CN-122022269-A - Printing production intelligent scheduling system and method based on real-time data

CN122022269ACN 122022269 ACN122022269 ACN 122022269ACN-122022269-A

Abstract

The application provides an intelligent scheduling system and method for printing production based on real-time data, wherein the intelligent scheduling system comprises the steps of extracting specific shutdown requirement time points from real-time monitoring information if the occurrence probability of potential abnormal symptoms exceeds a preset threshold value, judging an upcoming equipment maintenance window period, simulating equipment load distribution conditions in a production flow through an optimized task execution sequence, predicting overall production continuity indexes such as a production line uninterrupted operation rate after task transfer by adopting a time sequence data processing network, determining a final production scheduling scheme according to feedback data such as equipment utilization rate of task transfer if the overall production continuity indexes are lower than a target threshold value, iteratively adjusting task allocation proportion, acquiring the final production scheduling scheme, fusing updated data such as sudden load change of real-time factors, continuously monitoring equipment shutdown risks, and obtaining a production efficiency optimization path based on closed loop feedback control.

Inventors

YU XIANG
Xi Guangqi

Assignees

湖北大河文化发展有限公司

Dates

Publication Date: 20260512
Application Date: 20251231

Claims (8)

1. An intelligent scheduling system and method for printing production based on real-time data, wherein the method comprises the following steps: acquiring real-time monitoring information such as temperature data, vibration data, consumable allowance data and the like in the operation of industrial equipment through a temperature sensor and a vibration sensor, and analyzing the real-time monitoring information by adopting a time sequence data processing network to obtain the change trend characteristics of the operation state of the equipment; According to the change trend characteristics of the running states of the equipment, the interaction influencing factors such as environmental temperature fluctuation and load change among the equipment are fused, a decision tree algorithm is adopted to analyze the correlation mode of temperature rise and vibration aggravation, and the occurrence probability of potential abnormal symptoms such as equipment failure precursors is determined; if the occurrence probability of the potential abnormal symptom exceeds a preset threshold value, extracting a specific shutdown requirement time point from the real-time monitoring information, and judging an upcoming equipment maintenance window period; Acquiring the equipment maintenance window period, adopting a dynamic adjustment mechanism such as task priority sequencing aiming at order tasks in the current production schedule, and redistributing the order tasks to other available industrial equipment to obtain an optimized task execution sequence; simulating equipment load distribution conditions in a production flow through the optimized task execution sequence, and predicting overall production continuity indexes such as the uninterrupted operation rate of a production line after task transfer by adopting a time sequence data processing network; If the overall production continuity index is lower than a target threshold, iteratively adjusting the task allocation proportion according to feedback data of task transfer such as equipment utilization rate, and determining a final production scheduling scheme; And acquiring the final production scheduling scheme, fusing updated data of real-time factors such as sudden load change, and continuously monitoring equipment shutdown risk to obtain a production efficiency optimization path based on closed loop feedback control.
2. The intelligent scheduling system and method for printing production based on real-time data according to claim 1, wherein the real-time monitoring information such as temperature data, vibration data and consumable residue data in the operation of industrial equipment is obtained through a temperature sensor and a vibration sensor, and the real-time monitoring information is analyzed by adopting a time sequence data processing network to obtain the change trend characteristics of the operation state of the equipment, and the method comprises the following steps: Acquiring temperature data and vibration data and consumable allowance data in the operation of industrial equipment through a temperature sensor and a vibration sensor, and acquiring operation information monitored in real time; processing the collected operation information by adopting a time sequence data processing network, and extracting state change characteristics related to the operation of equipment; according to the extracted state change characteristics, the characteristics are subjected to sequence analysis by using a long-short-term memory network, so that trend characteristics of the running state of the equipment are obtained; if the trend characteristics obtained through analysis exceed a preset threshold range, weighting the temperature data and the vibration data, and determining a key source of abnormal fluctuation; According to the determined abnormal fluctuation source, carrying out association comparison on consumable residual data, and judging whether the operation abnormality caused by insufficient consumable exists or not; Through comprehensive analysis of trend characteristics and abnormal fluctuation sources, a prediction model of the running state of the equipment is generated, and potential risk points in a future period of time are obtained; and dynamically adjusting the operation parameters of the equipment according to the potential risk points acquired by the prediction model to obtain an optimized operation strategy.
3. The intelligent scheduling system and method for printing production based on real-time data according to claim 1, wherein the steps of fusing interaction influencing factors between devices such as environmental temperature fluctuation and load change according to the change trend characteristics of the running state of the devices, analyzing the association mode of temperature rise and vibration aggravation by adopting a decision tree algorithm, and determining the occurrence probability of potential abnormal symptoms such as device failure precursors comprise: acquiring historical time sequence data and real-time data of the running state of equipment; calculating a change trend according to the equipment operation state data; extracting a temperature rise sequence and a vibration aggravation sequence from the change trend; Acquiring environmental temperature fluctuation data and load change data; Training the temperature rise sequence, the vibration aggravation sequence, the environmental temperature fluctuation data and the load change data through a decision tree algorithm to obtain a correlation mode model; Predicting the current change trend, the current environmental temperature fluctuation data and the current load change data by adopting a correlation mode model, and judging whether a correlation mode of temperature rise and vibration aggravation exists or not; if the association mode is judged to exist, the occurrence probability of the fault precursor is calculated.
4. The intelligent scheduling system and method for printing production based on real-time data according to claim 1, wherein if the occurrence probability of the potential abnormal symptom exceeds a preset threshold, extracting a specific shutdown requirement time point from the real-time monitoring information, and judging an upcoming equipment maintenance window period, wherein the method comprises the following steps: Step one, acquiring equipment running state data from a real-time monitoring system, continuously tracking abnormal symptoms, and determining occurrence probability of the abnormal symptoms by comparing deviation of historical data and current data; If the occurrence probability of the abnormal symptom exceeds a preset threshold, extracting relevant time sequence information from the real-time monitoring data, and judging a specific shutdown demand time node by combining the equipment operation log; Step three, according to the shutdown demand time node, acquiring an available time period in the equipment maintenance plan, and determining an upcoming equipment maintenance window period by adopting a time matching method; Step four, aiming at the equipment maintenance window period, acquiring operation load data of related equipment, and judging the optimal maintenance time by analyzing the fit degree of load distribution and maintenance time; if the optimal maintenance time is consistent with the time period in the window period, extracting a resource list required for maintenance from the system, and determining the execution sequence of maintenance tasks by combining with the resource availability data; Step six, acquiring state monitoring data before and after equipment shutdown according to the execution sequence of maintenance tasks, and judging the execution integrity of the maintenance tasks through state comparison analysis; And step seven, if the execution integrity of the maintenance task meets the preset standard, inputting the maintenance time, the time node and the task sequence information into the system, and obtaining the optimization arrangement of the subsequent operation of the equipment by automatically updating the maintenance plan.
5. The intelligent scheduling system and method for printing production based on real-time data according to claim 1, wherein the obtaining the equipment maintenance window period, for the order task in the current production schedule, adopts a dynamic adjustment mechanism such as task priority ordering, redistributes the order task to other available industrial equipment, and obtains an optimized task execution sequence, and includes: analyzing a list of unavailable devices during maintenance by extracting relevant data of device maintenance and window deadlines from a production scheduling system to obtain an affected order task set; according to the affected order task set, combining task priority data, adopting a preset priority ordering rule to evaluate importance of the tasks and determining a task list needing urgent reassignment; Aiming at the urgent reassigned task list, acquiring the running state and load data of the current available equipment, judging whether the equipment meets the task execution condition, and if the load is lower than a preset threshold, assigning the task to the corresponding industrial equipment to obtain a preliminary assignment scheme; Extracting task execution time and equipment matching information from the primary allocation scheme, and performing secondary allocation on equipment with time conflict or load overrun through a dynamic adjustment mechanism to determine an adjusted task execution sequence; Acquiring an adjusted task execution sequence, analyzing the operation efficiency of each industrial device in the sequence, and optimizing the sequence by adopting a genetic algorithm to obtain an optimized execution sequence; generating a final task allocation instruction according to the optimized execution sequence, transmitting the final task allocation instruction to an industrial equipment control system, and completing the reallocation of order tasks; by monitoring the execution condition of the final task allocation instruction in real time, recording the running state of the equipment and the task completion progress, judging whether an execution deviation exists, and if the deviation is detected, triggering a dynamic adjustment mechanism to regenerate the allocation instruction.
6. The intelligent scheduling system and method for printing production based on real-time data according to claim 1, wherein the simulating the equipment load distribution in the production process by the optimized task execution sequence and predicting the overall production continuity index after task transfer, such as the production line no-interruption operation rate, by using a time sequence data processing network comprises: acquiring the time point and the duration time of each task distributed to the equipment in the optimized sequence; According to the optimized sequence, arranging the task execution sequence, calculating the load accumulation duration of each device in the whole time period, and forming the device load distribution condition; Overlapping time sequence dimensions according to equipment load distribution conditions to generate multi-equipment parallel load time sequence data; adopting a time sequence data processing network to forward propagate multi-equipment parallel load time sequence data to obtain load fluctuation predicted values before and after a task transfer point; if the load fluctuation predicted value of the task transfer point exceeds a preset threshold, marking the corresponding transfer point as a potential interruption risk point; inserting a transfer adjustment window into the time sequence data according to the potential interruption risk point position, reallocating the execution sequence of adjacent tasks, and updating the optimized sequence; And re-simulating the load distribution condition of the equipment through the updated optimized sequence, generating new multi-equipment parallel load time sequence data, and inputting the new multi-equipment parallel load time sequence data into a time sequence data processing network to obtain the uninterrupted operation rate predicted value of the production line.
7. The intelligent scheduling system and method for printing production based on real-time data according to claim 1, wherein if the overall production continuity indicator is lower than a target threshold, iteratively adjusting the task allocation ratio according to the feedback data of the task transfer, such as the equipment utilization, to determine a final production scheduling scheme, comprising: acquiring overall index data in a production system, classifying and sorting records related to production continuity, and determining the difference between the current index and a preset threshold; if the current index is lower than a preset threshold, extracting feedback data from the task transfer record, and analyzing the distribution condition of the equipment utilization rate to obtain a preliminary evaluation result of the equipment load; calculating the adjustment direction of the task allocation proportion according to the evaluation result of the equipment load, and adopting a pre-established linear regression model to process the matching relationship between the task and the equipment to determine a preliminary task allocation adjustment scheme; simulating operation is carried out on the preliminary task allocation adjustment scheme, the simulated production continuity index change is obtained, and whether the target setting is approached or not is judged; If the simulated index still does not reach the target setting, iteratively updating the task allocation proportion according to the equipment utilization rate data fed back in the simulation operation to obtain a new scheduling scheme; Aiming at the new scheduling scheme, analyzing the influence degree of the new scheduling scheme on the whole index, acquiring the execution parameters of the final production scheduling, and determining the formal production scheduling scheme; And automatically updating task allocation and equipment matching records by writing data of the formal production scheduling scheme into a system scheduling module to obtain real-time operation configuration of the production system.
8. The intelligent scheduling system and method for printing production based on real-time data according to claim 1, wherein the obtaining the final production scheduling scheme, integrating updated data of real-time factors such as sudden load change, continuously monitoring equipment shutdown risk, and obtaining a production efficiency optimization path based on closed-loop feedback control, comprises: Acquiring real-time data of a production field, acquiring equipment running state and load fluctuation information, and classifying and storing the acquired information through a data stream processing technology to obtain a running data set after preliminary arrangement; According to the operation data set after preliminary arrangement, analyzing hidden danger of equipment shutdown and fluctuation conditions of sudden load, triggering an abnormal mark if detecting that the load fluctuation exceeds a preset threshold value, screening high-risk equipment through the abnormal mark, and determining potential shutdown risk points; Aiming at potential shutdown risk points, combining with continuously monitored data streams, adopting a pre-established prediction model to perform risk assessment, analyzing the running trend of the equipment through a support vector machine algorithm, and judging the shutdown probability distribution of the equipment in a future time period; According to the equipment shutdown probability distribution, a closed loop feedback mechanism is combined to dynamically adjust a production scheduling scheme, and an adjusted scheduling strategy is obtained through the comparative analysis of real-time data and a historical record; aiming at the adjusted scheduling strategy, an optimized path of production efficiency is analyzed, and key nodes for improving the efficiency are determined through comprehensive evaluation of load fluctuation and equipment states; And generating a final production scheduling optimization scheme according to the key nodes with improved efficiency, and judging a bottleneck link in the production flow by combining the optimization path with real-time monitoring data to complete closed-loop control of efficiency optimization.

Description

Printing production intelligent scheduling system and method based on real-time data Technical Field The invention relates to the technical field of information, in particular to an intelligent printing production scheduling system and method based on real-time data. Background The intelligent scheduling system for printing production belongs to the core field of digital transformation of industrial manufacturing, directly relates to enterprise delivery capacity and resource utilization efficiency, and has key strategic significance in the printing industry with vigorous competition. Although a certain digital tool is introduced in the current scheduling mode, the equipment shutdown event frequently occurring in the production process is still difficult to effectively cope with, so that order delay and productivity waste are caused. The existing scheduling methods mainly rely on historical experience or fixed rules to make a plan, and when the methods face sudden changes of the running state of equipment, abnormal symptoms cannot be timely perceived, so that serious conflict between shutdown maintenance and production tasks is caused. Signals such as equipment temperature rise, vibration aggravation or consumable exhaustion are difficult to capture in advance, passive shutdown in the production process is forced, and the whole scheduling sequence is disturbed. Accurate prediction of downtime becomes an important bottleneck limiting system intelligence, because equipment downtime involves the interactive effects of various real-time factors, including operating conditions, consumable supply margin, and sensor-monitored temperature vibration data, which are interrelated, and subtle fluctuations in operating conditions can be rapidly transmitted to consumable consumption rates and changes in equipment physical parameters. If these associated dynamics cannot be captured in a comprehensive way, it is difficult to determine when operations that have to be stopped, such as changing paper, changing ink or cleaning the cylinder, occur, resulting in overlapping of the maintenance window with the order production, and frequent interruption of the production rhythm. In an actual printing workshop, when consumable materials are about to be exhausted or the temperature of one device is abnormally increased in the middle of executing a large batch order, the original schedule cannot reserve maintenance time in advance, and only emergency shutdown processing is performed, so that subsequent order backlog is caused, and other idle devices are idle due to untimely task allocation. Therefore, how to accurately predict the shutdown requirement according to the real-time running state and the sensor data in the scheduling and rapidly transfer the task to other equipment when the abnormal symptoms appear becomes a key problem for realizing the improvement of the continuity and the efficiency of the printing production. Disclosure of Invention The invention provides an intelligent printing production scheduling system and method based on real-time data, which mainly comprise the following steps: acquiring real-time monitoring information such as temperature data, vibration data, consumable allowance data and the like in the operation of industrial equipment through a temperature sensor and a vibration sensor, and analyzing the real-time monitoring information by adopting a time sequence data processing network to obtain the change trend characteristics of the operation state of the equipment; According to the change trend characteristics of the running states of the equipment, the interaction influencing factors such as environmental temperature fluctuation and load change among the equipment are fused, a decision tree algorithm is adopted to analyze the correlation mode of temperature rise and vibration aggravation, and the occurrence probability of potential abnormal symptoms such as equipment failure precursors is determined; if the occurrence probability of the potential abnormal symptom exceeds a preset threshold value, extracting a specific shutdown requirement time point from the real-time monitoring information, and judging an upcoming equipment maintenance window period; Acquiring the equipment maintenance window period, adopting a dynamic adjustment mechanism such as task priority sequencing aiming at order tasks in the current production schedule, and redistributing the order tasks to other available industrial equipment to obtain an optimized task execution sequence; simulating equipment load distribution conditions in a production flow through the optimized task execution sequence, and predicting overall production continuity indexes such as the uninterrupted operation rate of a production line after task transfer by adopting a time sequence data processing network; If the overall production continuity index is lower than a target threshold, iteratively adjusting the task allocation proportion according to feedback d