Product production method and equipment based on intelligent agent and computer program product

Abstract

The invention is suitable for the technical field of computers, and provides an intelligent agent-based product production method, computer equipment and a computer program product.The intelligent agent-based product production method comprises the steps of collecting working state data of production equipment on a specified production line in real time based on a production task; respectively sending the working state data to each intelligent agent associated with the specified production line, wherein the intelligent agents represent specific equipment or processes on the specified production line; based on an operation result of a specified operation executed by each intelligent agent based on the working state data, detecting an output product of each production process to obtain a detection result; and if the detection result represents that the output product of each production process is a qualified product, determining the output product of the last production process as a target product required by the to-be-executed order. The product production efficiency of the manufacturing industry can be improved.

Description

Technical Field

[0001] This application belongs to the field of computer technology, and particularly relates to a product production method, device, and computer program product based on an agent. Background Art

[0002] The application of agent platforms in the manufacturing industry is extensive and profound. From production process optimization to quality control, from predictive maintenance to supply chain collaboration, agents have demonstrated their powerful potential and value.

[0003] Taking production process optimization as an example, traditional production lines often have problems such as long waiting times and low equipment utilization. In terms of quality control, the agent platform also plays an important role. Traditional quality control methods often rely on manual inspection, but this method is error-prone and inefficient. Summary of the Invention

[0004] In view of this, embodiments of this application provide a product production method, device, and computer program product based on an agent, which can improve the production efficiency of products in the manufacturing industry.

[0005] The first aspect of the embodiments of this application provides a product production method based on an agent, including: Based on the production task, real-time collect the working status data of the production equipment on the specified production line; Send the working status data to each agent associated with the specified production line, where the agent represents a specific device or process on the specified production line; Based on the operation results of the specified operations performed by each agent based on the working status data, detect the output products of each production process to obtain a detection result; If the detection result indicates that the output products of each production process are qualified products, determine the output product of the last production process as the target product required for the order to be executed.

[0006] In one implementation of the first aspect, the real-time collection of the working status data of the production equipment on the specified production line based on the production task includes: Based on the production task, select a specified sensor adapted to the specified production line from the sensor network, where the specified sensor is used to monitor the working status of the production equipment; Based on a preset data integration method, integrate the data of each specified sensor to obtain the working status data of the production equipment on the specified production line.

[0007] In one implementation of the first aspect, the data integration method includes at least one of the following: time series alignment method, data source merging method, multi-table integration method, cross-data integration.

[0008] In an implementation of the first aspect, the method further includes: obtaining preset constraint conditions for each production process; determining a designated production line for each production task based on the order to be executed, the current available resource situation, and the preset constraint conditions.

[0009] In an implementation of the first aspect, the method further includes: if the order to be executed includes multiple different models of products, then based on the differences between the different models, perform production planning to determine the target production method for each model of product.

[0010] In an implementation of the first aspect, the agent associated with the production equipment further includes a fault detection agent, and the method further includes: Predicting the fault phenomena that will occur to the targeted production equipment based on each of the fault detection agents; Determining a target maintenance method for repairing the fault phenomena based on each of the fault phenomena; Performing predictive maintenance based on the target maintenance method to obtain a maintenance result.

[0011] In an implementation of the first aspect, the predicting the fault phenomena that will occur to the targeted production equipment based on each of the fault detection agents includes: Collecting at least one image of the key parts of each production equipment on the designated production line through an image acquisition device; Performing image recognition on the image based on the image recognition model integrated in each of the fault detection agents to obtain a recognition result; Predicting the fault phenomena that will occur to the targeted production equipment based on the recognition result.

[0012] In an implementation of the first aspect, the method further includes: In the case of an emergency during the production process, performing priority sorting on the production tasks based on the degree to which the production tasks are affected to obtain a sorting result; Allocating the remaining resources on the designated production line based on the sorting result.

[0013] A second aspect of the embodiments of the present application provides an agent-based product production device, including: An acquisition module, configured to collect working state data of production equipment on a designated production line in real time based on production tasks; A sending module, configured to send the working state data to each agent associated with the designated production line respectively, where the agent represents a specific device or process on the designated production line; A production module, configured to detect the output products of each production process based on the operation results of the specified operations performed by each agent based on the working status data, so as to obtain detection results; A determination module, configured to, if the detection results indicate that the output products of each production process are qualified products, determine the output product of the last production process as the target product required for the to-be-executed order.

[0014] A third aspect of the embodiments of the present application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the computer device implements the method as described in the first aspect above.

[0015] A fourth aspect of the embodiments of the present application provides a computer program product, characterized by including a computer program, and when the computer program is run, the method as described in the first aspect above is executed.

[0016] A fifth aspect of the embodiments of the present application provides a computer-readable storage medium, storing a computer program, and when the computer program is executed by a processor, the method as described in the first aspect above is implemented.

[0017] In the first aspect of the embodiments of the present application, the product production method based on agents realizes precise monitoring and optimal control of the production process by collecting the working status data of production equipment in real time and sending this data to agents representing specific equipment or processes. Each agent performs specified operations according to the received working status data to ensure that the equipment operates in the best state, thereby improving production efficiency and product quality. By analyzing the detection results of the output products of each production process, potential problems can be discovered and corrected in a timely manner to ensure the qualification of the final product. This method not only improves the automation level and flexibility of production, but also reduces errors and delays caused by human intervention, ensures that the output product of the last production process meets the requirements of the to-be-executed order, and ultimately improves the overall production efficiency.

[0018] It can be understood that the beneficial effects of the second to fifth aspects above can refer to the relevant descriptions in the first aspect above, and will not be elaborated here. Description of the Drawings

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained according to these drawings.

[0020] Figure 1 is the implementation flowchart of the agent-based product production method provided by the embodiments of the present application; Figure 2 is the implementation flowchart of the method for obtaining working state data provided by the embodiments of the present application; Figure 3 is the implementation flowchart of the method for determining the target execution path provided by the embodiments of the present application; Figure 4 is the implementation flowchart of the predictive maintenance method provided by the embodiments of the present application; Figure 5 is the implementation flowchart of the method for predicting fault phenomena provided by the embodiments of the present application; Figure 6 is a schematic diagram of the agent-based product production device provided by the embodiments of the present application; Figure 7 is a schematic diagram of the computer device provided by the embodiments of the present application. Detailed implementation manners

[0021] In the following description, for the purpose of illustration rather than limitation, specific details such as specific system architectures and technologies are proposed to thoroughly understand the embodiments of the present application. However, those skilled in the art should clearly understand that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present application.

[0022] It should be understood that when used in the specification and appended claims of the present application, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0023] It should also be understood that the term "and / or" used in the specification and appended claims of the present application refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0024] As used in the specification and appended claims of the present application, the term "if" can be interpreted as "when...", "once", "in response to determining", or "in response to detecting" according to the context. Similarly, the phrase "if determined" or "if [the described condition or event] is detected" can be interpreted as meaning "once determined", "in response to determining", "once [the described condition or event] is detected", or "in response to detecting [the described condition or event]" according to the context.

[0025] In addition, in the description of the specification and the appended claims of the present application, the terms "first", "second", "third", etc. are only used for distinguishing descriptions and cannot be construed as indicating or implying relative importance.

[0026] The reference to "one embodiment" or "some embodiments" etc. described in the specification of the present application means that a specific feature, structure or characteristic described in connection with the embodiment is included in one or more embodiments of the present application. Thus, the statements "in one embodiment", "in some embodiments", "in other some embodiments", "in still other some embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "comprising", "including", "having" and their variants all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0027] In one embodiment, the agent-based product production system includes: a multi-agent system (MAS), where each agent represents a specific device or process on a production line; a sensor network for deploying various types of sensors such as temperature, humidity, vibration, etc.; an Internet of Things (IoT) platform for connecting all sensors to ensure that data can be collected, transmitted and stored in real time; and multiple edge computing nodes arranged near the data source to preliminarily process the raw data and reduce unnecessary bandwidth occupancy.

[0028] In one embodiment, as Figure 1 shown, a method for agent-based product production is provided. Taking the example that this method is applied to a computer device, it includes the following steps S101 to step S104: Step S101, based on the production task, collect the working state data of the production equipment on the specified production line in real time.

[0029] In applications, to enable agents to accurately perceive the environment and make decisions, manufacturing enterprises have established a complete data collection system. This data collection system consists of three key parts: a sensor network, an Internet of Things (IoT) platform, and edge computing nodes. Among them, the sensor network: A large number of high-precision sensors are deployed on the production line to monitor the working status of production equipment. These sensors can collect various physical parameters such as temperature, pressure, and vibration frequency in real time and transmit the information to subsequent processing links. For example, in the process of automobile manufacturing, the force sensors installed on the stamping machine can accurately measure the force changes generated by each operation, helping maintenance personnel detect potential faults in advance; while the vision sensors on the assembly line can identify whether the parts are correctly installed in place to ensure product quality. The Internet of Things (IoT) platform: As the central nervous system connecting all sensors, the IoT platform not only realizes the interconnection and interoperability of devices but also provides powerful data management functions. It supports multiple communication protocols (such as MQTT, CoAP, etc.) to ensure that data from different sources can be smoothly aggregated. More importantly, the IoT platform has powerful data analysis capabilities and can perform preliminary cleaning and transformation on the raw data through built-in algorithms to extract valuable information for further use. For example, by analyzing historical operation records, it can predict when a certain machine needs maintenance or component replacement to avoid unplanned downtime. Edge computing nodes: Considering the network latency problems that may be brought about by large-scale data transmission, enterprises have set up multiple edge computing nodes near the data source. These nodes can complete some simple preprocessing tasks locally, such as anomaly detection, trend analysis, etc., reducing the upload of unnecessary redundant information to the cloud and effectively reducing the bandwidth occupancy rate. At the same time, the edge computing nodes can also act as temporary storage media to save important data when the network is unstable and upload it later when conditions permit.

[0030] In applications, the computer device receives a production instruction for any specified production task and real-time collects the working status data of all production equipment on the specified production line corresponding to the production task through sensors. The production equipment associated with each production line can be preset according to actual needs. Multiple production equipment for any production line have a sequential relationship in the production process.

[0031] Step S102, send the working status data to each agent associated with the specified production line, where the agent represents a specific device or process on the specified production line.

[0032] In an application, production equipment on each production line used to perform production tasks and the processes of the production equipment can be managed by one or more agents. The computer device distributes the collected working status data to each agent associated with the specified production line. The data integration and processing of the working status data can be performed by each received agent, such as filtering and deduplication, or can be integrated by the computer device and then sent to each receiving agent.

[0033] Exemplarily, taking the agent-based production manufacturing scenario as the product manufacturing scenario in the automotive industry. In this scenario, due to the strict requirements of the automotive industry for real-time performance and reliability, the enterprise selects an agent platform with powerful computing capabilities and good scalability, such as JADE (Java Agent Development Framework) or Jason, etc. Such platforms not only support multi-threaded concurrent processing but also provide rich API interfaces and services, facilitating seamless docking with other business systems (such as ERP, MES). Based on the selected agent framework, the enterprise constructs a multi-agent system (MAS). Each agent in the MAS represents a specific device or process on a production line. For example: The parts supply agent: responsible for managing the raw material warehouse and timely providing the required parts to each workstation according to the production plan. It needs to continuously monitor the inventory level, predict future material requirements, and initiate replenishment requests in a timely manner. The welding robot agent: specifically processes the connection between the body structure parts. It adjusts its position and posture according to the received task instructions to ensure that the quality of the weld seam meets the standards. At the same time, this agent also needs to synchronize with other adjacent processes to avoid affecting the overall progress due to excessive waiting time. The painting workshop agent: controls the working parameters of the painting equipment, including color mixing, spraying thickness, and drying temperature, etc.

[0034] These agents communicate with each other through a message passing mechanism and follow predefined behavior rules to cooperate to complete complex production tasks.

[0035] Step S103, based on the operation results of the specified operations performed by each of the agents based on the working status data, detect the output products of each of the production processes to obtain detection results.

[0036] In an application, each production process has clearly defined detection standards and acceptance criteria set in advance. For example, in the welding process, standard ranges for parameters such as weld width and thickness can be set. Each agent performs the specified operations corresponding to the production equipment or process based on the adapted working state data, obtains the operation results, and can feedback the operation results through sensors or other means. The computer device acquires the operation results and can detect the output products of each production process based on the operation results of each agent to obtain the detection results. The detection results are used to determine whether the products meet the standards. If the products are unqualified, the specific defect types and locations are recorded. The products output by non - the last production process are intermediate products, i.e., semi - finished products. The computer device needs to perform product detection on the intermediate products output by each production process to determine whether the intermediate products of each production process meet the conditions of the preset qualified products.

[0037] Exemplarily, assume that a certain automobile manufacturer is assembling a new type of electric vehicle. Then the (multiple agents) MAS on its production line may operate as follows: After receiving the latest production instructions, the component supply agent quickly checks whether the current inventory is sufficient. If it is not sufficient, it immediately issues a purchase order to the supplier and tracks the logistics status until the goods are in stock. After receiving the task, the welding robot agent starts to perform precise welding operations at the specified positions. During this period, it closely monitors the data fed back by the sensors to ensure that each welding reaches the expected effect. If there is any deviation, it immediately stops the operation and sends a warning signal to the adjacent process. The painting workshop agent takes over after the body frame is assembled, carefully formulates the paint formula and applies it evenly. After completion, it also carefully checks the coating uniformity and flatness and performs secondary repairs if necessary until it is completely qualified.

[0038] Step S104, if the detection results indicate that the output products of each production process are qualified products, then determine the output product of the last production process as the target product required by the to - be - executed order.

[0039] In an application, a product quality traceability system is established to record the production process and detection results of each product. Ensure that once a product is judged to be qualified, its information can be completely saved and traced. After the last production process is completed, a final quality inspection is carried out to ensure that the product fully meets the order requirements.

[0040] In this embodiment, the agent-based product production method realizes precise monitoring and optimized control of the production process by collecting the working state data of production equipment in real time and sending this data to agents representing specific equipment or processes. Each agent performs specified operations based on the received working state data to ensure that the equipment operates in the best state, thereby improving production efficiency and product quality. By analyzing the detection results of the products output by each production process, potential problems can be discovered and corrected in a timely manner to ensure the qualification of the final products. This method not only improves the automation level and flexibility of production, but also reduces errors and delays caused by human intervention, ensures that the products output by the last production process meet the requirements of the orders to be executed, and ultimately improves the overall production efficiency and customer satisfaction. In addition, this method can further enhance the adaptive ability and intelligent level of the system through continuous data accumulation and technology optimization, creating more value for the enterprise.

[0041] In one embodiment, as Figure 2 shown, the implementation process of step S101 includes the following steps S201 to step S202: Step S201, based on the production task, select a specified sensor adapted to the specified production line from the sensor network, and the specified sensor is used to monitor the working state of the production equipment.

[0042] In application, for a specified production line, the computer device selects one or more specified sensors adapted to the specified production line from the sensor network, and each production equipment or process on the specified production line can be controlled by one or more agents.

[0043] Step S202, based on a preset data integration method, integrate the data of each of the specified sensors to obtain the working state data of the production equipment on the specified production line.

[0044] In application, the computer device integrates the acquired sensor data based on a preset data integration method to obtain high-quality working state data.

[0045] In one embodiment, the data integration method includes at least one of the following: time series alignment method, data source merging method, multi-table integration method, cross-data integration.

[0046] In application, according to the characteristics of the data itself and the specific task requirements, perform one or more of the following data integration operations on the data: time series alignment method, data source merging method, multi-table integration method, cross-data integration.

[0047] Among them, time series alignment: Integrate different data to ensure consistent timestamps and the same time granularity (such as summarizing by month, quarter, year). Data source merging: According to different task requirements, select key merging (merging through common fields in the data) or vertical merging (merging data from different tables). Multi-table integration: Associate different related data (such as sales, production, inventory data, etc.) to form a comprehensive management data set. Cross-data integration: If data is stored in different databases, ETL tools or writing scripts can be used to achieve cross-data integration.

[0048] In this embodiment, through various data integration methods, high-quality data can be provided to ensure the accuracy of subsequent operations.

[0049] In one embodiment, as Figure 3 shown, the computer device can also perform the following steps S301 to step S303 to determine the designated production line for each production task: Step S301, obtain the preset constraint conditions for each production process.

[0050] In the application, each production task includes various production processes, and each production process is associated with various production parameters. These production parameters include but are not limited to: environmental parameters: such as temperature, humidity, etc., to ensure that the production environment meets the process requirements; operating parameters of production equipment: for example, the speed, pressure, power, etc. of machine tools to ensure that the equipment operates efficiently within a safe range; product parameters: related to raw material specifications, processing accuracy, etc., to ensure that the product quality meets the standards, etc. According to the actual business logic, several constraint conditions are set to limit the possibility of variable values. These constraint conditions can be divided into the following categories: resource allocation constraints, time-related constraints, quality control constraints, etc. For example, resource allocation constraints at least include the following: 1) The same piece of equipment cannot serve two different processes simultaneously. 2) Certain special processes must be performed by certified operators. 3) Specific processes require the use of specific types of tools or materials. Time-related constraints at least include the following: 1) Sufficient time must be reserved for quality inspection and packaging before the finished product delivery date. 2) The production line changeover time should be as short as possible to reduce downtime. Quality control constraints at least include the following: 1) The output of each process must meet a certain quality standard to enter the next process. 2) Key processes may require additional quality inspection steps.

[0051] In the application, the above-defined production processes and the corresponding preset constraint conditions are pre-stored in the corresponding storage space (such as a database, etc.), and when the computer device executes a production task, it reads the preset constraint conditions for each production process from the storage space.

[0052] Step S302: Based on the orders to be executed, the current available resource situation, and the preset constraint conditions, determine the designated production line for each production task.

[0053] In the application, the computer device parses the orders to be executed to obtain the corresponding order information, including but not limited to: basic information such as order quantity, specifications, delivery date, etc., and special requirements such as customization needs, priorities, etc. Collect and analyze the available resource status on the current production line, mainly including: Equipment status: Which equipment is idle, and which is under repair or maintenance. Human resources: Which operators have the corresponding qualifications to perform specific processes. Material inventory: Whether the quantity of existing raw materials is sufficient to complete the order, and whether there are enough spare parts.

[0054] In the application, based on the above information, an appropriate algorithm is used to determine the best execution path for each production task, that is, the designated production line that matches the actual production requirements of each task. Commonly used algorithms include but are not limited to: Heuristic search algorithm: Minimum Remaining Value (MRV) heuristic: Prioritize tasks with the fewest alternative solutions to expose potential conflicts earlier. Degree heuristic: Tend to schedule activities with more associations with other unassigned tasks first to reduce the number of subsequent branches. Constraint Satisfaction Problem (CSP) solver: Use advanced CSP solvers (such as MiniZinc, Choco Solver, etc.). These tools provide a variety of built-in optimization techniques that can significantly improve the solving speed and ensure finding the global optimal solution or an approximate optimal solution.

[0055] Exemplarily, assume that a factory receives an order for a new product and is required to produce 500 pieces within two weeks. The production of this product involves three main processes: cutting, welding, and assembly. Parse the order: Confirm the order quantity, delivery date, and special requirements. Analyze the resources: Cutting machine A is currently idle, but welding robot B is performing other tasks and is expected to be available in three days. There are two qualified operators in the assembly workshop, but one of them will take a vacation in a week. Apply the constraint conditions: Welding must be carried out immediately after the cutting process, so we need to wait for welding robot B to be idle. The assembly process needs to be completed before the operator takes a vacation. Determine the best (target) path: First, arrange the cutting process using cutting machine A. When welding robot B becomes idle, start the welding process immediately. When the welding process is nearing completion, arrange the assembly process to ensure that all assembly work is completed before the operator takes a vacation.

[0056] In this embodiment, production tasks can be effectively planned to ensure that orders are completed on time and with high quality, while maximizing the utilization of existing resources and reducing unnecessary downtime and costs.

[0057] In one embodiment, if the order to be executed includes multiple products of different models, production planning is carried out based on the differences between the different models to determine the target production method for each model of product.

[0058] In application, in view of the fact that the models of products to be produced in the order to be executed may be different, the computer device determines the differences between different signals and can carry out production planning to determine the production method for each model of product. The production method refers to the production strategy for producing products of a specified model. Among them, the way of carrying out production planning can be realized through a genetic algorithm.

[0059] Exemplarily, taking the production of automobiles as an example, due to the differences between different vehicle models, the existing production line needs to frequently switch between multiple product models, resulting in a reduction in production efficiency. For this reason, a genetic algorithm is introduced into the intelligent agent to carry out production planning. The genetic algorithm simulates the natural selection mechanism in the process of biological evolution and aims to find the global optimal solution, especially effective when facing complex scheduling problems with multiple objectives and multiple constraints. Population initialization: Randomly generate a set of initial production plans as the members of the first-generation population. Each plan represents a possible product combination method and its corresponding production order. Fitness evaluation: According to the established objective function (such as minimizing the total cost, maximizing the output, etc.), calculate the fitness scores of each plan. The higher the score, the closer the plan is to the ideal state. Selection operation: Select individuals with higher fitness into the next reproduction stage according to a certain proportion. Common methods include roulette wheel selection, tournament selection, etc., to ensure the inheritance of excellent genes. Crossover and mutation: Imitating the principle of DNA recombination, perform a crossover operation on the selected parent individuals, exchange some chromosome segments to generate new offspring; at the same time, appropriately introduce random mutations to increase the population diversity and prevent premature convergence to a local optimal solution. Iterative optimization: Repeat the above process until a predetermined termination condition (such as the maximum number of iterations, fitness threshold, etc.) is reached, and finally output the optimal production plan.

[0060] In this embodiment, this method is particularly suitable for solving complex scheduling problems with multiple objectives and multiple constraints.

[0061] In one embodiment, as Figure 4 shown, the implementation process of fault detection in the product production process includes the following steps S401 to step S403: Step S401, based on each of the fault detection intelligent agents, predict the fault phenomena that the targeted production equipment is about to occur.

[0062] In the application, high-precision image acquisition devices (such as high-definition cameras, industrial cameras) and sensors (such as temperature, pressure, vibration sensors, etc.) are used to collect real-time data from key parts of each production device on the specified production line. The collected data is cleaned to remove noise and outliers, ensuring the consistency and accuracy of the data. The data from different sources is converted into a unified format for subsequent processing.

[0063] In the application, a pre-trained deep learning model (such as a convolutional neural network CNN) is used to analyze the collected images to identify changes in the states of key components. Key features that help determine the health status of the device are extracted from the image recognition results, such as surface cracks, part wear degrees, etc. At the same time, features are extracted from the sensor data, such as temperature fluctuations, vibration frequencies, etc. Based on historical data and the feature extraction results, a suitable machine learning or deep learning model (such as LSTM, random forest, support vector machine, etc.) is selected for fault prediction. The current feature data is input into the prediction model, and the probability and type of possible faults in a future period are output. This can help detect potential problems in advance and avoid production line downtime caused by sudden failures.

[0064] Step S402: Based on each of the fault phenomena, determine the target maintenance method for repairing the fault phenomenon.

[0065] In the application, a computer device performs fault classification and priority ranking on the fault phenomena. Fault classification means classifying the faults into different types according to the results output by the prediction model, such as mechanical faults, electrical faults, software faults, etc. Priority ranking means evaluating the severity and urgency of each fault to determine its repair priority. Usually, serious faults that directly affect production are given higher priorities. And by referring to the maintenance manuals provided by the equipment manufacturer or the internal maintenance guidelines of the enterprise, understanding the standard maintenance processes for different types of faults, and based on the fault type and severity, the most suitable maintenance method is selected. For example, minor wear may only require simple cleaning and lubrication, while serious mechanical faults require component replacement or major repairs.

[0066] Step S403: Based on the target maintenance method, perform predictive maintenance to obtain maintenance results.

[0067] In an application, resource scheduling and preparation are first carried out according to the target maintenance method, such as determining the required tools, materials, labor, and other resources, arranging relevant personnel and materials, and ensuring that maintenance tasks can be quickly responded to and executed when needed. And according to the production plan of the production line and the operating status of the equipment, a suitable time window is selected for maintenance operations to minimize the impact on normal production. Collect the operation data of the equipment after maintenance, analyze the maintenance effect, and feedback the results to the system. By continuously accumulating experience and optimizing the algorithm, the accuracy and efficiency of predictive maintenance are improved.

[0068] In this embodiment, not only can equipment failures be detected and prevented in advance, but also the losses caused by unexpected shutdowns can be significantly reduced, the service life of the equipment can be extended, and the overall production efficiency can be improved. In addition, continuous data collection and analysis contribute to the continuous improvement of maintenance strategies, making the operation of the enterprise more intelligent and efficient.

[0069] In one embodiment, as Figure 5 shown, the implementation process of step S401 includes the following steps S501 to step S503: Step S501, at least one image of the key parts of each production equipment on the specified production line is collected through an image acquisition device.

[0070] In an application, high-precision image acquisition devices (such as high-definition cameras, industrial cameras, etc.) need to be installed on the specified production line. These devices will be deployed at key parts, such as the core working area of the machine or places where wear and failures are likely to occur. The specific implementation includes: 1) Equipment selection: Select a suitable image acquisition device according to the characteristics of the production line to ensure that sufficient details can be captured for subsequent analysis. 2) Location determination: Cooperate with production engineers to determine which are the key monitoring points, which are usually areas with frequent failures or areas that have the greatest impact on production efficiency. 3) Timed / real-time acquisition: Set the time interval for image acquisition. Higher frequencies may be required for key components to ensure the real-time nature of the data. At the same time, ensure that the system can operate efficiently without affecting normal production. The computer device can collect at least one image of the key parts of each production equipment on the specified production line through the image acquisition device.

[0071] Step S502, based on the image recognition models integrated in each of the fault detection agents, perform image recognition on the image to obtain a recognition result.

[0072] In an application, each fault detection agent can integrate a specially trained image recognition model. This typically involves the application of deep learning techniques, such as convolutional neural networks (CNNs). For the image recognition model, during the data collection and annotation phase, a large number of device images covering different states (normal and abnormal) are collected and accurately annotated to provide a high-quality dataset. In terms of model selection and architecture design, a suitable deep learning framework (such as TensorFlow, PyTorch, etc.) is selected according to actual needs, and the corresponding network structure is designed. For specific types of faults, customized network layers and parameter adjustments may be required. During the model training phase, the labeled dataset is used to train the model, and the weights are continuously optimized through the backpropagation algorithm until the model reaches a satisfactory accuracy rate. The image recognition model is evaluated and verified, and the cross-validation method is used to evaluate the model performance to ensure its good generalization ability and avoid overfitting problems.

[0073] Step S503, based on the recognition result, predict the fault phenomenon that is about to occur in the production equipment targeted.

[0074] In an application, when the image acquisition is completed and preliminary results are obtained through the image recognition model, these results are used to predict the upcoming fault phenomenon. The specific implementation is as follows: 1) Feature extraction: Extract key features from the results of image recognition that are helpful for judging the health status of the equipment. For example, the length and width of surface cracks, the change in the gap between parts, etc. 2) Establish a fault model: Build a mathematical model or rule base based on historical fault data to describe the relationship between various fault precursors and their corresponding severity and development trends. 3) Application of prediction algorithms: Use machine learning algorithms (such as decision trees, support vector machines SVM, random forests, etc.) or more complex deep learning methods (such as LSTM for time series analysis), combined with the current feature values, to predict possible future problems and their probabilities. 4) Risk assessment and early warning: Calculate the risk level based on the prediction results and set thresholds to trigger the alarm mechanism. Once a potential fault is predicted and the risk exceeds the set threshold, relevant personnel are immediately notified to take preventive measures.

[0075] In an application, to ensure the effectiveness of the entire system, a real-time monitoring system can also be established to continuously track the status of the equipment and adjust the strategy according to the actual situation: 1) Automatic control: Embed the fault detection agent into the production line control system so that it can not only give early warnings but also automatically perform some basic maintenance operations, such as pausing the production line for inspection. 2) Continuous improvement: Regularly update the model and rule base, and continuously iterate and optimize the entire fault prediction system as new fault types appear and the performance of the existing model changes. 3) User interface development: Provide an intuitive and easy-to-use interface for managers to display monitoring information and alarm details, facilitating their quick response.

[0076] In this embodiment, the fault prediction function of production equipment based on the fault detection agent can be effectively realized, thereby improving production efficiency, reducing unexpected downtime, and lowering maintenance costs.

[0077] In one embodiment, the computer device can also perform the following operations: in the case of an emergency during the production process, based on the degree to which the production task is affected, prioritize the production tasks to obtain a sorting result; based on the sorting result, allocate the remaining resources on the specified production line.

[0078] In applications, when an emergency occurs, the intelligent agent will immediately activate the emergency response mechanism, re-evaluate the existing conditions, and quickly adjust the subsequent steps. Specific measures include, but are not limited to: prioritization: once an unexpected event (such as equipment failure, raw material shortage, etc.) occurs, the intelligent agent will quickly determine which tasks are most affected and mark them as high-priority. For such tasks, the system will prioritize the allocation of resources to support them and minimize the impact on the overall progress.

[0079] The embodiment of the present application also provides an intelligent agent-based product production device for performing the steps in the embodiment of the intelligent agent-based product production method described above. As Figure 6 shown, the intelligent agent-based product production device 600 provided by the embodiment of the present application includes: An acquisition module 610, configured to collect the working state data of the production equipment on the specified production line in real time based on the production task.

[0080] A sending module 620, configured to send the working state data to each intelligent agent associated with the specified production line, where the intelligent agent represents a specific device or process on the specified production line.

[0081] A production module 630, configured to detect the output products of each production process based on the operation results of the specified operations performed by each intelligent agent based on the working state data, to obtain a detection result.

[0082] A determination module 640, configured to, if the detection result indicates that the output products of each production process are qualified products, determine the output product of the last production process as the target product required for the order to be executed.

[0083] In one embodiment, the acquisition module is further configured to select a specified sensor adapted to the specified production line from the sensor network based on the production task, where the specified sensor is used to monitor the working state of the production equipment; based on a preset data integration method, integrate the data of each specified sensor to obtain the working state data of the production equipment on the specified production line.

[0084] In one embodiment, the data integration methods used in the acquisition module include at least one of the following: time series alignment method, data source merging method, multi-table integration method, cross-data integration.

[0085] In one embodiment, the production module is further configured to obtain preset constraint conditions for each production process; determine the designated production line for each production task based on the to-be-executed order, the current available resource situation, and the preset constraint conditions.

[0086] In one embodiment, the production module is further configured to, if the to-be-executed order includes multiple different models of products, perform production planning based on the differences between the different models, and determine the target production method for each model of product.

[0087] In one embodiment, the agent associated with the production equipment further includes a fault detection agent, and the production module is further configured to predict the upcoming fault phenomena of the production equipment targeted based on each of the fault detection agents; determine the target maintenance method for repairing the fault phenomena based on each of the fault phenomena; perform predictive maintenance based on the target maintenance method to obtain a maintenance result.

[0088] In one embodiment, the production module is further configured to collect at least one image of the key parts of each production equipment on the designated production line through an image acquisition device; perform image recognition on the image based on the image recognition model integrated in each of the fault detection agents to obtain a recognition result; predict the upcoming fault phenomena of the production equipment targeted based on the recognition result.

[0089] In one embodiment, the production module is further configured to, in the case of an emergency during the production process, rank the production tasks according to the degree of impact on the production tasks to obtain a ranking result; allocate the remaining resources on the designated production line based on the ranking result.

[0090] In an application, each module in the product production device based on an agent can be a software program module, can also be implemented by different logic circuits integrated in a processor, or can also be implemented by multiple distributed processors.

[0091] Figure 7 The figure is a schematic structural diagram of a computer device for implementing a product production method based on an agent provided in an embodiment of the present application. As Figure 7 shown, the computer device 700 in this embodiment includes: at least one processor 710 ( Figure 7(only one is shown), a memory 720, and a computer program 730 stored in the memory 720 and executable on the at least one processor 710. When the processor 710 executes the computer program 730, the steps in any of the above-described embodiments of the exception handling method are implemented.

[0092] The computer device 700 may be a computing device such as a desktop computer, a notebook, a palm computer, or a cloud server. The computer device 700 may include, but is not limited to, a processor 710 and a memory 720. Those skilled in the art can understand that Figure 7 This is only an example of the computer device 700 and does not limit the computer device 700. It may include more or fewer components than shown in the figure, or combine certain components, or different components. For example, it may also include input / output devices, network access devices, etc.

[0093] The so-called processor 710 may be a central processing unit (CPU), and the processor 710 may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0094] In some embodiments, the memory 720 may be an internal storage unit of the computer device 700, such as the hard disk or memory of the computer device 700. In other embodiments, the memory 720 may also be an external storage device of the computer device 700, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the computer device 700. Further, the memory 720 may also include both the internal storage unit and the external storage device of the computer device 700. The memory 720 is used to store an operating system, application programs, a boot loader, data, and other programs, such as the program code of the computer program. The memory 720 may also be used to temporarily store data that has been output or will be output.

[0095] It should be noted that, regarding the information interaction, execution process, etc. between the above-mentioned devices / units, since they are based on the same concept as the method embodiments of this application, for their specific functions and the technical effects brought about, reference can be made to the method embodiment section for details, and thus will not be elaborated here.

[0096] Those skilled in the art can clearly understand that, for the sake of convenience and brevity of description, only the above-mentioned division of each functional unit and module is used as an example. In actual applications, the above functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiments can be integrated into a processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction and do not limit the protection scope of this application. The specific working processes of the units and modules in the above system can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated here.

[0097] The embodiments of this application also provide a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.

[0098] The embodiments of this application provide a computer program product. When the computer program product runs on a mobile terminal, the mobile terminal can implement the steps in the above-mentioned method embodiments when executed.

[0099] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above method embodiments of this application, a computer program can be used to instruct relevant hardware to complete. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the device / terminal device, recording medium, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium. For example, USB flash drive, mobile hard disk, magnetic disk or optical disc, etc. In some jurisdictions, according to legislation and patent practice, the computer-readable medium cannot be electrical carrier signal and telecommunication signal.

[0100] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0101] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed in this document can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0102] In the embodiments provided in this application, it should be understood that the disclosed device / network device and method can be implemented in other ways. For example, the device / network device embodiments described above are merely illustrative. For example, the division of the modules or units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of the device or unit can be in an electrical, mechanical or other form.

[0103] The unit described as a separation component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, that is, it may be located in one place or may be distributed over multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0104] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included within the protection scope of the present application.

Claims

1. A product production method based on an intelligent agent, characterized in that: include: Based on production tasks, real-time collection of working status data of production equipment on designated production lines; Sending the work status data to each intelligent agent associated with the designated production line, wherein the intelligent agent represents a specific device or production process on the designated production line; Based on the operation results of the designated operations performed by each of the intelligent agents based on the work status data, the output products of each production process are tested to obtain the test results; If the detection result indicates that the output products of each of the production processes are qualified products, the output product of the last production process is determined as the target product required for the order to be executed.

2. The agent-based product production method according to claim 1, characterized in that: The real-time collection of working status data of production equipment on a specified production line based on production tasks includes: Based on the production task, a designated sensor adapted to the designated production line is selected from the sensor network, wherein the designated sensor is used to monitor the working status of the production equipment; Based on a preset data integration method, the data of each of the designated sensors is integrated to obtain the working status data of the production equipment on the designated production line.

3. The agent-based product production method according to claim 2, characterized in that: The data integration method includes at least one of the following: time series alignment method, data source merging method, multi-table integration method, and cross-data integration.

4. The agent-based product production method according to claim 1, characterized in that: The method further comprises: Get the preset constraints for each production process; The designated production line for each production task is determined based on the orders to be executed, the current available resources and the preset constraints.

5. The agent-based product production method according to claim 3, characterized in that: The method further comprises: If the to-be-executed order includes products of multiple different models, production planning is performed based on the differences between the different models to determine the target production method for each model of the product.

6. The agent-based product production method according to claim 1, characterized in that: The intelligent agent associated with the production equipment also includes a fault detection intelligent agent, and the method further includes: Based on each of the fault detection intelligent agents, predict the impending fault phenomenon of the targeted production equipment; Based on each of the fault phenomena, determining a target maintenance method for repairing the fault phenomenon; Based on the target maintenance method, predictive maintenance is performed to obtain a maintenance result.

7. The agent-based product production method according to claim 6, characterized in that: The predicting of impending failure phenomena of the production equipment based on each of the fault detection intelligent agents includes: By means of an image acquisition device, at least one image of a key part of each production equipment on the designated production line is acquired; Based on the image recognition model integrated by each of the fault detection agents, performing image recognition on the image to obtain a recognition result; Based on the identification result, an upcoming failure phenomenon of the targeted production equipment is predicted.

8. The agent-based product production method according to claim 1, characterized in that: The method further comprises: In the event of an emergency during the production process, the production tasks are prioritized based on the degree of impact on the production tasks to obtain the ranking results; Based on the sorting result, the remaining resources on the designated production line are allocated.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the computer device is caused to implement the method according to any one of claims 1 to 8.

10. A computer program product, characterized in that The invention comprises a computer program, which, when being executed, enables the method according to any one of claims 1 to 8 to be performed.

Images