Intelligent production scheduling method and device for equipment manufacturing industry

Abstract

The present invention discloses an intelligent production scheduling method and device for the equipment manufacturing industry, and relates to the technical field of production scheduling. The method includes: establishing an equipment archive information library to record the information of active equipment and standby equipment of the target production line; and constructing an equipment topology network according to the equipment archive information library and the process information of the production line. A lightweight prediction model is configured for each edge device node, and the equipment health status analysis and short-term maintenance demand prediction are performed in combination with real-time equipment operation data to obtain a preliminary evaluation status. The evaluation status is corrected through data sharing of edge device nodes to obtain the final evaluation result of the equipment. According to the evaluation results and the equipment archive library, equipment scheduling and maintenance decisions are generated to realize intelligent production scheduling. The technical problem of lack of flexibility and predictability in equipment maintenance and production scheduling in the prior art is solved, and the technical effect of improving equipment utilization, optimizing production scheduling and ensuring efficient operation of the production line is achieved.

Description

Technical Field

[0001] The present application relates to the technical field of production scheduling, and in particular to an intelligent production scheduling method and device for the equipment manufacturing industry. Background Art

[0002] In the equipment manufacturing industry, with the increasing complexity and diversity of production line equipment, the scheduling problem of production equipment has become increasingly prominent. How to reasonably schedule equipment during the production process has become a key link in ensuring production efficiency. Traditional equipment scheduling methods often rely on fixed plans and manual experience, lack of accurate assessment of real-time equipment status, resulting in the inability to effectively foresee equipment failures and maintenance needs, which in turn affects the continuity and efficiency of production. Faced with a complex production environment, there is an urgent need for an intelligent scheduling method that can evaluate equipment status in real time and predict maintenance needs in advance, so as to improve equipment utilization and reduce production interruptions.

[0003] At the current stage, relevant technologies have the technical problem of lack of flexibility and predictability in equipment maintenance and production scheduling. Summary of the invention

[0004] This application provides an intelligent production scheduling method and device for the equipment manufacturing industry. First, an equipment archive information library is established to record the information of the active equipment and spare equipment of the target production line, and each device has an independent code; then, according to the equipment archive information library and the process information of the production line, an equipment topology network is constructed, and the network contains multiple edge device nodes. A lightweight prediction model is configured for each edge device node, and the equipment health status analysis and short-term maintenance demand prediction are performed in combination with real-time equipment operation data to obtain a preliminary evaluation status. Through data sharing of edge device nodes, the evaluation status is corrected to obtain the final evaluation result of the equipment. Finally, based on the evaluation results and the equipment archive library, equipment scheduling and maintenance decisions are generated to realize intelligent production scheduling, achieving the technical effect of improving equipment utilization, optimizing production scheduling, and ensuring continuous and efficient operation of the production line.

[0005] This application provides an intelligent production scheduling method for equipment manufacturing industry, including:

[0006] An equipment archive information library is established, wherein the equipment archive information library contains the active equipment information and spare equipment information of the target production line, and each equipment has an independent code; based on the equipment archive information library, combined with the process information of the target production line, an equipment topology network is constructed, wherein the equipment topology network contains multiple edge device nodes; a lightweight prediction model is configured for the multiple edge device nodes, and primary equipment status assessment is performed in combination with real-time equipment operation data, including equipment health status analysis and short-term maintenance demand prediction, to obtain multiple primary equipment assessment states; through the sharing of associated equipment data of the multiple edge device nodes, the assessment status is corrected to obtain an edge device status assessment result; based on the edge device status assessment result, combined with the production equipment archive, equipment scheduling and maintenance decisions are generated to perform intelligent production scheduling.

[0007] The present application also provides an intelligent production scheduling device for equipment manufacturing industry, including:

[0008] An equipment archive information library establishment module, the equipment archive information library establishment module is used to establish an equipment archive information library, the equipment archive information library contains the active equipment information and spare equipment information of the target production line, and each equipment has an independent code; an equipment topology network construction module, the equipment topology network construction module is used to construct an equipment topology network based on the equipment archive information library and the process information of the target production line, and the equipment topology network includes multiple edge device nodes; a primary equipment status evaluation module, the primary equipment status evaluation module is used to configure a lightweight prediction model for the multiple edge device nodes, and perform primary equipment status evaluation in combination with real-time equipment operation data, including equipment health status analysis and short-term maintenance demand prediction, to obtain multiple primary equipment evaluation states; an evaluation status correction module, the evaluation status correction module is used to perform evaluation status correction through the sharing of associated equipment data of the multiple edge device nodes, and obtain edge device status evaluation results; an intelligent production scheduling module, the intelligent production scheduling module is used to generate equipment scheduling and maintenance decisions based on the edge device status evaluation results and in combination with the production equipment archive library for intelligent production scheduling.

[0009] The intelligent production scheduling method and device for the equipment manufacturing industry proposed in this application first establish an equipment archive information library to record the information of the active equipment and spare equipment of the target production line, and each device has an independent code; then, according to the equipment archive information library and the process information of the production line, a device topology network is constructed, and the network contains multiple edge device nodes. A lightweight prediction model is configured for each edge device node, and the real-time equipment operation data is combined to perform equipment health status analysis and short-term maintenance demand prediction to obtain a preliminary evaluation status. Through data sharing of edge device nodes, the evaluation status is corrected to obtain the final evaluation result of the equipment. Finally, based on the evaluation results and the equipment archive, equipment scheduling and maintenance decisions are generated to realize intelligent production scheduling, achieving the technical effect of improving equipment utilization, optimizing production scheduling, and ensuring continuous and efficient operation of the production line. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the device according to the embodiment of the present application. It should be understood that the previous or following operations are not necessarily performed precisely in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0011] Figure 1 A schematic diagram of a process flow of an intelligent production scheduling method for equipment manufacturing industry provided in an embodiment of the present application;

[0012] Figure 2 A schematic diagram of the structure of an intelligent production scheduling device for equipment manufacturing provided in an embodiment of the present application. Explanation of reference numerals: equipment archive information base establishment module 10, equipment topology network construction module 20, primary equipment status evaluation module 30, evaluation status correction module 40, intelligent production scheduling module 50. DETAILED DESCRIPTION

[0013] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0014] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0015] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, device, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0016] The present application embodiment provides an intelligent production scheduling method for equipment manufacturing industry, such as Figure 1 As shown, the method includes:

[0017] Step S100, establish an equipment archive information library, which contains the active equipment information and spare equipment information of the target production line, and each equipment has an independent code. Specifically, when establishing the equipment archive information library, first collect information on the active equipment of the target production line. Carry out detailed classification according to the equipment type, identify and record it in the production site; obtain model and function information by consulting nameplates, manuals, etc., and describe the processing technology, production capacity and functional positioning in the production line in detail; accurately calculate the service life and record influencing factors such as maintenance and modification; collect fault reports and maintenance work orders to sort out fault time, phenomenon, measures and post-maintenance status, and store them according to rules. For spare equipment, register information as required, including storage location, environmental conditions, activation conditions, etc., and conduct regular inspections and tests and record the results. Then formulate coding rules, such as using letters to represent production areas or production line categories, numbers to distinguish equipment types and sequential numbers, etc., attach coding labels to each device, establish a correspondence table between codes and equipment information in the information database, and set up automatic identification and retrieval functions to facilitate information entry, query, and update operations. When new equipment is added or information changes, related information should be updated in a timely manner to ensure that the information database is accurate and complete.

[0018] Step S200, according to the equipment archive information library, combined with the process information of the target production line, a device topology network is constructed, and the device topology network includes multiple edge device nodes. Specifically, firstly, the process information is obtained by studying the process flow chart of the target production line, each link is disassembled and analyzed, and the sub-process is divided into sub-processes, and the input and output material characteristics of each sub-process are clarified. Then, the key process equipment of each sub-process is determined, its technical parameters and other information are recorded, and the corresponding relationship between the equipment and the process is established. Then, the edge device node attributes are set according to factors such as the importance of the equipment in the process, and the node connection relationship is configured according to the process sequence and logical relationship, considering the data flow, collaborative work and control dependency. After that, the network hierarchy is determined according to the process characteristics and management requirements, which is divided into the equipment layer, control layer, management layer, etc., and the edge device nodes and their connection relationships are integrated to form a complete network, and optimization tests are performed during the integration process to check data transmission, evaluate reliability and fault tolerance, formulate backup and recovery strategies, and continuously adjust and improve the network according to production feedback to improve adaptability and operation efficiency.

[0019] In one possible implementation, based on the equipment archive information library and combined with the process information of the target production line, an equipment topology network is constructed, and the equipment topology network includes multiple edge device nodes. Step S200 further includes step S210, according to the process flow chart of the target production line, the process information of the target production line is obtained to divide the process into sections, and multiple section production processes are obtained. The section production process is an independent production process of a single component, and the section production process includes corresponding key process nodes and key process equipment. Specifically, study the design drawings and technical specifications of the product, and determine its key components and functional modules. For example, in the production of electronic products, the motherboard can be regarded as a key component, and the screen display function is a functional module. For mechanical products, the manufacture of the engine and the assembly of the transmission system can be regarded as key parts. Taking these as units, the entire production process can be split according to the production order of components or functions. For the machining sub-process, it is necessary to clarify the specific shape of the processed parts, the dimensional accuracy requirements and the required processing technology, such as cutting parameters, tool selection, etc. In the circuit assembly process, determine the type, layout and welding process specifications of electronic components. For the shell spraying process, it is necessary to clarify the parameters such as the type of paint, spraying thickness, color requirements, etc. Detailed operation instructions are formulated for each sub-process, and records are kept. Record operating steps, quality inspection standards and other information for accurate execution and subsequent traceability. In each sub-process, determine the key process nodes through detailed analysis of the production process. Nodes are often the key links that determine component quality or production efficiency. For example, in machining, the last finishing process of precision parts may be a key node; in circuit assembly, the chip welding link is usually crucial. For the determined key process nodes, clarify the corresponding key process equipment. The performance, stability and accuracy of key equipment directly affect product quality. Focus on the management of key equipment, establish special equipment files, record detailed information such as equipment operating parameters, maintenance history, fault records, etc., and formulate emergency plans for key equipment to respond quickly when equipment fails to reduce the impact on production.

[0020] Step S220, based on the multiple divisional production processes, configure the corresponding multiple edge device nodes, and based on the multiple edge device nodes, build a device topology network, the device topology network includes multiple levels, corresponding to different process priorities. Specifically, for each divisional production process, configure the edge device nodes according to the process characteristics and data collection and transmission requirements. In the machining process, configure sensor nodes for the processing equipment to collect the equipment's operating status data, such as speed, temperature, vibration and other information. In the circuit assembly process, set up detection device nodes, which can detect and collect data on the parameters of electronic components, welding quality, etc. Each edge device node has data processing and communication functions, can perform preliminary processing on the collected data, and transmit the data to the network according to the prescribed communication protocol. A unique identifier is assigned to each node to facilitate identification and management in the network. A multi-level network is constructed according to the priority of the process and the criticality of the node. For the nodes corresponding to the key process equipment, they are placed in the core layer of the network, and the bandwidth and speed of data transmission are prioritized to ensure real-time monitoring and data processing of key equipment. For example, in automobile production, the equipment nodes of the engine assembly process are at a higher level, and non-key nodes are distributed in the peripheral layer of the network. Data collection and transmission are carried out at certain time intervals, which are mainly used to monitor the overall status of the production process. In the network layer division, data routing rules are established to ensure that data can be transmitted accurately and efficiently between different layers. At the same time, a network security protection mechanism is set up to control access to nodes at different levels to prevent data leakage and malicious attacks, and ensure the stable operation of the equipment topology network and data security.

[0021] Step S300, configure a lightweight prediction model for the multiple edge device nodes, combine the real-time equipment operation data to perform primary equipment status assessment, including equipment health status analysis and short-term maintenance demand prediction, and obtain multiple primary equipment assessment states. Specifically, first extract the edge device node related data from the equipment archive information library, including basic equipment parameters, historical operation and maintenance records, etc., conduct in-depth analysis to find out the rules and characteristics, and use data mining technology to extract key feature values. Then select a suitable machine learning algorithm to build a lightweight prediction model according to the equipment characteristics and data characteristics, such as a decision tree algorithm or an ARIMA model, and perform lightweight processing. Then collect the key parameters, working conditions and environmental data of the equipment operation in real time, input them into the model, evaluate the equipment health status by comparing with the normal parameter threshold and analyzing the change trend, and judge the fault risk. Combined with the maintenance cycle, common faults and equipment status at the time of occurrence in the historical maintenance data, predict whether maintenance and maintenance type are needed in the future short term based on the current health trend and production plan. The primary equipment assessment status is obtained by combining the above equipment health status analysis and short-term maintenance demand prediction, which provides a basis for equipment management and production scheduling.

[0022] In one possible implementation, a lightweight prediction model is configured for the multiple edge device nodes, and primary equipment status evaluation is performed in combination with real-time equipment operation data, including equipment health status analysis and short-term maintenance demand prediction, to obtain multiple primary equipment evaluation states, and step S300 further includes step S310, extracting the first edge device node based on the multiple edge device nodes. Specifically, among the multiple edge device nodes, a starting node is determined as the research object, which is defined as the first edge device node, and the selection follows the following rules, such as according to the order of the equipment in the production process, the importance of the equipment, or random selection, etc., and the selection method is determined according to the actual production needs and research purposes. For example, if you want to start the analysis from key equipment, the edge nodes corresponding to the equipment in the core process or the equipment that has a greater impact on product quality are preferred.

[0023] Step S320, for the first edge device node, traverse the device archive information library to extract identity information and collect the first device archive data. Specifically, with the first edge device node as the target, conduct a comprehensive search in the device archive information library, use the device's independent code or other unique identification information as the search keyword, quickly and accurately locate all records related to the node, extract the device's identity information from the device archive information library, including the device's name, number, production line and other basic identification content, and then collect detailed first device archive data, such as the device's initial installation information, past operation logs, maintenance records, parts replacement lists, etc. The data provides basic information for subsequent analysis, and the data collection process must ensure completeness and accuracy to avoid missing key information.

[0024] Step S330, according to the first device archive data, identify the first device data characteristics, and obtain the first edge device level according to the level identification of the device topology network. Specifically, the collected first device archive data is deeply analyzed to check the distribution of the data and determine whether the data shows periodic changes, such as the fluctuation of the operating temperature of the equipment in a fixed time period every day, analyze the stability of the data, whether there are abnormal values ​​or mutation points, such as the sudden increase of the equipment current, etc.; study the correlation between different parameters, such as the correlation between the equipment speed and energy consumption, and determine the data characteristics of the first device through analysis to provide a basis for subsequent model construction. According to the structure and identification rules of the equipment topology network, find the level identification of the first edge device node in the network, have a clear understanding of the entire topology network, and clarify the meaning of each level. For example, a high level may represent a device that is close to the production source or has an important control effect on the entire production process. By accurately obtaining the level identification, the position and importance of the first edge device node in the entire production system are determined.

[0025] Step S340, based on the first device data characteristics and the first edge device level, configure the first model convergence constraint, and according to the first model convergence constraint, combined with the first device archive data, train to obtain a lightweight prediction model for the first edge device node; and so on, configure the lightweight prediction models of the multiple edge device nodes. Specifically, based on the identified data characteristics of the first device and the obtained hierarchical information of the first edge device, the model convergence constraints are determined. For devices with relatively stable data characteristics and higher hierarchical importance, stricter convergence accuracy requirements may be set to ensure the accuracy of model prediction; for some devices with large data fluctuations and relatively minor hierarchies, the convergence conditions can be appropriately relaxed, but the basic performance of the model must also be guaranteed. The convergence constraints also include restrictions on the number of training iterations, the setting of error ranges, etc. According to the set first model convergence constraints, the model is trained in combination with the collected first device archive data, and the data is divided into a training set, a validation set, and a test set. The training set data is used to input the model for parameter learning and optimization, and the performance of the model is monitored and the parameters are adjusted through the validation set data. Finally, the accuracy and generalization ability of the model are evaluated using the test set data. During the training process, the structure and parameters of the model are continuously adjusted to meet the convergence constraints. After multiple iterations, the lightweight prediction model of the first edge device node is obtained. According to the same steps and methods, other edge device nodes are operated in turn, so as to configure respective lightweight prediction models for multiple edge device nodes, providing strong support for subsequent equipment status evaluation, production scheduling, etc.

[0026] In one possible implementation, a lightweight prediction model is configured for the multiple edge device nodes, and primary equipment status assessment is performed in combination with real-time equipment operation data, including equipment health status analysis and short-term maintenance demand prediction, to obtain multiple primary equipment assessment states, and step S300 further includes step S350, obtaining real-time equipment operation data for key process equipment of the first edge device node. Specifically, the specific sources of data obtained from the key process equipment of the first edge device node are clarified, including various sensors installed on the equipment, such as temperature sensors, pressure sensors, vibration sensors, etc., which are used to collect physical parameters during the operation of the equipment; the operation data recorded by the equipment's own control system, such as the equipment's speed, working hours, energy consumption and other information; and the data feedback from other equipment or systems on the production line associated with the equipment, such as the interaction data between upstream and downstream equipment, to fully understand the operating status of the equipment in the actual production environment. According to the importance of the equipment, the operating characteristics and the demand for real-time data, set an appropriate data collection frequency. For critical equipment with rapidly changing operating conditions, a higher collection frequency can be used, such as collecting data once per second; for relatively stable equipment, the collection frequency can be appropriately reduced, such as collecting data once per minute, to balance the timeliness of data acquisition and the data processing pressure of the system. During the collection process, the accuracy and completeness of the data must be ensured, and data verification and fault-tolerant mechanisms must be adopted to prevent subsequent analysis from being affected by data transmission errors or losses.

[0027] Step S360, based on the real-time equipment operation data, perform data volume analysis, perform data slicing according to the volume analysis results, and obtain multiple equipment operation sequences arranged in sequence. Specifically, after obtaining the real-time equipment operation data, immediately parse the data volume, analyze the factors such as the size of the data, the diversity of data types, and the rate of data generation, and use data statistical analysis methods to calculate the total amount of data, the proportion of different types of data, and the growth of data in a unit time. Through analysis, understand the complexity and processing difficulty of the data, and provide a basis for subsequent data slicing. According to the results of the volume analysis, perform data slicing according to certain rules, and perform slicing according to the time series of the data, and divide the data within a continuous period of time into a segment; it can also be classified and sliced ​​according to the attributes or characteristics of the data, for example, temperature data, pressure data, etc. are divided into different slices. In the slicing process, add identification information to each data slice to ensure the order and correlation between the data slices, so that they can be accurately processed and analyzed later, and finally obtain multiple equipment operation sequences arranged in sequence.

[0028] Step S370, using the multiple equipment operation sequences as input data, respectively performing equipment health status analysis through the lightweight prediction model, and obtaining multiple state analysis results. Specifically, multiple equipment operation sequences are sequentially input into the lightweight prediction model configured for the equipment node, and the model analyzes and processes each sequence according to the pre-set algorithm and model parameters, identifies abnormal points and potential problems in the data by comparing the data in the equipment operation sequence with the normal equipment operation data mode stored in the model, and uses data analysis algorithms such as cluster analysis and anomaly detection algorithms to deeply mine the data, analyze whether the various operating parameters of the equipment are within the normal range, and whether the relationship between the parameters conforms to the logic of normal operation of the equipment, and obtain multiple state analysis results after the model analysis and processing, including the current operation status evaluation of the equipment (such as normal operation, slight abnormality, severe abnormality, etc.), the working status analysis of each component of the equipment (such as the working efficiency of a certain component, the degree of wear, etc.) and the preliminary evaluation of the overall performance of the equipment. These results are recorded and sorted to provide basic data for subsequent health trend analysis.

[0029] Step S380, perform health trend analysis based on the multi-segment state analysis results to obtain the device state change trend. Specifically, based on the multi-segment state analysis results, use a trend analysis algorithm to analyze the change in the device state, and draw a trend graph of the device operating parameters over time to observe whether the parameter change trend is stable, rising, falling, or fluctuating. Analyze the correlation and mutual influence between different parameter trends, and predict the health status trend of the device in the future. For example, if the temperature of the device has been on an upward trend and is close to the warning value, it can be predicted that the device may be at risk of overheating failure.

[0030] Step S390, extract the historical equipment operation and maintenance records of the key process equipment, combine the equipment status change trend, perform short-term maintenance demand forecasting, and obtain the primary equipment assessment status. Specifically, extract the historical equipment operation and maintenance records of the key process equipment, including previous maintenance time, maintenance content, and the operating status of the equipment after maintenance. Combine the equipment status change trend with the historical operation and maintenance records, and comprehensively consider factors such as the service life of the equipment, the number of maintenances that have been performed, and the importance of the equipment. Use prediction algorithms (such as time series prediction, machine learning prediction algorithms, etc.) to predict short-term maintenance needs, predict whether the equipment will need maintenance in the future, the time point when failures may occur, and the parts and types that need maintenance, etc., and finally obtain the primary equipment assessment status, which provides a favorable basis for equipment management and maintenance decisions.

[0031] In one possible implementation, historical equipment operation and maintenance records of the key process equipment are extracted, and short-term maintenance demand forecasts are performed in combination with the equipment status change trend to obtain the primary equipment assessment status. Step S390 further includes step S391, obtaining historical equipment operation and maintenance records for the key process equipment of the first edge device node. Specifically, the storage location and source channels of historical equipment operation and maintenance records are clarified. The records are stored in the enterprise's equipment management system database, covering data after electronicization of paper documents, data manually entered by equipment maintenance personnel, and data extracted from the equipment's own storage module, etc. The records are classified and sorted, such as classification by maintenance type (routine maintenance, fault repair, regular maintenance, etc.), maintenance time sequence, maintenance parts, etc., so as to extract and analyze data more efficiently in the future. From the selected data source, the relevant historical operation and maintenance records are accurately extracted according to the unique identification of the key process equipment of the first edge device node (such as equipment number, serial number, etc.). The extracted content includes the specific date of each maintenance, maintenance duration, maintenance reasons (such as equipment parts damage, software system failure, operating errors, etc.), maintenance measures (replacement of parts model, maintenance technical means, software upgrade version, etc.) and equipment operation status evaluation after maintenance (such as equipment performance recovery, whether there are still potential problems, etc.), the extracted data is cleaned and preprocessed to remove duplicate, erroneous or incomplete data records, to ensure the accuracy and completeness of the data, and provide a reliable data basis for subsequent analysis.

[0032] Step S392, according to the production plan of the target product, set the forecast window. Specifically, the production plan of the target product is studied, including key information such as the production quantity, production cycle, and delivery time of the product, and the characteristics and requirements of different production stages such as peak period, trough period, and special production task stage in the production plan are analyzed. According to the production process and process requirements of the product, the requirements for the stability and reliability of the equipment operation in each production stage, as well as the changes in the operating load of the equipment under different production intensities, are determined. In combination with the operating characteristics of the equipment and the requirements of the production plan, a reasonable forecast window is set. The length of the forecast window can be determined according to factors such as the maintenance cycle of the equipment, the urgency of product production, and the availability of historical data. For example, for equipment with a short maintenance cycle, urgent production tasks, and rich historical data, a shorter forecast window (such as one week or half a month) can be set; and for equipment with a long maintenance cycle, relatively stable production tasks, and limited historical data, the forecast window (such as one month or one quarter) can be appropriately extended. When determining the forecast window, the balance between the timeliness of the data and the accuracy of the forecast should also be considered to ensure that the forecast results can provide timely and effective support for production decisions.

[0033] Step S393, based on the historical equipment operation and maintenance records and the equipment status change trend, perform equipment failure prediction according to the prediction window to obtain the equipment's short-term maintenance needs. Specifically, integrate the data of historical equipment operation and maintenance records and equipment status change trends, and use data analysis algorithms to mine the laws and patterns in historical operation and maintenance data, such as the periodicity of certain failures, the correlation between failures and changes in equipment operating parameters, etc. At the same time, combined with the equipment status change trend, analyze the similarities and differences between the current equipment status and the status before the historical failure, and use machine learning algorithms (such as neural network algorithms, decision tree algorithms, etc.) to build a fault prediction model. Input the integrated data into the model for training and prediction. Within the prediction window, based on the output results of the fault prediction model, evaluate the possibility and type of equipment failure in the future, such as If the prediction model shows that the equipment may fail at a certain point in time, further analyze the production links that may be affected by the failure, the severity of the failure, and the possible losses. According to the failure prediction results, determine the short-term maintenance needs of the equipment, including whether preventive maintenance is needed in advance, the specific content of maintenance (such as inspection, replacement or maintenance of equipment parts, debugging or upgrading of software systems, etc.), maintenance schedule (reasonably select maintenance time according to the production plan to minimize the impact on production) and required maintenance resources (such as manpower, material resources, spare parts, etc.). By accurately predicting equipment failures and clarifying short-term maintenance needs, we can provide guarantees for the stable operation of equipment and the smooth progress of production.

[0034] Step S400, through the data sharing of the associated devices of the plurality of edge device nodes, the evaluation status is corrected to obtain the edge device status evaluation result. Specifically, firstly, based on the device topology network, a communication link is built using wired or wireless network technology, and a suitable connection mode is selected according to factors such as device distribution and a relay device is set to enhance the signal stability. At the same time, a unified data transmission protocol is formulated to standardize the data format, encoding, transmission rate and verification rules, and data encryption is also used to ensure security. Then, each edge device node sets the collection frequency according to the criticality of the device and the speed of data change, collects its own operation, status and environmental data, and publishes it to the associated device node after encapsulation and encoding according to the protocol and adds a timestamp and device identification. After receiving the data, the associated device node parses it according to the protocol, verifies the validity, and retransmits it when abnormal, and stores the parsed data for subsequent processing. Then, the edge device impact matrix is ​​constructed based on the shared data of the associated devices and the environmental monitoring data of the device itself, and the correlation between the devices and the influence of environmental factors are analyzed by the algorithm. Based on this, the device correlation impact analysis is performed, and the primary equipment evaluation status is adjusted according to the analysis results to make the evaluation results more in line with the actual production environment. The actual situation, the edge device status evaluation results are finally obtained for subsequent production scheduling and equipment maintenance decisions.

[0035] In a possible implementation, the evaluation status is corrected by sharing the associated device data of the multiple edge device nodes to obtain the edge device status evaluation result, and step S400 further includes step S410, setting a data sharing channel between the multiple edge device nodes through the device topology network. Specifically, relying on the existing physical connection and logical architecture of the device topology network, the path and method of data transmission are determined. For edge device nodes connected to the wired network, the connectivity and bandwidth of the network line are checked to ensure that the data transmission requirements can be met. For the wireless network connection part, the signal strength and frequency band settings are optimized to reduce signal interference. Data switches, routers and other equipment are installed at key nodes of the network to reasonably allocate and manage data traffic. Different network areas are divided according to the distribution location of edge device nodes and the frequency of data interaction. Corresponding network addresses and subnet masks are set to ensure accurate transmission and addressing of data between areas. A security protection mechanism is established for the data sharing channel. Firewall technology is used to block external illegal network access, prevent data leakage and malicious attacks, and set up user authentication and authorization mechanisms. Only authorized device nodes can transmit data in the channel. At the same time, data backup and redundancy technology is used to ensure that data is not lost and transmission can be quickly restored when a network failure occurs. Performance testing and maintenance of data sharing channels are performed regularly to promptly discover and repair network failures, bandwidth bottlenecks and other problems to ensure channel stability and reliability.

[0036] Step S420, based on the data sharing channel, the data of the associated devices is shared, and the evaluation status data of the associated devices is obtained. Specifically, a unified data format and transmission protocol are formulated to standardize the data sharing between the edge device nodes, and to clarify the field structure, data type, unit and encoding method of the data. For example, it is stipulated that the device operation status data is transmitted in a specific binary encoding format, which contains specific fields of parameters such as device temperature and pressure, and a handshake protocol and response mechanism for data transmission are formulated to ensure that the sender and the receiver can accurately identify the sending and receiving status of the data. Each edge device node encapsulates its own data according to a unified data format and protocol to facilitate transmission in the data sharing channel. Each edge device node actively pushes or shares data in a request-response manner through the data sharing channel. When the device status changes or reaches a certain data collection cycle, it actively sends its own device evaluation status data to the associated device node. At the same time, it can respond to data requests from other device nodes and accurately provide the required data according to the request content. In the process of data sharing, a data cache mechanism is established to temporarily store frequently requested data to improve the speed of data acquisition. Data compression technology is used to reduce the amount of data transmission, improve data transmission efficiency, and reduce network load.

[0037] Step S430: construct an edge device impact matrix based on the associated device evaluation status data and the real-time environmental monitoring data of each edge device node. Specifically, the acquired associated device evaluation status data is integrated with the real-time environmental monitoring data of each edge device node itself, and the potential correlation between the data is mined through the data analysis algorithm to analyze the correlation of different device status changes in time, space and function. For example, during the operation of adjacent equipment on the same production line, it is studied whether the failure of one device will cause an abnormal change in the state of another device; how changes in environmental factors such as temperature and humidity affect the operating state of the equipment and the interaction between the devices. According to the analysis results, the influencing factors and degree of influence between the devices are determined, the elements of the influence matrix are constructed, and the structure of the edge device influence matrix is ​​designed. The rows and columns of the matrix correspond to different edge device nodes, and the element values ​​in the matrix represent the degree of influence of one device on another, which can be quantified by numerical values, levels or probabilities. At the same time, an additional dimension of environmental factor influence is set in the matrix to record the comprehensive impact of environmental factors on the equipment status. The influence matrix is ​​regularly updated and optimized. With the continuous accumulation of data and changes in the production process, the element values ​​in the matrix are adjusted in time to ensure that the influence matrix can accurately reflect the actual impact relationship between devices and between devices and the environment.

[0038] Step S440: Perform device association impact analysis based on the edge device impact matrix, and correct the primary device evaluation status of each edge device node according to the analysis result. Specifically, based on the constructed edge device impact matrix, mathematical models and data analysis methods are used to perform device association impact analysis. For example, a multivariate regression analysis method is used to study the relationship between multiple device state variables; a causal relationship analysis model is used to determine the causal impact path between devices. By analyzing the impact matrix, key device nodes and strong association chains between devices are identified. For the primary device evaluation status of each edge device node, a comprehensive evaluation and correction are performed based on the impact relationship between devices and the role of environmental factors. If the state change of the associated device has a positive or negative impact on the evaluation status of a certain device, the evaluation status of the device is adjusted according to the quantitative data in the impact matrix. For example, when an associated device fails and causes an increase in the load of surrounding devices, the health status evaluation of the corresponding device may decrease, and its evaluation index is adjusted according to the degree of impact. At the same time, the correction effect of environmental factors on the equipment evaluation status is considered. For example, the failure rate of equipment in a high temperature environment may increase, and the correction coefficient of environmental factors is introduced into the evaluation status. After comprehensive analysis and correction, a more accurate and actual production evaluation status of each edge device node is obtained, which provides a reliable basis for subsequent production scheduling, equipment maintenance and other decisions.

[0039] Step S500, based on the edge device status evaluation results, combined with the production equipment archive, generate equipment scheduling and maintenance decisions for intelligent production scheduling. Specifically, firstly, the edge device status evaluation results are deeply analyzed, and abnormal, faulty or inefficient equipment is identified by setting thresholds, and the scheduling priority is determined by considering the urgency of the production plan and the importance of the equipment. Then, the equipment scheduling demand type is clarified, such as maintenance, replacement or calling of spare equipment, etc., as well as the layout adjustment demand. Then, the idle equipment information is obtained from the production equipment archive, including specifications, performance, status and storage location, etc., and matching rules such as function, capacity, time, and cost are established to screen the idle equipment combination that meets the conditions and formulate alternative scheduling strategies, covering scheduling paths, time, personnel and maintenance and replacement plans. After that, a mathematical model is established with scheduling cost (including transportation, installation and commissioning, maintenance and production interruption costs, etc.) and scheduling efficiency (measured by equipment production time, production line recovery speed and task completion time) as the optimization target for quantitative processing, and optimization algorithms such as genetic algorithms or simulated annealing algorithms are used to evaluate and screen alternative strategies. Individuals are selected through fitness evaluation during iteration for crossover and mutation operations to converge to the optimal strategy. Finally, the optimal decision is evaluated for risk and feasibility, clearly presented to management personnel, and a monitoring mechanism is established to track progress in real time to ensure the smooth implementation of intelligent production scheduling.

[0040] In a possible implementation, according to the edge device status evaluation results, combined with the production equipment archive, equipment scheduling and maintenance decisions are generated to perform intelligent production scheduling, and step S500 further includes step S510, based on the edge device status evaluation results, equipment scheduling requirements are screened and extracted. Specifically, the edge device status evaluation results are analyzed in detail, and the health status indicators, performance parameter change trends, and fault prediction information of the equipment are checked. For the health status indicators, if they are lower than the preset safety threshold, such as the wear of key components of the equipment reaches a certain level, or the energy consumption of the equipment operation exceeds the normal range, they are marked as equipment that needs to be scheduled and paid attention to, and the performance parameter change trends are analyzed. If the production efficiency of the equipment continues to decline, or the product quality-related parameters are unstable and exceed the tolerance range, they are also included in the scheduling requirements. According to the fault prediction information, if it is predicted that the equipment may have a serious fault in the short term that affects production, it will be treated as an emergency scheduling requirement. At the same time, combined with the production plan, if the current production task is tight and higher requirements are placed on the stability and production efficiency of the equipment, priority will be given to those equipment scheduling needs that have a greater impact on the production progress. The equipment scheduling needs will be classified and extracted, which can be divided into maintenance needs, that is, the equipment needs to be repaired and maintained to restore normal operation, including minor repairs, major repairs, and preventive maintenance; replacement needs, when the equipment is seriously aged or the fault cannot be repaired, the equipment needs to be replaced; deployment needs, for example, due to changes in the production task volume, the work tasks or work areas of the equipment need to be reallocated, etc., and the key information of each type of demand must be clarified, such as maintenance needs to determine the urgency of the maintenance, the required maintenance resources and time, etc.; replacement needs must clarify the selection requirements and replacement time requirements of the new equipment; deployment needs must clarify the new work task goals and deployment paths of the equipment and other information.

[0041] Step S520, according to the production equipment archive, obtain the basic information of idle equipment, match it with the equipment scheduling requirements, and generate multiple alternative scheduling strategies. Specifically, the basic information of idle equipment is fully retrieved from the production equipment archive, and the specification parameters of idle equipment are obtained, including detailed information such as the size, power, and production capacity of the equipment, to determine whether it meets the production process requirements, check the technical performance indicators of the equipment, such as accuracy level, running speed, etc., to ensure that it matches the scheduling requirements, understand the current status of the equipment, whether it has been maintained, the last use time, and the storage environment, etc., to evaluate the possibility of its immediate use, and record the location information of the idle equipment to calculate the scheduling transportation cost and time cost, and match the equipment scheduling requirements with the basic information of idle equipment. Taking maintenance requirements as an example, if the equipment needs to be repaired because a certain component is damaged, find out whether there are replaceable components or auxiliary equipment that can be used for maintenance from the idle equipment. For replacement requirements, according to the production process requirements and equipment performance parameters, select suitable equipment models from the idle equipment as replacement candidates. In the deployment requirements, according to the change in the production task and the capacity of the equipment, match the idle equipment that can meet the new task requirements. For each matching situation, a corresponding scheduling strategy is formulated, including equipment transportation plan, installation and commissioning plan, personnel arrangement, and integration plan with existing production lines, thereby generating multiple alternative scheduling strategies.

[0042] Step S530, for the multiple alternative scheduling strategies, optimization is performed with scheduling cost and scheduling efficiency as optimization targets, and the optimal scheduling strategy is obtained as the equipment scheduling and maintenance decision. Specifically, an optimization model is constructed with scheduling cost and scheduling efficiency as the core. For scheduling cost, the transportation cost, equipment loading and unloading cost, possible equipment damage risk cost, etc. during the equipment transportation process are considered. The maintenance cost includes the cost of parts required for maintenance, maintenance personnel time cost, and maintenance equipment rental cost, etc. The production interruption cost is calculated based on the production plan, product price, and interruption time. For scheduling efficiency, a time model is established to calculate the time from equipment preparation to production, including transportation time, installation and commissioning time, etc. The production recovery model evaluates the speed at which the production line recovers normal production efficiency after equipment scheduling. These cost and efficiency factors are quantified and integrated by establishing mathematical formulas or algorithm models, and intelligent optimization algorithms such as genetic algorithms or simulated annealing algorithms are used for calculation. In the genetic algorithm, each candidate scheduling strategy is encoded as a gene sequence, and the optimal solution is found by simulating the selection, crossover and mutation operations in the biological evolution process. The simulated annealing algorithm starts from an initial solution, accepts a poor solution with a certain probability to avoid falling into the local optimal solution, and gradually iterates and optimizes. During the algorithm iteration process, it continuously evaluates the objective functions of scheduling cost and scheduling efficiency and updates the state of the solution. After multiple iterations, the optimal scheduling strategy is obtained. This strategy achieves the lowest scheduling cost and the highest scheduling efficiency under the premise of meeting production needs, which serves as the final equipment scheduling and maintenance decision.

[0043] The embodiment of the present application first establishes an equipment archive information library to record the information of the active equipment and spare equipment of the target production line, and each device has an independent code; then, based on the equipment archive information library and the process information of the production line, a device topology network is constructed, and the network contains multiple edge device nodes. A lightweight prediction model is configured for each edge device node, and the equipment health status analysis and short-term maintenance demand prediction are performed in combination with real-time equipment operation data to obtain a preliminary evaluation status. Through data sharing of edge device nodes, the evaluation status is corrected to obtain the final evaluation result of the equipment. Finally, based on the evaluation results and the equipment archive library, equipment scheduling and maintenance decisions are generated to realize intelligent production scheduling, achieving the technical effect of improving equipment utilization, optimizing production scheduling, and ensuring continuous and efficient operation of the production line.

[0044] In the above, refer to Figure 1 The intelligent production scheduling method for the equipment manufacturing industry according to the embodiment of the present invention is described in detail. Figure 2 An intelligent production scheduling device for equipment manufacturing industry according to an embodiment of the present invention is described.

[0045] The intelligent production scheduling device for the equipment manufacturing industry according to the embodiment of the present invention is used to solve the technical problem of the lack of flexibility and predictability of equipment maintenance and production scheduling in the prior art, and achieves the technical effect of improving equipment utilization, optimizing production scheduling, and ensuring continuous and efficient operation of the production line. The intelligent production scheduling device for the equipment manufacturing industry includes: an equipment archive information library establishment module 10, an equipment topology network construction module 20, a primary equipment status evaluation module 30, an evaluation status correction module 40, and an intelligent production scheduling module 50.

[0046] The equipment archive information base establishing module 10 is used to establish an equipment archive information base, which contains the active equipment information and spare equipment information of the target production line, and each equipment has an independent code.

[0047] The device topology network construction module 20 is used to construct a device topology network according to the device archive information library and in combination with the process information of the target production line. The device topology network includes a plurality of edge device nodes.

[0048] The primary equipment status assessment module 30 is used to configure a lightweight prediction model for the multiple edge device nodes, and perform primary equipment status assessment in combination with real-time equipment operation data, including equipment health status analysis and short-term maintenance demand prediction, to obtain multiple primary equipment assessment states.

[0049] The evaluation status correction module 40 is used to perform evaluation status correction by sharing associated device data of the plurality of edge device nodes to obtain edge device status evaluation results.

[0050] The intelligent production scheduling module 50 is used to generate equipment scheduling and maintenance decisions for intelligent production scheduling based on the edge device status evaluation results and in combination with the production equipment archive.

[0051] The specific configuration of the device topology network construction module 20 will be described in detail below. As described above, according to the device archive information library, combined with the process information of the target production line, a device topology network is constructed, and the device topology network includes multiple edge device nodes. The device topology network construction module 20 further includes: a sub-process division unit, the sub-process division unit is used to obtain the process information of the target production line according to the process flow chart of the target production line to divide the sub-process into sub-processes, and obtain multiple sub-production processes, the sub-production processes are independent production processes of single components, and the sub-production processes include corresponding key process nodes and key process equipment; a device node configuration unit, the device node configuration unit is used to configure the corresponding multiple edge device nodes based on the multiple sub-production processes, and build a device topology network based on the multiple edge device nodes, and the device topology network includes multiple levels corresponding to different process priorities.

[0052] The following will describe in detail the primary equipment status assessment module 30 for configuring a lightweight prediction model for the multiple edge device nodes, and performing primary equipment status assessment in combination with real-time equipment operation data, including performing equipment health status analysis and short-term maintenance demand prediction, and obtaining multiple primary equipment assessment states. The primary equipment status assessment module 30 further includes: a first edge device node extraction unit, the first edge device node extraction unit is used to extract a first edge device node based on the multiple edge device nodes; an identity information extraction unit, the identity information extraction unit is used to traverse the device archive information library for identity information extraction for the first edge device node, and extract the first edge device node. A first device archive data set; a first edge device level acquisition unit, the first edge device level acquisition unit is used to identify the first device data characteristics according to the first device archive data, and obtain the first edge device level according to the level identifier of the device topology network; a lightweight prediction model training unit, the lightweight prediction model training unit is used to configure a first model convergence constraint based on the first device data characteristics and the first edge device level, and train a lightweight prediction model of the first edge device node according to the first model convergence constraint and the first device archive data; and so on, configure the lightweight prediction models of the multiple edge device nodes.

[0053] Among them, the primary equipment status assessment module 30 further includes: a real-time equipment operation data acquisition unit, which is used to acquire real-time equipment operation data for the key process equipment of the first edge device node; a data volume analysis unit, which is used to perform data volume analysis based on the real-time equipment operation data, and perform data segmentation according to the volume analysis results to obtain multiple equipment operation sequences arranged in sequence; a multi-segment status analysis result acquisition unit, which is used to use the multiple equipment operation sequences as input data, and respectively perform equipment health status analysis through the lightweight prediction model to obtain multi-segment status analysis results; a health trend analysis unit, which is used to perform health trend analysis based on the multi-segment status analysis results to obtain the equipment status change trend; a short-term maintenance demand prediction unit, which is used to extract the historical equipment operation and maintenance records of the key process equipment, and perform short-term maintenance demand prediction in combination with the equipment status change trend to obtain the primary equipment assessment status.

[0054] Among them, the historical equipment operation and maintenance records of the key process equipment are extracted, and combined with the equipment status change trend, short-term maintenance demand prediction is performed to obtain the primary equipment evaluation status. The short-term maintenance demand prediction unit further includes: a historical equipment operation and maintenance record acquisition subunit, the historical equipment operation and maintenance record acquisition subunit is used to obtain historical equipment operation and maintenance records for the key process equipment of the first edge device node; a prediction window setting subunit, the prediction window setting subunit is used to set the prediction window according to the production plan of the target product; an equipment failure prediction subunit, the equipment failure prediction subunit is used to predict equipment failure according to the prediction window based on the historical equipment operation and maintenance records and the equipment status change trend, and obtain the equipment short-term maintenance demand.

[0055] The specific configuration of the evaluation state correction module 40 will be described in detail below. As described above, the evaluation state correction is performed through the associated device data sharing of the multiple edge device nodes to obtain the edge device state evaluation result. The evaluation state correction module 40 further includes: a data sharing channel setting unit, the data sharing channel setting unit is used to set a data sharing channel between the multiple edge device nodes through the device topology network; an associated device evaluation state data acquisition unit, the associated device evaluation state data acquisition unit is used to perform associated device data sharing based on the data sharing channel to obtain associated device evaluation state data; an edge device impact matrix construction unit, the edge device impact matrix construction unit is used to construct an edge device impact matrix based on the associated device evaluation state data and the real-time environmental monitoring data of each edge device node; a device association impact analysis unit, the device association impact analysis unit is used to perform device association impact analysis based on the edge device impact matrix, and perform primary device evaluation state correction of each edge device node according to the analysis result.

[0056] The specific configuration of the intelligent production scheduling module 50 will be described in detail below. As described above, according to the edge device status evaluation result, combined with the production equipment archive, an equipment scheduling and maintenance decision is generated to perform intelligent production scheduling. The intelligent production scheduling module 50 further includes: an equipment scheduling demand screening unit, the equipment scheduling demand screening unit is used to screen and extract equipment scheduling requirements based on the edge device status evaluation result; an alternative scheduling strategy generation unit, the alternative scheduling strategy generation unit is used to obtain the basic information of idle equipment according to the production equipment archive, match it with the equipment scheduling requirements, and generate multiple alternative scheduling strategies; an equipment scheduling maintenance decision acquisition unit, the equipment scheduling maintenance decision acquisition unit is used to optimize the multiple alternative scheduling strategies with scheduling cost and scheduling efficiency as the optimization target, and obtain the optimal scheduling strategy as the equipment scheduling maintenance decision.

[0057] The intelligent production scheduling device for the equipment manufacturing industry provided in the embodiment of the present invention can execute the intelligent production scheduling method for the equipment manufacturing industry provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0058] Although the present application makes various references to certain modules in the device according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0059] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. An intelligent production scheduling method for equipment manufacturing industry, characterized in that: The method comprises: Establishing an equipment archive information database, which contains the information of active equipment and spare equipment of the target production line, and each equipment has an independent code; According to the equipment archive information library, combined with the process information of the target production line, a device topology network is constructed, wherein the device topology network includes a plurality of edge device nodes; Configuring a lightweight prediction model for the multiple edge device nodes, and performing primary equipment status assessment in combination with real-time equipment operation data, including performing equipment health status analysis and short-term maintenance demand forecasting, to obtain multiple primary equipment assessment states; By sharing the associated device data of the plurality of edge device nodes, the evaluation status is corrected to obtain an edge device status evaluation result; According to the edge device status evaluation results, combined with the production equipment archive, equipment scheduling and maintenance decisions are generated to perform intelligent production scheduling; According to the equipment archive information database, combined with the process information of the target production line, a equipment topology network is constructed, including: According to the process flow chart of the target production line, the process information of the target production line is obtained to divide the process into sections, and a plurality of section production processes are obtained, wherein the section production process is an independent production process of a single component, and the section production process includes corresponding key process nodes and key process equipment; Based on the multiple divisional production processes, a corresponding plurality of edge device nodes are configured, and based on the multiple edge device nodes, a device topology network is constructed, wherein the device topology network includes multiple levels corresponding to different process priorities; Configuring a lightweight prediction model for the plurality of edge device nodes includes: Based on the multiple edge device nodes, extracting a first edge device node; For the first edge device node, traverse the device archive information library to extract identity information and collect first device archive data; According to the first device profile data, identifying the first device data characteristic, and according to the level identifier of the device topology network, obtaining the first edge device level; Based on the first device data characteristics and the first edge device level, configure a first model convergence constraint, and train a lightweight prediction model of the first edge device node according to the first model convergence constraint and in combination with the first device archive data; By analogy, the lightweight prediction models of the multiple edge device nodes are configured.

2. The intelligent production scheduling method for equipment manufacturing industry according to claim 1, characterized in that: Based on lightweight prediction models, combined with real-time equipment operation data, primary equipment status assessment is performed, including equipment health status analysis and short-term maintenance demand prediction, including: For key process equipment of the first edge device node, obtain real-time equipment operation data; Based on the real-time equipment operation data, data volume analysis is performed, and data slicing is performed according to the volume analysis result to obtain a plurality of equipment operation sequences arranged in sequence; Using the multiple equipment operation sequences as input data, respectively performing equipment health status analysis through the lightweight prediction model to obtain multiple status analysis results; Perform health trend analysis based on the multiple-segment status analysis results to obtain the device status change trend; The historical equipment operation and maintenance records of the key process equipment are extracted, and combined with the equipment status change trend, short-term maintenance demand forecast is carried out to obtain the primary equipment assessment status.

3. The intelligent production scheduling method for equipment manufacturing industry according to claim 2, characterized in that: The short-term maintenance demand forecasting includes: For the key process equipment of the first edge device node, obtain historical equipment operation and maintenance records; Set the forecast window according to the production plan of the target product; Based on the historical equipment operation and maintenance records and the equipment status change trend, equipment failure prediction is performed according to the prediction window to obtain the short-term maintenance requirements of the equipment.

4. The intelligent production scheduling method for equipment manufacturing industry according to claim 1, characterized in that: By sharing the associated device data of the plurality of edge device nodes, the assessment status is corrected, including: Setting a data sharing channel between the plurality of edge device nodes through the device topology network; Based on the data sharing channel, the associated device data is shared to obtain the associated device evaluation status data; Constructing an edge device impact matrix based on the associated device evaluation status data and combining the real-time environmental monitoring data of each edge device node; Based on the edge device impact matrix, a device association impact analysis is performed, and the primary device assessment status of each edge device node is corrected according to the analysis results.

5. The intelligent production scheduling method for equipment manufacturing industry according to claim 1, characterized in that: According to the edge device status evaluation results, combined with the production equipment archive, equipment scheduling and maintenance decisions are generated, including: Based on the edge device status evaluation result, screen and extract device scheduling requirements; According to the production equipment archive, basic information of idle equipment is obtained, matched with the equipment scheduling requirements, and multiple alternative scheduling strategies are generated; For the multiple candidate scheduling strategies, optimization is performed with scheduling cost and scheduling efficiency as optimization objectives to obtain the optimal scheduling strategy as the equipment scheduling and maintenance decision.

6. Intelligent production scheduling device for equipment manufacturing industry, characterized in that: The device is used to implement the intelligent production scheduling method for the equipment manufacturing industry according to any one of claims 1 to 5, and the device comprises: An equipment archive information base establishment module, which is used to establish an equipment archive information base, wherein the equipment archive information base contains the active equipment information and spare equipment information of the target production line, and each equipment has an independent code; A device topology network construction module, wherein the device topology network construction module is used to construct a device topology network according to the device archive information library and in combination with process information of a target production line, wherein the device topology network includes a plurality of edge device nodes; A primary equipment status assessment module, which is used to configure a lightweight prediction model for the multiple edge device nodes, and perform primary equipment status assessment in combination with real-time equipment operation data, including performing equipment health status analysis and short-term maintenance demand prediction, to obtain multiple primary equipment assessment states; An evaluation status correction module, the evaluation status correction module is used to perform evaluation status correction through the associated device data sharing of the plurality of edge device nodes to obtain an edge device status evaluation result; An intelligent production scheduling module, which is used to generate equipment scheduling and maintenance decisions for intelligent production scheduling based on the edge device status evaluation results and in combination with the production equipment archive; The device topology network construction module further includes: a sub-process division unit, the sub-process division unit is used to obtain the process information of the target production line according to the process flow chart of the target production line to divide the sub-process into sub-processes, and obtain multiple sub-production processes, the sub-production processes are independent production processes of single components, and the sub-production processes include corresponding key process nodes and key process equipment; a device node configuration unit, the device node configuration unit is used to configure the corresponding multiple edge device nodes based on the multiple sub-production processes, and build a device topology network based on the multiple edge device nodes, the device topology network includes multiple levels, corresponding to different process priorities; The primary equipment status assessment module further includes: a first edge device node extraction unit, the first edge device node extraction unit is used to extract the first edge device node based on the multiple edge device nodes; an identity information extraction unit, the identity information extraction unit is used to traverse the device archive information library for the first edge device node to extract identity information and collect first device archive data; a first edge device level acquisition unit, the first edge device level acquisition unit is used to identify the first device data characteristics according to the first device archive data, and obtain the first edge device level according to the level identifier of the device topology network; a lightweight prediction model training unit, the lightweight prediction model training unit is used to configure the first model convergence constraint based on the first device data characteristics and the first edge device level, and according to the first model convergence constraint, combined with the first device archive data, train to obtain a lightweight prediction model of the first edge device node; and so on, configure the lightweight prediction models of the multiple edge device nodes.

Images