Intelligent manufacturing process management system based on big data

Abstract

The invention relates to the technical field of intelligent manufacturing, and discloses an intelligent manufacturing process management system based on big data, and the system comprises a data collection module which is used for obtaining the real-time data of a production workshop through deploying a sensor, industrial communication equipment and video collection equipment, and the collected data comprises the equipment operation state, environment parameters and production process feature data; the collected multi-source data is sent to the data fusion and processing module; and the data fusion and processing module is used for receiving the collected multi-source heterogeneous data and executing data cleaning, formatting and standardization processing. Through close cooperation among the modules, full-process data closed-loop management from data acquisition, data processing, intelligent optimization to equipment dynamic adjustment is realized, key parameters of a production process can be updated in real time, the dynamic adjustment capability and the execution effect of an optimization scheme are ensured, and the accuracy and the intelligent level of production management are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing technology, and in particular to an intelligent manufacturing process management system based on big data. Background Art

[0002] With the introduction of the concept of intelligent manufacturing, the manufacturing industry is transforming from traditional manual management and experience-based decision-making to data-driven and intelligent. The application of big data technology, artificial intelligence algorithms and industrial Internet of Things provides new development momentum for the manufacturing industry. By collecting, processing and analyzing the data generated during the production process, enterprises can achieve refined management of the entire production process. However, there are still problems and defects in the existing technology.

[0003] Most manufacturing companies have deployed various sensors and industrial communication equipment, but there is still a lack of systematic solutions for the collection, cleaning and standardization of multi-source heterogeneous data. The data formats from different sources are not unified and the time steps are not synchronized, which makes data integration and fusion difficult, affecting the accuracy and effectiveness of subsequent data analysis. Traditional production scheduling algorithms are usually based on static models and cannot respond to changes in task requirements or equipment failures in real time. If orders are inserted or production equipment fails, the existing system cannot quickly adjust task allocation and production resources, resulting in reduced production efficiency and waste of resources. Existing anomaly detection technologies mostly rely on simple rules or threshold monitoring. When faced with complex equipment operating conditions, the detection accuracy is low and early warning cannot be achieved. When an anomaly is detected, the existing system cannot locate the specific cause of the anomaly and needs to be manually checked one by one, which increases the difficulty of equipment maintenance and downtime; There is a high risk in directly applying the optimization strategy to actual production equipment. Traditional production systems lack the ability to verify the optimization plan, resulting in incorrect optimization strategies causing equipment damage or interruption of the production process, increasing the company's trial and error costs and production risks. The visualization functions of most existing systems are relatively simple, mainly staying at the data query level, lacking intuitive display of optimization results and abnormal conditions. At the same time, users lack effective means of interaction after noticing problems, and are unable to intervene and optimize the production process in a timely manner.

[0004] Therefore, those skilled in the art provide an intelligent manufacturing process management system based on big data to solve the above-mentioned problems. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides an intelligent manufacturing process management system based on big data to solve the problems raised in the above background technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: A big data-based intelligent manufacturing process management system, comprising: The data acquisition module is used to acquire real-time data of the production workshop by deploying sensors, industrial communication equipment, and video acquisition equipment. The collected data includes equipment operation status, environmental parameters, and production process characteristic data. The collected multi-source data is sent to the data fusion and processing module. The data fusion and processing module is used to receive the collected multi-source heterogeneous data, perform data cleaning, formatting and standardization processing, calculate and store the received data using real-time stream processing technology, and generate data streams that can be called by the intelligent optimization and prediction modules; Intelligent optimization and prediction module, which is used to call processed standardized data, realize dynamic scheduling of production tasks, detection of equipment operation anomalies and prediction of process energy consumption based on big data analysis and artificial intelligence algorithms, and send the generated optimization results to the dynamic response and control module; The dynamic response and control module is used to receive the optimization results generated by the intelligent optimization and prediction module, verify the optimization strategy by building a digital twin model, and adjust the production equipment parameters according to the verified strategy. The adjusted feedback information is returned to the data acquisition module for updating; The application display and visualization module is used to read the real-time processed data and the results of optimization analysis, generate a visual production process display interface and a user interaction interface, provide an operation interface for users and pass relevant instructions to the intelligent optimization and prediction module.

[0007] Preferably, the data acquisition module includes: Sensor unit, used to collect equipment operating status data , the collected data is transmitted to the video acquisition unit and the industrial communication unit for supplementation or correlation verification. The operating status data formula is as follows: , in, is the operating parameter of the i-th device at time t, It is the equipment operation status data; The video acquisition unit is used to receive the equipment operation status data collected by the sensor unit, and extract the production feature data through computer vision to identify the status of materials, equipment or personnel. The video feature extraction formula is as follows: , in, Represents the input video frame image, Indicates detected materials, equipment or personnel; The industrial communication unit is used to receive the output data of the sensor unit and the video acquisition unit, and uniformly fuse the equipment operation status data and the video acquisition feature data, and transmit them to the data fusion and processing module.

[0008] Preferably, the data fusion and processing module includes: The data cleaning unit is used to receive multi-source data from the data acquisition module, filter out the noise in the data, and generate cleaned data. ; The data cleaning formula is as follows: , in, is the original data, is the noise signal; The time series alignment unit is used to align the cleaned asynchronous data and generate the aligned time series data through interpolation calculation. , the time series alignment formula is as follows: , in, , is the time step, and At time step and The cleaned data, is the interpolation calculation function; The data formatting unit is used to standardize and convert the aligned data into a unified format to generate formatted data. The formatting formula is as follows: , in, is the data normalization function, For time series data, To format the data.

[0009] Preferably, the intelligent optimization and prediction module includes: Dynamic production scheduling unit, which receives cleaned, aligned and formatted production data and optimizes the production tasks based on reinforcement learning algorithms; The dynamic scheduling unit optimizes the scheduling and resource allocation of production tasks through the objective function: , in, is the maximum completion time of all production tasks, is the total energy consumption, is the energy consumption weight factor; , , in, represents the completion time of the i-th task, represents the power consumption of the i-th device at time t, n is the total number of tasks, and T is the total length of time; The task constraint unit is used to define and apply the following constraints during the optimization process to ensure the feasibility of the scheduling plan: Time constraints between tasks: , if task i and task j use the same device, in, is the start time of task j, is the completion time of task i; Equipment load constraints: , in, is the current load of the kth device at time t, is the maximum load capacity of the kth device.

[0010] Preferably, the intelligent optimization and prediction module includes: The anomaly detection unit is used to receive the cleaned and aligned time series data, predict the equipment operation status and detect anomalies, and detect anomalies through the following time series prediction models: , , in, Indicates the operating status of the device in the next time step predicted based on the data of the previous n time steps; represents the prediction function; Represents the time series data cleaned and aligned by the data fusion and processing module; Represents the anomaly score, which is used to measure the prediction value and actual value Deviation between Anomaly score analysis unit, used to receive anomaly scores And determine whether there is an abnormality based on the set abnormality threshold , if the following conditions are met, it is considered abnormal: , in, represents the abnormal threshold, like Exceed When an abnormality occurs, the abnormality information is sent to the dynamic response and control module for response; The root cause analysis unit is used to further analyze the root cause of the anomaly when it occurs. It infers the relationship between device parameters by constructing a knowledge graph. The mathematical description of the knowledge graph is: , in, For the knowledge graph, is the node set in the knowledge graph, is the edge set in the knowledge graph.

[0011] Preferably, the dynamic response and control module includes: The digital twin unit is used to receive the optimized task scheduling and equipment parameter adjustment strategies, build a virtual production environment, and simulate the effect of the optimization strategy in actual production; The digital twin unit builds a virtual environment through the following models: , in, For the digital twin model, For real-time device data, is the simulation model data, Build functions for digital twins; The optimization verification unit is used to verify the effectiveness of the optimization strategy in the virtual production environment and determine whether to implement the strategy by evaluating the impact of the optimization strategy on the production target; The optimization verification formula is as follows: , in, To optimize the performance evaluation value of strategy O, is the performance evaluation function, is the simulation data based on the optimization strategy O; A parameter control unit, used to receive the strategy confirmed by the optimization verification unit and generate an instruction to adjust the equipment operation parameters; The parameter adjustment is determined by the following formula: , in, Indicates the adjustment value of the equipment power or load. represents the control function, It is the current real-time operation data of the device.

[0012] Preferably, the application display and visualization module includes: A data visualization unit is used to receive processed real-time data and optimize analysis results and generate a visual display interface; The user interaction unit is used to provide a user operation interface to support users in querying, filtering and operating the visualized content.

[0013] Preferably, the specific functions of the user interaction unit include: Task priority adjustment: Users can manually intervene in the production scheduling strategy by adjusting the task priority according to the production scheduling plan displayed by the data visualization unit; Abnormal alarm management: Users can confirm the abnormal cause through the Web or mobile terminal according to the displayed abnormal alarm status and generate processing instructions; Control feedback function: The user's operation instructions are transmitted to the intelligent optimization and prediction module through the user interaction unit. The optimization module readjusts the production scheduling plan or exception handling strategy based on the user input.

[0014] Preferably, the data acquisition module further includes an edge computing unit for performing real-time processing and pre-processing on the data after the sensor unit, the video acquisition unit and the industrial communication unit collect the data.

[0015] Preferably, the functions of the edge computing unit include: Data aggregation: Downsample the high-frequency sensor data to generate low-frequency representative values. The data aggregation formula is as follows: , in, is the aggregated data, is the data of the i-th sampling point, is the total number of sampling points; Fast anomaly detection: Perform simple statistical analysis on the data locally to quickly detect anomalies using the following formula: ,like , it is considered that there is an abnormality. in, is the data value of the current time step, is the mean of the data, is the standard deviation of the data, is the threshold for anomaly detection; The data processed by the edge computing unit is transmitted to the data fusion and processing module through the industrial communication unit.

[0016] The present invention provides an intelligent manufacturing process management system based on big data. It has the following beneficial effects: 1. The present invention realizes closed-loop management of the entire process of data collection, data processing, intelligent optimization and dynamic adjustment of equipment through close collaboration between modules. It can update the key parameters of the production process in real time, ensure the dynamic adjustment capability and execution effect of the optimization plan, and significantly improve the accuracy and intelligence level of production management.

[0017] 2. The present invention introduces digital twin technology, simulates the operating effect of the optimization strategy in actual production through a virtual environment, and implements virtual verification before implementation, which effectively reduces the trial and error cost and production risk, ensures the feasibility and effectiveness of the optimization plan, and provides strong protection for the decision-making security in the intelligent manufacturing process.

[0018] 3. The dynamic task scheduling algorithm based on reinforcement learning of the present invention can intelligently allocate production tasks in combination with real-time data. The scheduling optimization method has stronger dynamic adaptability and flexibility, and can quickly re-plan resource allocation in the event of task changes or equipment abnormalities, thereby maximizing production efficiency.

[0019] 4. The present invention achieves high-precision detection and root cause analysis of equipment anomalies through the combination of time series prediction models and knowledge graphs, and can provide early warning before equipment failure occurs. Root cause analysis can quickly locate the cause of the anomaly, provide direct guidance for production maintenance, reduce downtime caused by equipment failure, and improve the stability and reliability of the production line. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a system framework diagram of the present invention; Figure 2 It is a functional schematic diagram of the user interaction unit of the present invention. DETAILED DESCRIPTION

[0021] In order to make the technical personnel in the technical field understand the scheme of the present invention, the technical scheme in the embodiment of the present invention will be clearly and completely described below in combination with the drawings in the embodiment of the present invention. Obviously, the described embodiment is a partial embodiment of the present invention, not a complete embodiment. Based on the embodiment of the present invention, other embodiments obtained by ordinary technicians in the field without creative work should fall within the scope of protection of the present invention.

[0022] The present invention is described in detail below in conjunction with the accompanying drawings: Example: Please refer to the attached Figure 1 and attached Figure 2 The embodiment of the present invention provides an intelligent manufacturing process management system based on big data, including: The data acquisition module is used to acquire real-time data of the production workshop by deploying sensors, industrial communication equipment, and video acquisition equipment. The collected data includes equipment operation status, environmental parameters, and production process characteristic data. The collected multi-source data is sent to the data fusion and processing module. The data fusion and processing module is used to receive the collected multi-source heterogeneous data, perform data cleaning, formatting and standardization processing, calculate and store the received data using real-time stream processing technology, and generate data streams that can be called by the intelligent optimization and prediction modules; Intelligent optimization and prediction module, which is used to call processed standardized data, realize dynamic scheduling of production tasks, detection of equipment operation anomalies and prediction of process energy consumption based on big data analysis and artificial intelligence algorithms, and send the generated optimization results to the dynamic response and control module; The dynamic response and control module is used to receive the optimization results generated by the intelligent optimization and prediction module, verify the optimization strategy by building a digital twin model, and adjust the production equipment parameters according to the verified strategy. The adjusted feedback information is returned to the data acquisition module for updating; The application display and visualization module is used to read the real-time processed data and the results of optimization analysis, generate a visual production process display interface and a user interaction interface, provide an operation interface for users and pass relevant instructions to the intelligent optimization and prediction module.

[0023] The data acquisition module deploys sensors, industrial communication equipment and video acquisition equipment to comprehensively acquire equipment operating status, environmental parameters and production process characteristic data, ensuring that key data points of the entire production process are monitored in real time. The high-frequency acquisition and stable transmission of multi-source data provide a comprehensive and reliable foundation for subsequent data processing and optimization. The data fusion and processing module cleans, formats, and aligns time series of multi-source heterogeneous data, unifies data formats, eliminates noise and redundant information in data, and improves data quality. At the same time, it uses real-time stream processing technology to achieve second-level computing and storage, providing high-quality data support for dynamic optimization and real-time anomaly detection. The intelligent optimization and prediction module uses standardized data to achieve dynamic scheduling of production tasks, high-precision detection of equipment anomalies, and prediction and optimization of energy consumption based on big data analysis and artificial intelligence algorithms, thereby improving production efficiency, reducing equipment downtime and energy consumption costs, and providing technical support for the realization of green manufacturing; The dynamic response and control module verifies the optimization strategy in a virtual environment through digital twin technology to ensure the feasibility and safety of the strategy. At the same time, it optimizes the production process by adjusting the production equipment parameters in real time, forming a data closed loop, and significantly improving the dynamic adaptability and continuous optimization capabilities of the system. The application display and visualization module displays key production performance indicators and optimization results through an intuitive visualization interface, and provides users with interactive functions such as task priority adjustment and abnormal alarm management, helping users to quickly understand the production status and make decisions, further enhancing the system's human-machine collaboration capabilities and operational convenience.

[0024] The data acquisition module includes: Sensor unit, used to collect equipment operating status data , the collected data is transmitted to the video acquisition unit and the industrial communication unit for supplementation or correlation verification. The operating status data formula is as follows: , in, is the operating parameter of the i-th device at time t, It is the equipment operation status data; The video acquisition unit is used to receive the equipment operation status data collected by the sensor unit, and extract the production feature data through computer vision to identify the status of materials, equipment or personnel. The video feature extraction formula is as follows: , in, Represents the input video frame image, Indicates detected materials, equipment or personnel; The industrial communication unit is used to receive the output data of the sensor unit and the video acquisition unit, and uniformly fuse the equipment operation status data and the video acquisition feature data, and transmit them to the data fusion and processing module; The data acquisition module further includes an edge computing unit for real-time processing and preprocessing of data after the sensor unit, the video acquisition unit, and the industrial communication unit acquire the data; The functions of the edge computing unit include: Data aggregation: Downsample the high-frequency sensor data to generate low-frequency representative values. The data aggregation formula is as follows: , in, is the aggregated data, is the data of the i-th sampling point, is the total number of sampling points; Fast anomaly detection: Perform simple statistical analysis on the data locally to quickly detect anomalies using the following formula: ,like , it is considered that there is an abnormality. in, is the data value of the current time step, is the mean of the data, is the standard deviation of the data, is the threshold for anomaly detection; The data processed by the edge computing unit is transmitted to the data fusion and processing module through the industrial communication unit.

[0025] The sensor unit realizes high-precision monitoring of the equipment's operating status, including collecting key information such as equipment operating parameters and environmental data. It also enhances the integrity and reliability of data through collaboration with the video acquisition unit and industrial communication unit, providing an accurate data basis for subsequent analysis and optimization. The video acquisition unit uses computer vision technology to identify and monitor the status of equipment, materials and personnel in real time, which can capture dynamic changes in the production process, supplement features that cannot be captured by sensor data, and further improve the visualization and comprehensiveness of the production process; The industrial communication unit effectively integrates the output data of the sensor and video acquisition unit, uniformly processes the multi-source data and transmits it to the data fusion and processing module, solves the problem of heterogeneous data fusion, and ensures efficient data transmission and seamless connection in complex industrial environments; The edge computing unit reduces the redundancy of high-frequency data through data aggregation function and optimizes the bandwidth requirements for data transmission. At the same time, it realizes rapid anomaly detection locally, can discover potential problems in production in advance, reduce the response time of equipment and systems, and improve the real-time and intelligence level of the data acquisition module.

[0026] The data fusion and processing modules include: The data cleaning unit is used to receive multi-source data from the data acquisition module, filter out the noise in the data, and generate cleaned data. ; The data cleaning formula is as follows: , in, is the original data, is the noise signal; The time series alignment unit is used to align the cleaned asynchronous data and generate the aligned time series data through interpolation calculation. , the time series alignment formula is as follows: , in, , is the time step, and At time step and The cleaned data, is the interpolation calculation function; The data formatting unit is used to standardize and convert the aligned data into a unified format to generate formatted data. The formatting formula is as follows: , in, is the data normalization function, For time series data, To format the data.

[0027] The data cleaning unit filters out noise from multi-source data from the data acquisition module, effectively improving data quality, eliminating redundancy and errors in the data, providing a reliable data basis for subsequent time series alignment and formatting, and ensuring the accuracy and reliability of data processing; The time series alignment unit aligns the cleaned asynchronous data through interpolation calculation to solve the problem of asynchrony of multi-source data in time steps, making the data consistent in the time dimension and providing accurate data support for dynamic analysis and real-time optimization; The data formatting unit standardizes and converts the aligned data into a unified format, eliminating the heterogeneity of the data format and ensuring the structuring and normalization of the data, so as to facilitate direct calling by subsequent intelligent optimization modules and improve analysis efficiency.

[0028] Intelligent optimization and prediction modules include: Dynamic production scheduling unit, which receives cleaned, aligned and formatted production data and optimizes the production tasks based on reinforcement learning algorithms; The dynamic scheduling unit optimizes the scheduling and resource allocation of production tasks through the objective function: , in, is the maximum completion time of all production tasks, is the total energy consumption, is the energy consumption weight factor; , , in, represents the completion time of the i-th task, represents the power consumption of the i-th device at time t, n is the total number of tasks, and T is the total length of time; The task constraint unit is used to define and apply the following constraints during the optimization process to ensure the feasibility of the scheduling plan: Time constraints between tasks: , if task i and task j use the same device, in, is the start time of task j, is the completion time of task i; Equipment load constraints: , in, is the current load of the kth device at time t, is the maximum load capacity of the kth device.

[0029] The dynamic production scheduling unit optimizes and schedules production tasks in real time through reinforcement learning algorithms. It can dynamically adjust task allocation plans based on equipment operating status, task priority, and resource usage. It comprehensively considers task completion time and minimization of energy consumption through objective functions, significantly improving production efficiency and resource utilization. At the same time, it reduces production costs and adapts to dynamic demand changes in complex manufacturing environments. The task constraint unit applies time constraints and equipment load constraints during the optimization process to ensure the feasibility of the production schedule and the safety of equipment operation. Time constraints can ensure the reasonable connection of tasks and avoid resource conflicts; equipment load constraints prevent equipment from overloading, improve the service life and operating stability of the equipment, and provide more reliable scheduling guarantees for the production process.

[0030] Intelligent optimization and prediction modules include: The anomaly detection unit is used to receive the cleaned and aligned time series data, predict the equipment operation status and detect anomalies, and detect anomalies through the following time series prediction models: , , in, Indicates the operating status of the device in the next time step predicted based on the data of the previous n time steps; represents the prediction function; Represents the time series data cleaned and aligned by the data fusion and processing module; Represents the anomaly score, which is used to measure the prediction value and actual value Deviation between Anomaly score analysis unit, used to receive anomaly scores And determine whether there is an abnormality based on the set abnormality threshold , if the following conditions are met, it is considered abnormal: , in, represents the abnormal threshold, like Exceed When an abnormality occurs, the abnormality information is sent to the dynamic response and control module for response; The root cause analysis unit is used to further analyze the root cause of the anomaly when it occurs. It infers the relationship between device parameters by constructing a knowledge graph. The mathematical description of the knowledge graph is: , in, For the knowledge graph, is the node set in the knowledge graph, is the edge set in the knowledge graph.

[0031] The anomaly detection unit predicts and monitors the operating status of the equipment in real time through the time series prediction model. It can accurately measure the deviation between the predicted value and the actual value, promptly discover abnormal trends in equipment operation, and provide early warning before equipment failure occurs, effectively reducing production interruption time caused by equipment failure and improving the continuity and reliability of the production line. The abnormality scoring analysis unit combines the abnormality score and the preset threshold to accurately determine whether the equipment is in an abnormal state. When an abnormality is found, the abnormal information is quickly transmitted to the dynamic response and control module for processing, which can significantly shorten the time from abnormality detection to response time, improve the system's rapid response capability, and reduce the impact of abnormalities on the production process; The root cause analysis unit constructs a knowledge graph and performs associative reasoning on the relationship between equipment parameters. It can quickly locate the root cause of the anomaly, help maintenance personnel accurately determine the problematic equipment and key parameters, reduce the time and labor costs of anomaly troubleshooting, optimize the fault handling process, and improve the efficiency and accuracy of production maintenance.

[0032] The dynamic response and control module includes: The digital twin unit is used to receive the optimized task scheduling and equipment parameter adjustment strategies, build a virtual production environment, and simulate the effect of the optimization strategy in actual production; The digital twin unit builds a virtual environment through the following models: , in, For the digital twin model, For real-time device data, is the simulation model data, Build functions for digital twins; The optimization verification unit is used to verify the effectiveness of the optimization strategy in the virtual production environment and determine whether to implement the strategy by evaluating the impact of the optimization strategy on the production target; The optimization verification formula is as follows: , in, To optimize the performance evaluation value of strategy O, is the performance evaluation function, is the simulation data based on the optimization strategy O; A parameter control unit, used to receive the strategy confirmed by the optimization verification unit and generate an instruction to adjust the equipment operation parameters; The parameter adjustment is determined by the following formula: , in, Indicates the adjustment value of the equipment power or load. represents the control function, It is the current real-time operation data of the device.

[0033] The digital twin unit combines real-time equipment data and simulation models by building a virtual production environment to achieve simulation verification of the optimization strategy, effectively reducing the trial and error costs and risks when the strategy is directly applied to actual equipment, ensuring the feasibility and safety of the optimization strategy in actual production, and providing strong guarantees for production decisions; The optimization verification unit evaluates the performance of the optimization strategy in a virtual production environment, determines whether to implement the strategy by quantifying the impact of the optimization plan on the production target, and selects the optimal scheduling and equipment adjustment plan to avoid interference with the production process caused by unreasonable strategies, thereby improving the optimization effect of the system and the scientific nature of decision-making; The parameter control unit generates specific equipment operating parameter adjustment instructions based on the strategy confirmed by the optimization verification unit, and dynamically adjusts key parameters such as power and load in combination with real-time equipment operation data to ensure that the equipment always operates in the best state, improve the operating efficiency of the production line and the service life of the equipment, and at the same time enhance the system's adaptability to changing environments.

[0034] Application display and visualization modules include: A data visualization unit is used to receive processed real-time data and optimize analysis results and generate a visual display interface; A user interaction unit is used to provide a user operation interface to support users in querying, filtering and operating the visualized content; The specific functions of the user interaction unit include: Task priority adjustment: Users can manually intervene in the production scheduling strategy by adjusting the task priority according to the production scheduling plan displayed by the data visualization unit; Abnormal alarm management: Users can confirm the abnormal cause through the Web or mobile terminal according to the displayed abnormal alarm status and generate processing instructions; Control feedback function: The user's operation instructions are transmitted to the intelligent optimization and prediction module through the user interaction unit. The optimization module readjusts the production scheduling plan or exception handling strategy based on the user input.

[0035] The data visualization unit generates an intuitive graphical interface for real-time production data and optimization analysis results, helping users quickly understand key performance indicators in the production process, improving their ability to understand complex data, facilitating rapid problem identification and decision-making, and enhancing the transparency of production management; The user interaction unit provides users with a convenient operation interface and supports real-time intervention in production scheduling and exception management. Users can directly adjust task priorities, query exception status and generate processing instructions through the Web or mobile terminal. At the same time, the operation is fed back to the intelligent optimization module for optimization and adjustment, fully realizing the ability of human-machine collaborative optimization and meeting the flexible scheduling needs in a dynamic production environment. Users can adjust task priorities according to the production scheduling plan displayed on the data visualization interface to make the production scheduling strategy more in line with actual needs. In the case of sudden order insertion or task changes, the task priority adjustment function can respond quickly, improving the flexibility and real-time performance of production scheduling; The abnormal alarm management function helps users quickly understand the cause of the abnormality and take treatment measures by displaying the abnormal alarm status. Users can directly generate processing instructions and feed back abnormal information to the optimization module, shortening the time from abnormality discovery to response processing, and reducing downtime and economic losses caused by abnormalities. The control feedback function links the user's operating instructions with the optimization module to ensure that the user's adjustment needs can be incorporated into the optimization strategy in real time. On the basis of the optimization algorithm, the system further combines the user's experience and actual needs to improve the accuracy and applicability of the optimization plan.

[0036] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent manufacturing process management system based on big data, characterized in that: include: The data acquisition module is used to acquire real-time data of the production workshop by deploying sensors, industrial communication equipment, and video acquisition equipment. The collected data includes equipment operation status, environmental parameters, and production process characteristic data. The collected multi-source data is sent to the data fusion and processing module. The data fusion and processing module is used to receive the collected multi-source heterogeneous data, perform data cleaning, formatting and standardization processing, calculate and store the received data using real-time stream processing technology, and generate data streams that can be called by the intelligent optimization and prediction modules; Intelligent optimization and prediction module, which is used to call processed standardized data, realize dynamic scheduling of production tasks, detection of equipment operation anomalies and prediction of process energy consumption based on big data analysis and artificial intelligence algorithms, and send the generated optimization results to the dynamic response and control module; The dynamic response and control module is used to receive the optimization results generated by the intelligent optimization and prediction module, verify the optimization strategy by building a digital twin model, and adjust the production equipment parameters according to the verified strategy. The adjusted feedback information is returned to the data acquisition module for updating; The application display and visualization module is used to read the real-time processed data and the results of optimization analysis, generate a visual production process display interface and a user interaction interface, provide an operation interface for users and pass relevant instructions to the intelligent optimization and prediction module.

2. According to claim 1, a smart manufacturing process management system based on big data is characterized in that: The data acquisition module comprises: Sensor unit, used to collect equipment operating status data , the collected data is transmitted to the video acquisition unit and the industrial communication unit for supplementation or correlation verification. The operating status data formula is as follows: ， in, is the operating parameter of the i-th device at time t, It is the equipment operation status data; The video acquisition unit is used to receive the equipment operation status data collected by the sensor unit, and extract the production feature data through computer vision to identify the status of materials, equipment or personnel. The video feature extraction formula is as follows: ， in, Represents the input video frame image, Indicates detected materials, equipment or personnel; The industrial communication unit is used to receive the output data of the sensor unit and the video acquisition unit, and uniformly fuse the equipment operation status data and the video acquisition feature data, and transmit them to the data fusion and processing module.

3. According to claim 1, the intelligent manufacturing process management system based on big data is characterized in that: The data fusion and processing module includes: The data cleaning unit is used to receive multi-source data from the data acquisition module, filter out the noise in the data, and generate cleaned data. ; The data cleaning formula is as follows: , in, is the original data, is the noise signal; The time series alignment unit is used to align the cleaned asynchronous data and generate the aligned time series data through interpolation calculation. , the time series alignment formula is as follows: ， in, , is the time step, and At time step and The cleaned data, is the interpolation calculation function; The data formatting unit is used to standardize and convert the aligned data into a unified format to generate formatted data. The formatting formula is as follows: ， in, is the data normalization function, For time series data, To format the data.

4. According to claim 1, the intelligent manufacturing process management system based on big data is characterized in that: The intelligent optimization and prediction module includes: Dynamic production scheduling unit, which receives cleaned, aligned and formatted production data and optimizes the production tasks based on reinforcement learning algorithms; The dynamic scheduling unit optimizes the scheduling and resource allocation of production tasks through the objective function: ， in, is the maximum completion time of all production tasks, is the total energy consumption, is the energy consumption weight factor; ， ， in, represents the completion time of the i-th task, represents the power consumption of the i-th device at time t, n is the total number of tasks, and T is the total length of time; The task constraint unit is used to define and apply the following constraints during the optimization process to ensure the feasibility of the scheduling plan: Time constraints between tasks: , if task i and task j use the same device, in, is the start time of task j, is the completion time of task i; Equipment load constraints: ， in, is the current load of the kth device at time t, is the maximum load capacity of the kth device.

5. The intelligent manufacturing process management system based on big data according to claim 1 is characterized in that: The intelligent optimization and prediction module includes: The anomaly detection unit is used to receive the cleaned and aligned time series data, predict the equipment operation status and detect anomalies, and detect anomalies through the following time series prediction models: ， ， in, Indicates the operating status of the device in the next time step predicted based on the data of the previous n time steps; represents the prediction function; Represents the time series data cleaned and aligned by the data fusion and processing module; Represents the anomaly score, which is used to measure the prediction value and actual value Deviation between Anomaly score analysis unit, used to receive anomaly scores And determine whether there is an abnormality based on the set abnormality threshold , if the following conditions are met, it is considered abnormal: ， in, represents the abnormal threshold, like Exceed When an abnormality occurs, the abnormality information is sent to the dynamic response and control module for response; The root cause analysis unit is used to further analyze the root cause of the anomaly when it occurs. It infers the relationship between device parameters by constructing a knowledge graph. The mathematical description of the knowledge graph is: ， in, is the knowledge graph, is the node set in the knowledge graph, is the edge set in the knowledge graph.

6. The intelligent manufacturing process management system based on big data according to claim 1 is characterized in that: The dynamic response and control module includes: The digital twin unit is used to receive the optimized task scheduling and equipment parameter adjustment strategies, build a virtual production environment, and simulate the effect of the optimization strategy in actual production; The digital twin unit builds a virtual environment through the following models: ， in, For the digital twin model, For real-time device data, is the simulation model data, Build functions for digital twins; The optimization verification unit is used to verify the effectiveness of the optimization strategy in the virtual production environment and determine whether to implement the strategy by evaluating the impact of the optimization strategy on the production target; The optimization verification formula is as follows: , in, To optimize the performance evaluation value of strategy O, is the performance evaluation function, is the simulation data based on the optimization strategy O; A parameter control unit, used to receive the strategy confirmed by the optimization verification unit and generate an instruction to adjust the equipment operation parameters; The parameter adjustment is determined by the following formula: , in, Indicates the adjustment value of the equipment power or load. represents the control function, It is the current real-time operation data of the device.

7. The intelligent manufacturing process management system based on big data according to claim 1 is characterized in that: The application display and visualization module includes: A data visualization unit is used to receive processed real-time data and optimize analysis results and generate a visual display interface; The user interaction unit is used to provide a user operation interface to support users in querying, filtering and operating the visualized content.

8. The intelligent manufacturing process management system based on big data according to claim 7 is characterized in that: The specific functions of the user interaction unit include: Task priority adjustment: Users can manually intervene in the production scheduling strategy by adjusting the task priority according to the production scheduling plan displayed by the data visualization unit; Abnormal alarm management: Users can confirm the abnormal cause through the Web or mobile terminal according to the displayed abnormal alarm status and generate processing instructions; Control feedback function: The user's operation instructions are transmitted to the intelligent optimization and prediction module through the user interaction unit. The optimization module readjusts the production scheduling plan or exception handling strategy based on the user input.

9. The intelligent manufacturing process management system based on big data according to claim 2 is characterized in that: The data acquisition module further includes an edge computing unit, which is used to perform real-time processing and pre-processing on the data after the sensor unit, the video acquisition unit and the industrial communication unit collect the data.

10. The intelligent manufacturing process management system based on big data according to claim 9 is characterized in that: The functions of the edge computing unit include: Data aggregation: Downsample the high-frequency sensor data to generate low-frequency representative values. The data aggregation formula is as follows: ， in, is the aggregated data, is the data of the i-th sampling point, is the total number of sampling points; Fast anomaly detection: Perform simple statistical analysis on the data locally to quickly detect anomalies using the following formula: ,like , it is considered that there is an abnormality. in, is the data value of the current time step, is the mean of the data, is the standard deviation of the data, is the threshold for anomaly detection; The data processed by the edge computing unit is transmitted to the data fusion and processing module through the industrial communication unit.

Images