Intelligent workshop data acquisition and analysis system

Abstract

The invention discloses an intelligent workshop data acquisition and analysis system, and belongs to the technical field of intelligent manufacturing. The system comprises a data management module, a performance prediction module and a performance analysis and monitoring module. The data management module provides necessary data basis and support for performance prediction, analysis and optimization of the intelligent workshop by efficiently collecting, transmitting and managing workshop production data; the performance prediction module realizes accurate prediction of production line performance and early identification of potential problems through parameter selection, index prediction visualization and bottleneck identification functions; the performance analysis and monitoring module supports global or local performance visualization after production line selection, integrates a prediction result and actual performance data through an intuitive visual interface and a real-time data flow updating mechanism, ensures that a user can know the production state and trend in time, provides decision support and scheduling optimization suggestions, and improves the production efficiency. And continuous improvement of the production process is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent manufacturing, and in particular relates to an intelligent workshop data collection and analysis system. Background Art

[0002] With the rapid development of Industry 4.0 and intelligent manufacturing, modern manufacturing is gradually moving towards intelligence and automation. As an important part of intelligent manufacturing, the smart workshop has become a key means to improve production efficiency and product quality by introducing advanced sensor technology, data collection and analysis technology to achieve all-round monitoring and optimization of the production process.

[0003] At present, smart workshops have made certain progress in data collection and analysis. For example, a smart factory data intelligent analysis management system disclosed in China's invention patent CN 118333580A includes a data collection unit, a data processing unit, a data analysis unit, a recommendation analysis unit, and a data display unit. The data collection unit is used to collect factory production data in multiple historical target cycles and order data in the current target cycle in the target factory; the data processing unit is used to perform data calculation and processing on the factory production data in the previous cycle, and obtain the single product time analysis value of each different specified product; the data analysis unit is used to perform pending order determination analysis based on the required production volume of the received order data and the pending order data and combined with the data calculation and processing results of the data processing unit, and judge whether the pending order data can be completed within the current target cycle based on the order determination analysis results, and then send the judgment results to the data display unit; the data display unit is used to display the analysis results obtained by the data analysis unit to production managers. This solution provides data-driven decision support to help production managers analyze results based on objective data. Another example is a data acquisition and control method and system for a production line in an intelligent workshop disclosed in Chinese invention patent CN118311914A, which includes the following steps: acquiring multi-source data on equipment status; selecting multi-source feature data of equipment status from the multi-source data to generate multi-source feature data of equipment status; integrating time series data of the multi-source feature data of equipment status to generate time series multi-source feature data of equipment status; acquiring equipment control parameters of the workshop production line; establishing an equipment dynamic drive data model based on the equipment control parameters of the workshop production line and the equipment status time series multi-source feature data; analyzing and optimizing the equipment adaptive control strategy based on the equipment dynamic drive data model to generate an optimized equipment adaptive control strategy.

[0004] The systems provided by the above technical solutions can collect various parameters in production equipment and process flow in real time, and use big data analysis technology to warn of abnormal situations in the production process. However, these systems still have some problems in practical applications, mainly including:

[0005] (1) Insufficient data integration: The system can often only process data from a single production line and has difficulty integrating and analyzing data across production lines, making it impossible to provide a global optimization solution.

[0006] (2) Limited prediction and analysis capabilities: The prediction model of the existing system is relatively simple and cannot accurately predict performance and identify bottlenecks, which affects the optimization effect of the production process.

[0007] (3) Poor data visualization: Although some systems have data visualization functions, their display effects are relatively simple and cannot intuitively present key performance indicators and abnormal situations in the production process. Summary of the invention

[0008] In order to solve at least one of the problems existing in the prior art, the present invention provides a smart workshop data collection and analysis system and method, which can realize data integration and analysis across production lines and accurately and quickly identify performance bottlenecks. At the same time, it can intuitively display key performance indicators and abnormal situations, significantly improve the production efficiency of smart workshops, and provide strong support for the realization of intelligent manufacturing.

[0009] To achieve the purpose of the present invention, the present invention provides an intelligent workshop data collection and analysis system, including a data management module, including a data collection submodule, a data transmission submodule and a data storage submodule, the data collection submodule is used to collect factory production data of multiple production lines in a target workshop within multiple historical target cycles and order data and equipment information data within a current target cycle, the data transmission submodule adopts a multi-level network architecture, and is used to establish a mapping between the data management module and the performance prediction module, the data management module and the performance analysis and monitoring module, and the data storage submodule is used to store all data generated by the data management module and the performance prediction module;

[0010] The performance prediction module includes a parameter selection submodule, an indicator prediction visualization submodule and a bottleneck identification submodule. The parameter selection submodule is used to allow the user to select any one or more of the production line name, performance prediction type, prediction algorithm and bottleneck identification method through an interactive interface, and to predict the performance of the production line equipment by accessing the historical production data and current real-time production data stored in the data management module in combination with the algorithm selected by the user, and to generate relevant prediction results. The indicator prediction visualization submodule is used to update the display content, and the bottleneck identification submodule is used to identify the bottleneck by evaluating the prediction results;

[0011] The performance analysis and monitoring module includes a production line selection submodule, a global visualization submodule, a local visualization submodule, and a decision support and scheduling optimization submodule. The production line selection submodule is used to enable users to select the target production line from a preset production line list. The global visualization submodule is used to generate and display the overall performance data of the selected production line. The local visualization submodule is used to make adjustments based on the production line selection and provide detailed performance data at each equipment level of the selected production line. The decision support and scheduling optimization submodule is used to provide decision suggestions and dynamic scheduling optimization solutions by comparing the results of the performance prediction module with the actual operation data.

[0012] The data management module provides the necessary data foundation and support for the performance prediction, analysis and optimization of the smart workshop by efficiently collecting, transmitting and managing workshop production data; the performance prediction module realizes accurate prediction of production line performance and early identification of potential problems through parameter selection, indicator prediction visualization and bottleneck identification functions; the performance analysis and monitoring module supports global or local performance visualization after production line selection, integrates prediction results with actual performance data through an intuitive visualization interface and real-time data stream update mechanism, ensuring that users can understand production status and trends in a timely manner, and providing decision support and scheduling optimization suggestions to achieve continuous improvement of the production process.

[0013] According to one embodiment of the present invention, the data transmission submodule maps the data management module with the performance prediction module, the data management module and the performance analysis and monitoring module through the API interface to ensure seamless data transmission between the submodules and maintain data consistency and real-time performance.

[0014] According to one embodiment of the present invention, the performance prediction module includes three core parts: parameter selection, indicator prediction visualization, and bottleneck identification. The module provides a variety of customizable analysis options, allowing users to flexibly select single or multiple production lines, prediction types, algorithms, and bottleneck identification methods, and write these parameters to the database to trigger subsequent analysis. It can update and visualize the prediction results in real time, including bottleneck equipment identification, performance indicator trend charts, and production line bottleneck analysis charts, and supports integrated performance analysis of multiple production lines. The prediction results will be automatically transmitted to the performance analysis and monitoring module for real-time comparison and scheduling optimization. Through intuitive data display and flexible custom analysis options, the module provides users with accurate performance prediction and bottleneck identification capabilities, effectively supporting the efficient operation and decision-making optimization of smart workshops.

[0015] According to one embodiment of the present invention, the performance analysis and monitoring module includes a production line selection function and a global and local performance visualization function. The module allows users to select a single or multiple production lines through the production line selection function. The system automatically reads the corresponding performance data table from the database and displays it in real time. Through the real-time data stream update mechanism, the system can ensure the timeliness and accuracy of the data (the real-time data stream update mechanism ensures that the system can dynamically update the production status with millisecond or second accuracy through continuous data collection, processing and display, providing the latest and accurate visualization data to support real-time decision-making.). Global performance visualization includes throughput analysis, work-in-process quantity analysis and order progress display, while local performance visualization provides detailed status of a single machine, proportion of workpiece types in the buffer zone and power of the automatic guided vehicle (AGV, Automated Guided Vehicle) and other information. By comparing the analysis results with the prediction results of the performance prediction module, a basis is provided for decision support, helping users optimize the operating efficiency of the production line and make effective scheduling and management decisions.

[0016] Compared with the prior art, the present invention can at least achieve the following beneficial effects:

[0017] The technical solution provided by the embodiment of the present invention includes a data management module, a performance prediction module and a performance analysis and monitoring module. The three modules are precisely combined to realize cross-production line data collection, accurate performance prediction and real-time monitoring and analysis of the smart workshop, providing strong data support for production management decisions and helping to improve the efficiency and accuracy of smart workshop manufacturing.

[0018] The data acquisition submodule of the present invention can collect data from multiple production lines, that is, it can perform cross-line data collection, and can collect data from multiple production lines in multiple historical target cycles, including orders and equipment information data in the current cycle, so that the system can process long-term and real-time data from multiple production lines. In addition, the data transmission submodule mentions the use of a multi-level network architecture, in which Profibus fieldbus, EtherNet / IP industrial Ethernet, and AGV's wireless network (Bluetooth 5.0) jointly ensure efficient data transmission between different levels and different devices. These network technologies enable the system to cover data collection and transmission from a single device to multiple production lines, and realize data interoperability and real-time exchange, further supporting data integration across production lines. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a module schematic diagram of an intelligent workshop data collection and analysis system provided by an embodiment of the present invention.

[0020] Figure 2 It is a schematic diagram of the composition of the data management module in an embodiment of the present invention.

[0021] Figure 3 Schematic diagram of the composition of the performance prediction module in an embodiment of the present invention.

[0022] Figure 4 It is a schematic diagram of the composition of the performance analysis and monitoring module in an embodiment of the present invention. DETAILED DESCRIPTION

[0023] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] See also Figure 1 The present invention provides an intelligent workshop data collection and analysis system, which includes three parts: a data management module, a performance prediction module, and a performance analysis and monitoring module.

[0025] As the foundation of the system, the data management module can realize all-round and multi-dimensional production data collection, cleaning, storage and integration through advanced sensor networks and data processing technologies, providing high-quality and real-time data support for subsequent analysis. The performance prediction module allows users to select different prediction algorithms and bottleneck identification methods through a flexible interactive interface to achieve accurate prediction of production line performance and early identification of potential bottlenecks. The performance analysis and monitoring module processes the factory production line data and the operating status data of production equipment collected in real time by the data acquisition submodule through an intuitive visual interface. When the system receives the data, it immediately processes, filters, cleans and formats it, and stores the data in the data storage submodule. The performance analysis and monitoring module dynamically updates and displays the real-time data obtained from the workshop production line and equipment by reading the real-time data. This process is implemented through the ECharts open source library. ECharts provides powerful chart drawing and dynamic rendering capabilities. It can automatically generate various chart forms such as line charts, bar charts, pie charts, etc. from processed data. By converting complex data analysis results into easy-to-understand charts and indicators, the system integrates forecast results with actual performance data, allowing managers to quickly grasp production status, discover and solve problems in a timely manner, and provide decision support and scheduling optimization suggestions.

[0026] (1) Data Management Module

[0027] The data management module is the core component of the intelligent workshop data acquisition and analysis system. It is responsible for efficiently collecting, processing, transmitting, storing and managing workshop production data. The data management module includes data acquisition sub-module, data storage sub-module and data transmission sub-module, which are responsible for data acquisition, storage and transmission respectively.

[0028] The data collection submodule is used to collect factory production data of multiple production lines in the target workshop in multiple historical target cycles and order data and equipment information data in the current target cycle. Figure 2 As shown in the figure, the data acquisition submodule adopts a diversified acquisition strategy, including: using high-precision vibration sensors to monitor the operating status of the equipment in real time; using power sensors to record equipment energy consumption data; using RFID readers and displacement sensors to accurately track AGV transportation paths and material locations; and equipping the robot arm with torque sensors and position encoders to collect motion parameters and operation accuracy data in real time. In addition, high-speed industrial cameras are configured in key processes to capture workpiece processing details.

[0029] The data management module also preprocesses and filters the factory production line data and production equipment operation status data collected by the data acquisition submodule. The preprocessing steps include: data format conversion, converting heterogeneous data collected from multiple sources into JSON format; data cleaning, checking and processing missing values ​​and abnormal values ​​to ensure data integrity and accuracy; data standardization, using key-value pairs to store data, ensuring that it is expressed in a unified JSON format to improve data consistency and availability.

[0030] The data transmission submodule maps the data management module with the performance prediction module, the data management module, and the performance analysis and monitoring module through the API interface to ensure seamless data transmission between submodules and maintain data consistency and real-time performance.

[0031] In some embodiments of the present invention, the data transmission submodule adopts a multi-level network architecture to ensure efficient and reliable data transmission. At the device level, the industrial field bus technology Profibus is used to achieve real-time communication between sensors and local controllers. The controller layer uses industrial Ethernet technology EtherNet / IP to ensure high-speed data transmission and interoperability between devices. To improve the flexibility of the system, an industrial wireless network (Bluetooth 5.0) is deployed on the AGV to achieve real-time data exchange between mobile devices.

[0032] The data storage submodule is used to store all data generated by the data management module and the performance prediction module.

[0033] The data storage submodule adopts a tiered storage strategy, combining high-performance hardware with a flexible software architecture to meet the needs of smart workshops for large-scale, high-speed, and diversified data storage. The data storage submodule includes a hardware layer, a database layer, and a cache layer, which are used for data storage performance optimization, data structured management, and data access speed improvement, respectively.

[0034] In terms of the hardware layer, the data storage submodule adopts a hybrid storage architecture. In some embodiments of the present invention, a high-speed solid-state drive (SSD) array and a large-capacity mechanical hard disk (HDD) array are used in combination to realize hierarchical storage and management of data. The SSD array is used to store hot data that needs to be accessed quickly, including but not limited to real-time production data, current order information and key performance indicators of the production line, and the key performance indicators include but are not limited to equipment utilization (EU, Equipment Utilization), work in process (WIP, Work in Process), on-time delivery rate (OTD, On-time Delivery), average production cycle (APCT, Average Production Cycle Time), equipment failure rate (FR, Failure Rate), mean time between failures (MTBF, Mean Time Between Failures), mean time to repair (MTTR, Mean Time To Repair), equipment availability (EAR, Equipment Availability Rate), blocking time (BT, Blocking Time) and starving time (ST, Starving Time) of each machine, etc.). The SSD array can ensure a millisecond response time to meet the needs of smart workshops for real-time data processing. The HDD array is used to store cold data with low access frequency, including but not limited to historical production records, energy consumption data and equipment life cycle data. Through this hierarchical storage architecture, the system can meet the storage needs of different types of data in an economical and efficient manner.

[0035] In terms of the database layer, the data storage submodule uses a combination architecture of multiple types of databases to process different types of data. In some embodiments of the present invention, for structured data such as process parameters, production order information, etc., MySQL relational database is used for management to ensure the integrity and consistency of the data; for high-frequency time series data from workshop equipment such as machine operation status, real-time location of AGV, etc., the time series database InfluxDB is used for storage and processing to ensure efficient management of large amounts of time series data and support real-time analysis and monitoring; for unstructured data and semi-structured data such as images and videos captured by industrial cameras, a distributed file system is used for storage to ensure reliable storage and flexible access to data.

[0036] In terms of the cache layer, the data storage submodule introduces the Redis memory database to cache frequently accessed hot data. The Redis memory database has high concurrent processing capabilities and is suitable for data scenarios in the workshop that require real-time decision support. By storing hot data in the Redis memory database, the system can significantly reduce data access latency and improve the real-time response capabilities of the production line.

[0037] The data management module collects workshop production data in real time through the data acquisition submodule, data storage submodule and data transmission submodule, and provides the workshop production data to the performance prediction module and the performance analysis and monitoring module. Among them, the real-time production data collected by the data management module is transmitted to the performance prediction module in real time through the data transmission submodule. The performance prediction module calls the existing prediction model for calculation based on the algorithm selected by the user and the input data, and generates the production line performance prediction and bottleneck identification results (this process uses the existing ARIMA, LSTM, and graph neural network algorithm models to perform inference calculations through the input data and output the corresponding prediction results). The prediction model is used to predict the performance of the production line and identify potential bottlenecks. The data management module can perform data preprocessing and filtering operations according to the needs of the performance prediction module to ensure the accuracy of the input data, thereby improving the reliability of the prediction results and further optimizing the accuracy of the prediction.

[0038] (2) Performance prediction module

[0039] like Figure 3 As shown in the figure, the performance prediction module integrates three sub-modules: parameter selection, indicator prediction visualization, and bottleneck identification, to achieve comprehensive analysis, accurate prediction, and visual display of production line performance.

[0040] The parameter selection submodule adopts a flexible user interaction interface. Users can select the production line name, performance prediction type, prediction algorithm and bottleneck identification method according to actual needs. The user sets the parameters and submits them to the backend through the interactive interface. The backend program writes these parameters into a specific database table. The backend algorithm reads these parameters to perform performance prediction and bottleneck identification operations. The backend algorithm predicts the performance of the production line equipment by accessing the historical production data and current real-time production data stored in the data management module, and combines the algorithm selected by the user to generate relevant prediction results. The prediction results include the operating status of the equipment, bottleneck point prediction, etc., which are stored in the prediction index data table in the form of structured data. The prediction index data table contains multiple columns of information, such as equipment name, bottleneck type, timestamp and various performance indicators. The indicator prediction visualization submodule dynamically refreshes and reads the above prediction index data table, and displays the performance index prediction curve of the bottleneck equipment in real time through the ECharts open source library. The indicator prediction visualization submodule automatically updates the display content every minute to ensure the timeliness and accuracy of the prediction results. The bottleneck identification submodule identifies bottlenecks by evaluating the prediction index data of each device in the prediction index data table. Specifically, the bottleneck identification submodule analyzes the predicted performance indicators of each device, such as utilization, blocking time, starvation time, etc., and determines the bottleneck device through the preset threshold. The identification results are presented in two forms: a bar graph and a production line layout diagram. In the bar graph, the horizontal axis represents time, and the vertical axis lists the names of each machine. The blocking and starvation time of each device is intuitively displayed through columns of different colors; in the production line layout diagram, the bottleneck device is highlighted with a preset color such as red, and the name of the currently identified bottleneck device is displayed in the actual layout.

[0041] The performance prediction type includes two candidate quantities: bottleneck and duration. Bottleneck prediction aims to identify and predict possible bottlenecks in the production line, that is, the links or equipment that limit the overall production efficiency; duration prediction aims to estimate the time required to complete a specific production task or order.

[0042] The prediction algorithm includes ARIMA (Auto Regressive Integrated Moving Average) and LSTM (Long Short-Term Memory). The ARIMA algorithm is used to process univariate time series data, including historical production line performance indicators (such as output, working hours, equipment utilization, etc.), which is suitable for predicting single indicators such as performance equipment and production line output; the LSTM algorithm is used to process multidimensional time series data, including historical production line performance indicators and other factors that may affect performance (such as environmental conditions, equipment status, etc.), which is suitable for predicting complex, multi-dimensional production indicators (such as equipment utilization, cycle time, equipment failure rate, etc.); the prediction algorithm can also consider using a prediction algorithm combined with a graph neural network (GNN) and a gated recurrent unit (GRU). The combination algorithm of the graph neural network and the gated recurrent unit captures the spatial interaction in the process line through the graph neural network, and combines the prediction ability of the recursive neural network in the time dimension. It performs well in the accuracy and generalization ability of the complex product duration prediction model. The algorithm can effectively model the complex relationship between the process line layout and the process, and is suitable for processing the duration prediction problem in a complex multi-process manufacturing environment.

[0043] The bottleneck identification method includes the turning point method and the utilization method. The turning point method identifies short-term bottlenecks based on the changes in the blocking and starvation states between workstations; the utilization method identifies long-term bottlenecks by analyzing the utilization of each workstation.

[0044] Specifically, when the ARIMA algorithm is selected, the user is allowed to set the order p of the autoregressive (AR) part, the order d of the difference (I) part, and the order q of the moving average (MA) part; when the LSTM algorithm is selected, the user is allowed to set parameters such as the number of network layers, the number of neurons, and the learning rate. When the graph neural network and gated recurrent unit combination algorithm are selected, the user is allowed to set the number of layers, the number of attention heads, the node feature dimension, the number of GRU units, the time step, the learning rate, and other parameters of the graph neural network. For the bottleneck identification method, when the turning point method is selected, the user is allowed to set the time window size and the state change threshold; when the utilization method is selected, the user is allowed to select the utilization calculation cycle and the bottleneck determination threshold. These settings enable the system to flexibly adjust the complexity and calculation accuracy of the network, capture the spatial and temporal interactions between workstations and process flows in the production line, and thus improve the bottleneck prediction and performance analysis effects of multi-process production.

[0045] (3) Performance analysis and monitoring module

[0046] like Figure 4As shown in the figure, the performance analysis and monitoring module includes the production line selection submodule, the global visualization submodule, the local visualization submodule, and the decision support and scheduling optimization submodule. The submodules are closely linked to provide users with a comprehensive and dynamic production line performance analysis tool. After the user selects the target production line in the production line selection submodule, the system will automatically trigger the extraction and display of relevant performance data, allowing users to intuitively and in real time grasp the key performance indicators and operating status of the selected production line.

[0047] Specifically, after the user selects the target production line from the preset production line list through the production line selection submodule, the system will immediately retrieve and load the performance data file related to the target production line from the corresponding database of the data storage submodule. The performance data file includes but is not limited to overall analysis data, equipment performance data, order execution status, etc. After the production line selection is completed, the system automatically activates the global visualization submodule and the local visualization submodule, and generates corresponding visualization charts based on the data of the selected production line, so that the user can fully understand the overall and detailed performance of the production line.

[0048] The global visualization submodule generates and displays the overall performance data of the selected production line by receiving instructions from the production line selection submodule. Specifically, the system automatically generates dynamic charts of throughput curves, work-in-process quantity curves, and order completion progress (the system continuously obtains the latest production data from the production line through a real-time data stream update mechanism, and with the help of the dynamic response and real-time rendering functions of the ECharts open source visualization library, these data are automatically converted into intuitive visualization charts. When the system collects new production data and equipment status data, the chart will automatically refresh and display the latest production status). The throughput curve is used to reflect the production efficiency of the production line in different time periods; the work-in-process quantity curve shows the number of materials in the buffer between each station on the production line and the number of semi-finished products being processed; the order completion progress presents the execution of the current production task in percentage form. Through global visualization, users can fully grasp the overall operating status of the selected production line and promptly identify possible bottlenecks or inefficiencies in the production process, thereby providing the necessary data basis for the analysis of the decision support and scheduling optimization submodules.

[0049] The multi-line comparative analysis function of the production line selection submodule has significant advantages. Users can view the key performance indicators of multiple production lines in the same interface, such as the capacity utilization rate, equipment failure rate, and order completion status of each production line. Through multi-dimensional data comparison, users can quickly identify the performance differences between production lines and optimize the production line configuration based on this information. For example, when the equipment failure rate of a production line is significantly higher than that of other production lines, users can further retrieve relevant data files to analyze the cause of the failure through the comparative analysis results provided by the system, and formulate corresponding improvement measures to improve the overall operating efficiency of the production line. Such data analysis results will be directly input into the decision support and scheduling optimization submodule to automatically generate corresponding optimization suggestions.

[0050] Among them, the multi-line comparison function of the production line selection submodule is realized through flexible data integration and visualization technology within the system. Specifically, this function supports users to select multiple production lines in the same interface through the production line selection submodule, and displays the corresponding key performance indicators such as capacity utilization, equipment failure rate and order completion status in multi-dimensional data tables or charts. The implementation of this function depends on the efficient data transmission mechanism of the data management module in the system, which can summarize the data of each production line in real time, and display and compare the performance indicators of multiple production lines side by side through the ECharts visualization library. Users can select any number of production lines in the interface, and the system will update the corresponding performance indicator comparison chart in real time according to the user's selection, thereby realizing the multi-line comparison analysis function. This function can effectively help users identify the performance differences between production lines, and then analyze the causes of failures and generate optimization suggestions.

[0051] The local visualization submodule automatically adjusts with the production line selection and provides detailed performance data at the equipment level of the selected production line. The display content of the local visualization submodule includes but is not limited to the operating time of each device, the number of failures, the proportion of workpiece types in the buffer zone, and the power status of the AGV. After the user selects the production line, the system automatically loads and displays the key performance indicators of each device in the production line to help the user have an in-depth understanding of the operating status of each device and ensure the smooth operation of the production line. The fine-grained data generated by the local visualization submodule not only provides users with accurate analysis at the equipment level, but also provides a basis for real-time monitoring and adjustment of production plans for the decision support and scheduling optimization submodules.

[0052] The decision support and scheduling optimization submodule provides intelligent decision suggestions and dynamic scheduling optimization solutions by comparing the results of the performance prediction module with the actual operation data. Based on the real-time data provided by the global visualization submodule and the local visualization submodule, the decision support and scheduling optimization submodule analyzes the deviation between production efficiency and expected values, and gives improvement suggestions, such as adjusting production plans, reallocating resources, or modifying process flows. In addition, the decision support and scheduling optimization submodule uses real-time data and historical trends to provide dynamic scheduling optimization suggestions to help managers flexibly respond to production changes and improve overall operational efficiency. Through close linkage with the production line selection submodule and the visualization submodule, the decision support and scheduling optimization submodule is able to generate effective optimization solutions in a timely manner based on data-driven, ensuring continuous improvement and efficient operation of the production line.

[0053] Specifically, the decision support and scheduling submodule automatically analyzes the prediction results of the performance prediction module and the actual production data, compares the deviation between the production efficiency and the expected value in real time, and generates intelligent decision suggestions and dynamic scheduling optimization solutions based on the analysis results. Based on the real-time production data and indicator data provided by the global visualization submodule, the local visualization submodule and the data management module, the decision support and scheduling optimization submodule first analyzes the actual operation status of each link in the production process (including equipment status, production plan execution status, order progress and other information). By comparing the real-time data with the historical prediction data, the decision support and scheduling optimization submodule can identify the bottlenecks in the current production process (the bottleneck identification submodule is based on the prediction results generated by historical data and prediction algorithms, identifies possible bottlenecks in the future, and focuses on prediction; the decision support and scheduling submodule compares the prediction results of the performance prediction module with the actual data to find the actual bottlenecks, focusing on the bottlenecks in the current production process), resource waste and efficiency deviations, and uses linear rules to identify bottlenecks in the current production process. The planning algorithm generates specific optimization plans, including but not limited to the following types of suggestions: production plan adjustment (based on production schedule deviations, the system will suggest adjusting the order or priority of the production plan to avoid bottlenecks blocking the overall production), resource reallocation (based on production line load analysis, the system suggests reallocating personnel, equipment or material resources to better balance the operating load of the production line and reduce idleness or overload), process flow optimization (by comparing the actual time consumption of each process stage with the standard expectation, the system identifies inefficient process steps and suggests adjusting process parameters or switching to a more efficient operation mode), equipment maintenance and scheduling (based on equipment status data (such as operating time, failure rate), the system can predict equipment failure or fatigue, and recommends arranging maintenance in advance to avoid production interruptions). Through close linkage with the production line selection submodule and the visualization submodule, the decision support and scheduling optimization submodule is able to generate effective optimization plans in a timely manner based on data-driven, helping managers to flexibly respond to production changes, improve overall operational efficiency, and ensure continuous improvement and efficient operation of the production line.

[0054] In summary, according to an embodiment of the present invention, a smart workshop data collection and analysis system is provided, which includes a data management module, a performance prediction module and a performance analysis and monitoring module. The data management module adopts a multi-source collection strategy and a multi-level network architecture to achieve comprehensive data collection and efficient transmission from the device layer to the controller layer, and meets the large-scale, high-speed and diversified data storage requirements through a hybrid storage architecture. The performance prediction module integrates three sub-modules: parameter selection, indicator prediction visualization and bottleneck identification, to achieve comprehensive analysis and accurate prediction of production line performance. The performance analysis and monitoring module supports the selection and comparative analysis of a single or multiple production lines, and provides managers with a full range of production line analysis tools from macro to micro through visualization at both global and local levels. In addition, the performance analysis and monitoring module also integrates decision support and scheduling optimization functions, and provides managers with intelligent decision-making suggestions and dynamic scheduling solutions by integrating prediction results with actual operation data. This integrated approach helps managers quickly identify production bottlenecks, optimize resource allocation, and monitor detailed performance indicators of each device to ensure efficient operation of the production line. Through real-time data processing, intelligent algorithm application and intuitive visual display, the system has significantly improved the operational efficiency, decision-making quality and problem diagnosis capabilities of smart workshops, providing strong technical support for the intelligent transformation of the manufacturing industry.

[0055] It should be noted that the above embodiments are only for explaining the technical principles of the present invention, and are not intended to limit the protection scope of the present invention. Although the present invention has been described in detail through specific embodiments, it should be understood by those skilled in the art that various deformations, modifications or equivalent substitutions may be made to the technical solutions without departing from the core idea of ​​the present invention. These deformations, modifications or equivalent substitutions shall also be regarded as the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the content of the claims.

Claims

1. An intelligent workshop data collection and analysis system, characterized in that: include: The data management module includes a data acquisition submodule, a data transmission submodule and a data storage submodule. The data acquisition submodule is used to collect factory production data of multiple production lines in the target workshop within multiple historical target cycles and order data and equipment information data within the current target cycle. The data transmission submodule adopts a multi-level network architecture to map the data management module with the performance prediction module, the data management module and the performance analysis and monitoring module. The data storage submodule is used to store all data generated by the data management module and the performance prediction module; The performance prediction module includes a parameter selection submodule, an indicator prediction visualization submodule and a bottleneck identification submodule. The parameter selection submodule is used to allow the user to select any one or more of the production line name, performance prediction type, prediction algorithm and bottleneck identification method through an interactive interface, and to predict the performance of the production line equipment by accessing the historical production data and current real-time production data stored in the data management module in combination with the algorithm selected by the user, and to generate relevant prediction results. The indicator prediction visualization submodule is used to update the display content, and the bottleneck identification submodule is used to identify the bottleneck by evaluating the prediction results; The performance analysis and monitoring module includes a production line selection submodule, a global visualization submodule, a local visualization submodule, and a decision support and scheduling optimization submodule. The production line selection submodule is used to enable users to select the target production line from a preset production line list. The global visualization submodule is used to generate and display the overall performance data of the selected production line. The local visualization submodule is used to make adjustments based on the production line selection and provide detailed performance data at each equipment level of the selected production line. The decision support and scheduling optimization submodule is used to provide decision suggestions and dynamic scheduling optimization solutions by comparing the results of the performance prediction module with the actual operation data.

2. The intelligent workshop data collection and analysis system according to claim 1 is characterized in that: The data acquisition submodule adopts diversified acquisition strategies for data acquisition, including but not limited to: real-time monitoring of the equipment's operating status through vibration sensors, recording equipment energy consumption data through power sensors, tracking AGV transportation paths and material locations through RFID readers and displacement sensors, real-time acquisition of robot arm motion parameters and operating accuracy data through torque sensors and position encoders, and configuration of high-speed industrial cameras to capture workpiece processing details.

3. The intelligent workshop data collection and analysis system according to claim 1 is characterized in that: After the data management module collects data through the data acquisition submodule, it pre-processes and filters the collected factory production line data and production equipment operation status data.

4. The intelligent workshop data collection and analysis system according to claim 1 is characterized in that: The multi-level network structure of the data transmission submodule includes: at the device level, the industrial fieldbus technology Profibus is used to achieve real-time communication between sensors and local controllers; at the controller level, the industrial Ethernet technology EtherNet / IP is used; and an industrial wireless network is deployed on the AGV.

5. The intelligent workshop data collection and analysis system according to claim 1 is characterized in that: The data storage submodule adopts a layered storage strategy. In terms of the hardware layer, the data storage submodule adopts a hybrid storage architecture. In terms of the database layer, the data storage submodule adopts a combined architecture of multiple types of databases. In terms of the cache layer, the data storage submodule introduces the Redis in-memory database to cache frequently accessed hot data.

6. The intelligent workshop data collection and analysis system according to claim 1 is characterized in that: The performance prediction type includes two candidate quantities: bottleneck and duration. Bottleneck prediction aims to identify and predict possible bottlenecks in the production line, and duration prediction aims to estimate the time required to complete a specific production task or order.

7. The intelligent workshop data collection and analysis system according to claim 1 is characterized in that: The prediction algorithm includes but is not limited to any one of an autoregressive integrated moving average model and a long short-term memory network.

8. The intelligent workshop data collection and analysis system according to claim 1, characterized in that: The bottleneck identification method includes a turning point method and a utilization rate method.

9. The intelligent workshop data collection and analysis system according to claim 1, characterized in that: The decision support and scheduling submodule analyzes the prediction results of the performance prediction module and the actual production data, compares the deviation between the production efficiency and the expected value in real time, and generates intelligent decision suggestions and dynamic scheduling optimization solutions based on the analysis results.

10. An intelligent workshop data collection and analysis system according to any one of claims 1 to 9, characterized in that: The decision support and scheduling optimization submodule is based on the real-time production data and indicator data provided by the global visualization submodule, the local visualization submodule and the data management module. By comparing the real-time data with the historical forecast data, it identifies the actual bottlenecks, resource waste and efficiency deviations in the current production process, and uses the linear programming algorithm to generate a specific optimization plan.

Images