding to the labor required to complete the production tasks, and the equipment depreciation costs corresponding to the production equipment required to complete the production tasks; Combining the previously acquired historical production efficiency of the production equipment and the time difference between the actual completion time of the production task and the target deadline, and assigning a corresponding penalty coefficient to the time difference, a second objective function with the goal of minimizing the production cost is constructed; Setting a second constraint condition corresponding to the second objective function, wherein the second constraint condition includes at least one of the following: the number of production task switching times is less than a preset switching threshold, the production equipment is not overloaded, and the number of idle production equipment is not less than a preset idle threshold; The system also includes building an edge computing platform, and the edge computing platform includes: A device management module, used to assign a unique identity identifier to the sensor device registered on the edge computing platform, and assign device permissions matching the identity identifier based on the identity identifier, and after the identity and permissions of the sensor device are verified, transmit the production information collected by the sensor device to the application running module; An application running module, used to receive deployment instructions from a cloud server through a pre-built container running environment, and control an application in the container running environment according to the deployment instructions, and upload the processing results of the application to the cloud server, wherein the application is used to determine production scheduling information according to the production information transmitted by the device management module in combination with a pre-deployed edge processing model; The model deployment module is used to deploy the edge processing model trained on the cloud server on the edge computing platform after conversion and optimization; The first objective function includes: Among them, Wtime i Indicates the production time corresponding to the i-th production item, ST i Indicates the production start time, PT i Indicates process time, proiority i Indicates priority, urgency i represents the urgency weight, Cpre represents the task completion time of each production task in the predecessor task list of production project i; The second objective function includes: Among them, C represents the production cost, n represents the number of production items, M i Indicates the cost of raw materials, L i Indicates labor cost, D i represents equipment depreciation expense, E i represents the historical production efficiency, P i represents the penalty term, T i represents the time difference, and i represents the production item.

4. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the method described in any one of claims 1 to 2.

5. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 2 is implemented.

Images