A remote monitoring and management system for MES equipment in rare earth production

Abstract

The present invention relates to the field of remote monitoring technology, specifically a remote monitoring and management system for MES equipment in rare earth production, the system comprising: a material scheduling remote monitoring module, which collects real-time information of each material interaction point, including loading and unloading, reaction equipment status and warehousing conditions, and generates a real-time interaction information overview; based on the real-time interaction information overview. In the present invention, by collecting real-time information of material interaction points, loading and unloading, reaction equipment status and warehousing conditions are integrated into a real-time overview, making the dynamic state of the material flow process transparent throughout. Based on this information, dynamic path optimization and task priority adjustment are performed to realize the allocation and scheduling of transportation tasks, reducing resource waste and time consumption during transportation. Combined with the analysis of the real-time status of the reaction equipment and the arrival time of the materials, the start and pause time of the production tasks can be accurately adjusted, reducing the impact of task waiting on production efficiency.

Description

Technical Field

[0001] The present invention relates to the field of remote monitoring technology, and in particular to a remote monitoring and management system for MES equipment in rare earth production. Background Art

[0002] Remote monitoring technology is an intelligent management method based on computer networks and communications technologies, widely used in industrial automation, smart cities, energy management, and other fields. It collects data from remote devices or systems in real time, combines sensor technology with communication protocols, and transmits this information to a central monitoring system for processing, analysis, and display. The Manufacturing Execution System (MSE) is an information system that connects the factory floor (workshop) with the enterprise level (such as the ERP system). However, when it comes to production task control, the inability to dynamically analyze equipment status and material arrival times makes it difficult to schedule task starts and pauses, which can easily lead to task delays and idle resources. Process monitoring relies heavily on static models, lacking dynamic update capabilities. This results in poor connectivity between production stages, which can easily lead to process breakpoints and task disorganization. Therefore, improvements are needed. Summary of the Invention

[0003] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a remote monitoring and management system for MES equipment in rare earth production.

[0004] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solution: A remote monitoring and management system for MES equipment in rare earth production includes:

[0005] The material scheduling remote monitoring module collects real-time information on each material interaction point, including loading and unloading, reaction equipment status, and storage conditions, and generates a real-time interaction information overview. Based on this real-time interaction information overview, it performs dynamic route planning, adjusts the priority and route of transportation tasks, and generates optimized transportation instructions.

[0006] The production process control optimization module analyzes the real-time status of the reaction equipment and the expected material arrival time based on the optimized transportation instructions, adjusts the task start and pause time points in the production process, and generates an adjusted production task sequence; according to the optimized transportation instructions, it synchronously updates the status of each production stage in the Petri net to generate the current process state diagram;

[0007] A deadlock detection and resolution module analyzes the state space of the Petri net based on the current process state diagram, detects deadlock risks, and generates deadlock risk results; based on the deadlock risk results, adjusts task priorities and generates an adjusted resource allocation plan;

[0008] The remote monitoring and real-time feedback module monitors the real-time operating status of production equipment and material flow based on the adjusted resource allocation plan, tracks the progress and execution of each task, and obtains the task execution status results.

[0009] Preferably, the steps of obtaining the real-time interaction information overview are:

[0010] Sensors installed in loading and unloading areas, reactors, and warehouses collect real-time information, including the status of loading and unloading operations, current operating parameters and status of reactors, and real-time data on storage capacity, which are aggregated into raw real-time information datasets.

[0011] Based on the original real-time information data set, invalid data and outliers are removed, and the loading and unloading operation status, the operating status data of the reaction equipment, and the storage capacity data are classified and labeled to form a structured and organized real-time information data set;

[0012] Based on the structured and organized real-time information data set, the efficiency of loading and unloading operations, the operating efficiency of equipment, and storage usage are evaluated to obtain a real-time interactive information overview.

[0013] Preferably, the steps for obtaining the optimized transport instructions are:

[0014] Compare the real-time interactive information overview with past route data, analyze the optimal route and potential bottlenecks under the current environment, and obtain a preliminary transportation route plan;

[0015] Based on the preliminary transport path plan, the adjusted path weight is calculated using the formula:

[0016] ;

[0017] in, For the The adjusted path weight of each task, Indicates the The original path weight corresponding to each task, For the Task and The efficiency score of each link, 、 and are adjustment coefficients respectively, and m is the total number of evaluation links;

[0018] Based on the adjusted path weights, resource allocation and scheduling plans are optimized, and the execution order and resource input of the transportation tasks are reconfigured in combination with transportation demand and resource availability to obtain optimized transportation instructions.

[0019] Preferably, the steps for obtaining the adjusted production task timing are:

[0020] Based on the optimized transport instructions, the time difference between the arrival of materials and the availability of equipment is calculated, and the start and pause time points of the task are adjusted. The formula is:

[0021] ;

[0022] in, It is The adjusted start time of each task, is the estimated arrival time of the material, is the estimated time the device will be available. is the task urgency index;

[0023] The adjusted start time is used to rearrange the production process to obtain an adjusted production task sequence.

[0024] Preferably, the steps for obtaining the current process state diagram are:

[0025] Based on the optimized transportation instructions, updating the nodes and edges in the Petri net, recording the state transition of each production stage, including the task start, in progress and completed status, and obtaining an updated Petri net state model;

[0026] The production process is analyzed according to the updated Petri net state model, and the state of the Petri net is converted into a process state diagram through mapping to obtain the current process state diagram.

[0027] Preferably, the steps for obtaining the deadlock risk result are:

[0028] Based on the current process state diagram, collect the status data of nodes in each production stage, including resource occupancy, task queuing and completion status, and obtain a complete Petri net state description;

[0029] According to the complete Petri net state description, the deadlock risk index is calculated using the formula:

[0030] ;

[0031] in, Indicates the The deadlock risk index of each node, For the Node The current state of the class resource, is the threshold state of the same type of resources, is the total number of resource categories;

[0032] The deadlock risk index of each node is used to conduct risk assessment on the nodes, and the nodes exceeding the risk threshold are screened to generate deadlock risk results.

[0033] Preferably, the steps for obtaining the adjusted resource allocation plan are:

[0034] Based on the deadlock risk results, correlation analysis is performed with the current task execution progress, risk tasks are identified and sorted into a priority list of risk tasks, and a deadlock risk task correlation table is generated;

[0035] Based on the deadlock risk task association table, the priority adjustment coefficient of the task is calculated using the following formula:

[0036] ;

[0037] in, For the The priority adjustment coefficient of each task, Indicates the The resource usage of each task, and Respectively The minimum resource requirements and maximum resource limits of each task, The first in history The average delay time of each task, Indicates the estimated completion time of the current task. Indicates the The deadline for each task;

[0038] According to the priority adjustment coefficient, the tasks are sorted, and resources are reconfigured to generate an adjusted resource allocation plan.

[0039] Preferably, the steps for obtaining the task execution status result are:

[0040] According to the adjusted resource allocation plan, extract the real-time operation data and material flow data of the production equipment, including the working status of the equipment, the material transmission speed and the time point of arrival at the node, and integrate the data to form the equipment operation and material flow data set;

[0041] Based on the equipment operation and material flow data set, the progress of each production task is tracked, the task start time, current progress status and stage completion rate are recorded, and real-time task progress tracking data is generated;

[0042] According to the real-time task progress tracking data, the execution status of each task is analyzed to check whether there are delays, stagnation or abnormalities, and the analysis results are summarized to obtain the task execution status results.

[0043] Compared with the prior art, the advantages and positive effects of the present invention are:

[0044] In the present invention, by collecting real-time information on material interaction points, loading and unloading, reaction equipment status and warehousing conditions are integrated into a real-time overview, making the dynamic status of the material flow process transparent throughout. Based on this information, dynamic path optimization and task priority adjustment are performed to achieve the allocation and scheduling of transportation tasks, reducing resource waste and time consumption during transportation. Combined with the analysis of the real-time status of the reaction equipment and the arrival time of the materials, the start and pause time of the production tasks can be accurately adjusted, reducing the impact of task waiting on production efficiency. By dynamically updating the status of each stage in the production process, a real-time flow chart is fully displayed, the synchronization and coordination of production tasks are optimized, and the process discontinuity caused by information lag is reduced. Combined with real-time adjustment of resource allocation and deadlock detection, the risk of resource conflicts is actively avoided to ensure the stable operation of the production process. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a system flow chart of the present invention. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0047] See also Figure 1 The present invention provides a technical solution: a remote monitoring and management system for MES equipment in rare earth production, comprising:

[0048] The material scheduling remote monitoring module collects real-time information on each material interaction point, including loading and unloading, reaction equipment status, and storage conditions, and generates a real-time interaction information overview. Based on this real-time interaction information overview, it performs dynamic route planning, adjusts the priority and route of transportation tasks, and generates optimized transportation instructions.

[0049] The production process control optimization module analyzes the real-time status of the reaction equipment and the estimated material arrival time based on the optimized transportation instructions, adjusts the start and pause times of tasks in the production process, and generates an adjusted production task sequence. Based on the optimized transportation instructions, it also synchronously updates the status of each production stage in the Petri net to generate the current process state diagram.

[0050] The deadlock detection and resolution module analyzes the state space of the Petri net based on the current process state diagram, detects deadlock risks, and generates deadlock risk results. Based on the deadlock risk results, it adjusts task priorities and generates an adjusted resource allocation plan.

[0051] The remote monitoring and real-time feedback module monitors the real-time operating status of production equipment and material flow based on the adjusted resource allocation plan, tracks the progress and execution of each task, and obtains the task execution status results.

[0052] The steps to obtain the real-time interaction information overview are as follows:

[0053] Sensors installed in loading and unloading areas, reactors, and warehouses collect real-time information, including the status of loading and unloading operations, current operating parameters and status of reactors, and real-time data on storage capacity, which are aggregated into raw real-time information datasets.

[0054] Based on the original real-time information data set, invalid data and outliers are removed, and the loading and unloading operation status, the operating status data of the reaction equipment and the storage capacity data are classified and labeled to form a structured and organized real-time information data set;

[0055] Based on the structured and organized real-time information data set, the efficiency of loading and unloading operations, the operating efficiency of equipment and the use of storage are evaluated to obtain a real-time interactive information overview.

[0056] Specifically, after collecting real-time information through sensors installed in the loading and unloading area, reaction equipment and warehouse locations, when processing the real-time data of the loading and unloading operation status, the current operating parameters and operating conditions of the reaction equipment, and the storage capacity, the specific settings of the numerical range of each type of information can be clarified first, such as comparing the temperature with the actual temperature. to Range, if the temperature exceeds It is considered abnormal, and then the pressure is compared to Range, if the pressure exceeds It is classified as abnormal and the current and voltage indicators are compared respectively. to as well as to If the value exceeds the preset range, it will be marked as out-of-limit data. Then, the loading and unloading operation status will be analyzed according to the actual collection order. For example, the number of actions and the corresponding speed of each lifting or placing of goods will be recorded. The established experience value and safety limit value will be compared to determine whether there is an abnormality. If the data readings of some sensors are far higher than the numerical range set by experience (for example, the temperature reaches or pressure reaches ), a secondary check is required at the acquisition end or data end to confirm whether the sensor is faulty or the installation position is deviated. Then, the current operating parameters and operating conditions of the reaction equipment are read and compared in the same way. If the actual monitoring value of the speed or power consumption deviates from the pre-set range (for example, the speed exceeds For example, if the storage capacity of raw materials and finished products is set at a range of 0 to 20,000 units, and the remaining storage capacity is less than 500 or more than 20,000, it is classified as high risk or overloaded. This is then correlated with other abnormal data to form a comprehensive reference for each abnormal indicator. Ultimately, this information is aggregated into the original real-time information data set.

[0057] Based on the original real-time information data set obtained above, when further inspecting and classifying the loading and unloading operation status, the operating status data of the reaction equipment and the storage capacity data, first set the valid data range and abnormal value judgment standard in the cleaning rule. For example, if the corresponding parameter of a sensor deviates from the above temperature or pressure range several times in a row, it will be temporarily classified as suspicious information range. Then, combine the experience method or compare with other normal data points in the actual business scenario to determine whether it is really abnormal. When it is confirmed that some values ​​are indeed beyond the available range, refer to the steady-state rules inferred in advance based on historical operation data and manufacturing manuals (for example, the speed should be kept at to Between, the pressure to ), remove or mark this part of the data one by one, and in this process, mark the loading and unloading status, equipment status, and storage status with classification labels. For example, data with a storage capacity exceeding 20,000 or less than 500 is marked as "storage critical", and values ​​exceeding the safe pressure are marked as "equipment safety". The labeled results are then merged according to time series and attribute classification. If some data is found to violate both temperature and pressure thresholds during the execution process, multi-label records are additionally performed during the final collation, and the values ​​of these multiple labels are separately collected. All the data with the classified values ​​are then used as structured collation content to obtain a structured and collated real-time information dataset.

[0058] According to the structured and organized real-time information data set obtained above, calculations and comparisons can be made based on the three aspects of loading and unloading efficiency, equipment operation efficiency, and storage usage. In order to clarify the judgment method of loading and unloading efficiency, a loading and unloading efficiency reference threshold range can be set. For example, the normal loading and unloading frequency is defined as 10 to 20 times per hour. If the number of loading and unloading times recorded in a certain period is less than 10 or more than 20, it will be marked in the results respectively. At the same time, in terms of equipment operation efficiency, relevant standard values ​​such as speed, energy consumption and production batches can be set. , power consumption , a maximum of 5 production batches within 1 hour are compared one by one. If it is found that the production batches corresponding to the equipment power consumption and speed within a certain period of time deviate from this standard, the abnormal situation will be further marked. The storage usage is divided according to the storage capacity interval. When some values ​​are detected to be less than 500 or more than 20,000, they are regarded as impact points for subsequent tasks and recorded in the corresponding category. Finally, after comparing and evaluating all relevant information, an overview of real-time interactive information is obtained.

[0059] The steps for obtaining optimized transportation instructions are as follows:

[0060] Compare the real-time interactive information overview with past route data to analyze the optimal route and potential bottlenecks under the current environment and obtain a preliminary transportation route plan;

[0061] Based on the preliminary transportation route plan, the adjusted route weight is calculated using the following formula:

[0062] ;

[0063] in, For the The adjusted path weight of each task, Indicates the The original path weight corresponding to each task, For the Task and The efficiency score of each link, 、 and are adjustment coefficients respectively, and m is the total number of evaluation links;

[0064] Based on the adjusted path weights, resource allocation and scheduling plans are optimized. Combined with transportation demand and resource availability, the execution order and resource input of transportation tasks are reconfigured to obtain optimized transportation instructions.

[0065] Specifically, after comparing the real-time interactive information overview with past path data, the process of analyzing the optimal path and potential bottlenecks in the current environment can first extract data including vehicle speed, loading and unloading time and road section capacity from historical transportation records, and match them with the current real-time interactive information data. By comparing the transportation speed of each time period and the waiting time when arriving at the node one by one, the locations where delays are likely to occur in the route can be identified. The pass rate of different road sections and the load degree of the loading and unloading area are then further marked. The time it takes for each vehicle to arrive at the previous road section and whether it exceeds the normal waiting interval that has been established after arrival (for example, the waiting time is between 5 minutes and 15 minutes) is recorded. If the waiting time is less than 5 minutes, it is marked as a short time interval. If the waiting time exceeds 15 minutes, it is marked as a long time interval. Then, based on the repeated high-congestion sections obtained by multiple comparisons, the results are summarized. The system then calculates and verifies whether the operating load of the loading and unloading area is at a high level within the same time period (for example, 20 operations per hour or more). If the number of operations exceeds 20, it is classified as high-load, and if it is less than 5, it is classified as low-load. These statistically high-congestion sections and high-load periods are regarded as potential bottleneck areas or time periods that require special attention. Based on this, the actual traffic time of each section and node is weightedly compared with the operating time of the loading and unloading area to form a corresponding bottleneck distribution map. In this process, the vehicle load factor and the required human resources can also be combined. If the actual load weight is close to the rated capacity of the vehicle (for example, 9.9 tons is close to the rated load of 10 tons) or the number of on-site loading and unloading personnel is insufficient, this situation is marked as load critical. Finally, by comparing all high-congestion sections, load criticality and vehicle load factors, a preliminary transportation route plan is obtained.

[0066] The benefit of the formula is that by introducing the linear combination values ​​of multiple parameters in the index part, it can simultaneously consider the efficiency scores of different links and the original path weights, thereby taking into account multiple factors within the same calculation framework.

[0067] The steps to obtain the parameters are as follows: in the test of multiple transport routes, continuously observe the vehicle operation process and record the basic transport difficulty values ​​of different routes, and record the basic transport difficulty value of each route as , and then Do weighted averaging to get a mean and perform dispersion analysis, remove samples with large dispersion and recalculate The reference range is determined by combining the statistical results of equipment operation cycle and loading and unloading area operation frequency, and then fine-tuning the reference range to determine the final The numerical value of .

[0068] The steps for obtaining parameters are as follows: monitor the execution process of all transportation tasks for several days, record the number and degree of task interference factors each time the same task is executed, and summarize them into a data list. Then, quantify the average waiting time in the section where interference factors frequently appear, and obtain the statistical data. , use the Measure the overall level of interference and then determine specific scope.

[0069] The steps for obtaining parameters are: measure each link step by step, such as loading and unloading link, measurement link, re-inspection link, etc., and record the operation time or operation difficulty score for each link, and quantify it into an interval between 0 and 1 to form These values ​​come from the statistical analysis process of the proportion of time spent in each link to the total time spent.

[0070] The steps to obtain the parameters are as follows: the historical path weights obtained previously already include the objective difficulty indexes of multiple routes, and the one that matches the first route can be selected. The route difficulty index that best matches the transport task is used as The index is quantified by the average driving speed and sudden delay period of the same type of vehicles in historical records.

[0071] The steps to obtain the parameters are as follows: in the efficiency score of each link obtained previously, corresponding quantitative indicators are set for dimensions such as loading and unloading speed, weighing accuracy, and dispatch response time. For example, the loading and unloading speed score is between 0 and 1, the weighing accuracy score is between 0 and 1, and the dispatch response score is between 0 and 1. Then the score of each link is recorded as , indicating that During the execution of a task, The efficiency value of each link.

[0072] Calculation process:

[0073] First This part combines the values, for example:

[0074] , , , , , , , ;

[0075] Substituting these values ​​into:

[0076] ;

[0077] ;

[0078] ;

[0079] ;

[0080] ;

[0081] then ;

[0082] Then take the negative exponent of that value: ;

[0083] Then calculate the denominator: ;

[0084] at last ;

[0085] The results show that under the current combination, the adjusted path weight is about 0.673, which means that the A relatively high comprehensive priority of a task. If the numerical result is close to 1, it means that the task is evaluated as a more important path weight. If the value is significantly less than 0.5, it means that the task has a relatively low priority in the scheduling arrangement.

[0086] Based on the adjusted path weights, combined with transportation demand and resource availability, when reconfiguring the execution order and resource input of transportation tasks, you can first obtain the quantity range of real-time allocable resources from the available number of transportation vehicles and the loading and unloading personnel schedule. For example, 5 to 10 vehicles can be put into use daily, and 10 to 20 loading and unloading personnel can be deployed. Then compare these quantity ranges with the transportation task list, evaluate the dependence of each transportation task on vehicles and personnel item by item, record the requirements of certain tasks for specific vehicle types or the need for additional operators, and if you encounter tasks that require large-capacity vehicles or more operators, mark them centrally to allocate corresponding resources, and then allocate all resources according to the execution period of each task. A preliminary list is compiled based on the scheduling order of the source, and compact tasks are arranged in the early schedule according to the principle of stage-by-stage arrangement. For periods with heavy loads, priority is given to checking whether there are vehicles that meet the conditions and can be reused within this period. For example, it is checked whether the current vehicle can return in time after completing the previous route. The vehicle's return time must also be compared here. If it exceeds the pre-set vehicle safety interval (for example, 30 minutes of inspection time should be reserved after each trip), staggered adjustments are made. After the overall scheduling is completed, the terminal will output the distribution of each vehicle and personnel in the corresponding time period. By comparing the resource allocation and task list of each time period, the final task execution order and resource scheduling plan are generated to obtain the optimized transportation instructions.

[0087] The steps for obtaining the adjusted production task timing are:

[0088] Based on the optimized transport instructions, the time difference between material arrival and equipment availability is calculated, and the start and pause times of the task are adjusted. The formula is:

[0089] ;

[0090] in, It is The adjusted start time of each task, is the estimated arrival time of the material, is the estimated time the device will be available. is the task urgency index;

[0091] The adjusted start time is used to rearrange the production process and obtain the adjusted production task sequence.

[0092] Specifically, the formula is useful in determining the When calculating the start time of a task, the time difference between the material arrival time and the equipment availability time is taken into account. By taking the absolute value of the time difference and then performing a square root operation, the time gap can be quantified more flexibly. Combined with the logarithmic processing of the task urgency index, tasks with higher urgency can also be given higher priority when the time difference is not large.

[0093] The acquisition steps are as follows: record the transportation process of each batch of materials, extract the estimated arrival time from the real-time location monitoring data provided by the transporter, compare these times with the average speed and congestion of the transport section obtained before, make appropriate corrections, and summarize them to form .

[0094] The steps to obtain the data are as follows: in the process of multiple observations of the equipment's schedule and available cycle, the idle period of the equipment and the current task occupancy are recorded respectively, the available time window is summarized, and according to the maintenance requirements of each equipment (for example, 30 minutes of maintenance and inspection are required for each 4-hour continuous operation), the time point when the equipment is expected to be put into the next task is calculated and recorded as .

[0095] The steps to obtain it are: when performing statistics in the factory workshop, record the historical task delay rate and the urgency of the demand for key materials. By calculating the number of expedited times per unit cycle, combined with the actual impact range of each task and the raw material consumption, the task urgency index is comprehensively obtained and digitized. It usually ranges from 0 to 5, and the higher the value, the more urgent it is.

[0096] Calculation process:

[0097] First introduce the example values:

[0098] (Hour), (Hour), ;

[0099] Calculate the absolute time difference between material arrival and equipment availability:

[0100] ;

[0101] Take the square root: ;

[0102] Execute on the emergency index Operation:

[0103] ;

[0104] ;

[0105] Finally, multiply the two together:

[0106] (Hour);

[0107] The results show that after calculating the time difference between material arrival and equipment availability combined with the task urgency, the task start time increased by 2.099 hours relative to the baseline starting point. If the calculated value is greater than 5, it means that the task needs to be started later due to a large time difference or an extremely high urgency index. If the calculated value is less than 1, it means that the task can be started faster.

[0108] Using the adjusted start time, the implementation process of the production process can be re-arranged. The adjusted time value of each task can be compared with the workshop's scheduled time period. The fixed time period of each production stage and the operating time line of the corresponding equipment can be read through the workshop's management system, and the effective range inspection rules can be set for each time period. For example, the morning shift is scheduled between 8 and 12 o'clock. If the calculated start time falls outside this range, the task is marked as not suitable for the morning shift. Then compare the available time periods of the afternoon or night shift. After confirming the final time period, compare whether there is any overlap with other previously confirmed tasks. If it is found that the established parallel threshold is exceeded (for example, the same equipment can only run two parallel processes at the same time) and the total number of current tasks has reached 2 or more, the time period needs to be adjusted. In order to postpone or rearrange the order appropriately, it is also necessary to make arrangements based on the storage status and supply quantity of different materials, and record the delivery time of multiple materials. If some materials will arrive at the workshop before a specific time period and the additional transportation cannot be completed at one time based on the previously obtained transportation data, the subsequent tasks must be arranged after their arrival time to avoid repeated waiting. By comparing each possible time period allocation and equipment idle status, the time period where resource conflicts occur is further optimized and sorted. Finally, the allocation information of all equipment in each time period is combined to see if there are any uncovered time periods or missed idle time. This part of the slack can also be allocated to subsequent tasks with a higher urgency index, forming a global timing diagram of the entire production process and obtaining the adjusted production task timing.

[0109] The steps for obtaining the current process state diagram are:

[0110] Based on the optimized transportation instructions, the nodes and edges in the Petri net are updated to record the state transition of each production stage, including the task start, in progress, and completed status, to obtain the updated Petri net state model;

[0111] According to the updated Petri net state model, the production process is analyzed, and the state of the Petri net is converted into a process state diagram through mapping to obtain the current process state diagram.

[0112] Specifically, based on the optimized transportation instructions, combined with the work requirements of each stage in the task list and equipment list, when the existing nodes and edges in the Petri net are updated one by one, the initial identification of each production node can be retrieved first and its previous timing and resource association information can be recorded. Then, the arrival time and material input method specified in the transportation instructions are read, and these times are compared with the records of equipment occupancy periods. If the production stage corresponding to a node has exceeded the pre-established parallel production threshold at a certain moment (for example, the number of registered parallel productions reaches 3 or more), then the node will be marked as waiting or queuing when it is updated, and the waiting period will be recorded on the edge of the Petri net to indicate that the task is in the queue. Here, it is necessary to set the parallel production threshold in combination with historical statistics. For example, a certain equipment can be observed to be in the queue multiple times within a month. At most, three similar products can be produced at the same time. The downtime risk and task delay records are summarized and three are determined as the threshold. If the current parallel number does not reach three, it can be directly put into the ongoing state and the corresponding edge is marked as the active state in the Petri net. For the completed stage, its end time and subsequent resource release information need to be recorded, and the release information is matched with the occupied period of the adjacent node to see if it can trigger the conditional activation of the subsequent node. If so, the identification of the subsequent node is updated to the startable state, otherwise the original state remains unchanged. All these state transition information will be recorded on the corresponding nodes and edges in the Petri net, including the start, ongoing and completed states. Finally, the timing and resource association updates of all stages are summarized to obtain the updated Petri net state model.

[0113] According to the updated Petri net state model, the process of analyzing the production process can first read the newly generated node identifiers and edge identifiers, identify the resource waiting time and working time recorded on each edge, and then compare the relationship between adjacent nodes one by one. If it is found that some nodes have been identified as waiting and their waiting time exceeds the specified reference standard (for example, the process guidelines stipulate that more than 2 hours on the same equipment may affect the material properties), this information will be written into the data set for subsequent analysis, and then the state information of all nodes will be converted into a visual mapping form. For example, the nodes in the ongoing state will be identified with different colors and highlighted in the diagram. The parallel production threshold can also be combined to judge the degree of resource occupancy in the same time period. If there are multiple If two nodes compete for the same equipment and one of them has been registered as in progress, the edges of other nodes on the equipment will be recorded as queued or ready. All these mappings need to be arranged with reference to the equipment type, material batch number and start and end time when drawing. Different production links in the same period can be divided by special lines or different shapes to ensure that analysts can see the connection relationship of each stage from the mapping. If a node is displayed as in progress but the status is not updated in time, it can be compared again with the previously obtained equipment operation status record, and the missing time or status identifier can be added to the corresponding edge, and then a global mapping can be performed. Finally, the sorted mapping graph is output as a process status diagram to obtain the current process status diagram.

[0114] The steps to obtain deadlock risk results are:

[0115] Based on the current process state diagram, collect the status data of nodes in each production stage, including resource usage, task queue and completion status, and obtain a complete Petri net state description;

[0116] According to the complete Petri net state description, the deadlock risk index is calculated using the formula:

[0117] ;

[0118] in, Indicates the The deadlock risk index of each node, For the Node The current state of the class resource, is the threshold state of the same type of resources, is the total number of resource categories;

[0119] The deadlock risk index of each node is used to conduct risk assessment on the nodes, and the nodes exceeding the risk threshold are screened to generate deadlock risk results.

[0120] Specifically, based on the current process status diagram, when collecting the status data of each production stage node, the resource occupancy, queue information, completion time and other contents can be extracted item by item from the node operation records obtained previously, and this information can be matched according to the node number. A specific threshold range is set for each type of resource. For example, the equipment operation time can be compared with the range of 0 hours to 24 hours. If the occupancy time of any node has reached 24 hours, it is marked as over-limit. The material conveying capacity can be compared with the range of 0 to 10,000 pieces. If it exceeds 10,000 pieces, it is marked as full load. The queue situation is further compared, and the number of consecutive queue states at multiple nodes is counted. If a node queues for more than three times in a row, it is classified as a high-waiting category so that the reason for its waiting can be accurately queried in subsequent analysis. If the node has entered the completion stage, its completion time is recorded and compared with the available cycle of the equipment to see if additional cooling or maintenance periods are required. If the maintenance period is set to leave 30 minutes for maintenance after each completion and this 30 minutes is within the mandatory maintenance standard given in the equipment manual, the completion status will be expanded to "Completed and in maintenance" in the node record. The above-mentioned annotation information is summarized in order to form a comprehensive data including all resource usage, queue times and completion node lists. These data are consistent with the current process status. Figure 1 One-to-one correspondence gives a complete description of the Petri net state.

[0121] The benefit of this formula is that by taking the cosine of the difference between resource occupancy and the threshold, averaging it across all resource categories, and then performing the maximum calculation, it can combine multiple resource states in the same expression, helping to identify nodes with a higher risk of deadlock.

[0122] The acquisition steps are: in the resource occupancy record obtained above, The first node under The resource usage of this type is monitored multiple times, and the usage or occupancy time values ​​are extracted. For example, the occupancy time of the equipment can be summarized and counted after each shift to obtain the average value of the cumulative occupancy time.

[0123] The steps to obtain the threshold value are as follows: in the equipment manual and production specification documents, set the threshold range for each resource, quantify the upper and lower limits of these ranges, and select the one that matches the first The standard value that matches the resource type is used as If the device is occupied for a certain period of time, you can refer to the specific limits of no more than 24 hours or no more than 48 hours and define them as threshold numbers respectively.

[0124] The acquisition steps are as follows: In the current production environment, all resource categories are counted and classified. If they include equipment, manpower, materials, etc., the serial numbers are set in the statistical table respectively, and the total number of resource categories counted is calculated as .

[0125] Calculation process:

[0126] make , indicating that there are two types of resources to consider, namely equipment occupancy time and material capacity occupancy. Assuming that a node Device usage time =28 (hours), its threshold =24 (hours), material capacity occupied =12000 (items), the threshold =10000 (items);

[0127] First calculate the difference:

[0128] ;

[0129] ;

[0130] Substituting the difference into the cosine term:

[0131] ;

[0132] ;

[0133] ;

[0134] and The same is also equal to 1;

[0135] Then find the average:

[0136] ;

[0137] Then take the maximum value: ;

[0138] therefore: ;

[0139] The results show that when the occupancy time and material occupancy of the node significantly exceed the set threshold, the deadlock risk index reaches 1. If the result is close to 1, it means that the resource difference is large and there is a high deadlock risk. If the result is significantly less than 0.5, it means that the resource status is close to the threshold and deadlock is not likely to occur.

[0140] Using the deadlock risk index of each node, the process of node risk assessment can first read the deadlock risk index of each node obtained earlier, compare it with the equipment type and occupancy category, and mark the resource type, node number and risk index in a comparison table. If it is found that the deadlock risk index of a node has exceeded the threshold value obtained through empirical statistics (for example, 0.8 is calculated as the limit point based on the equipment downtime record of a workshop for one quarter, and exceeding 0.8 may cause serious congestion), then it will be marked as high risk on the node label and the category of the actual difference will be recorded. For example, the node whose material occupancy exceeds the standard by a large margin will be marked as "high occupancy". Identify and mark the nodes whose equipment operation time seriously deviates from the limit range of the equipment manual with a "serious deviation" mark. Then, based on the number and specific distribution of all nodes marked as high-risk, screen them and collect them into a special list. Then, refer to the production process sequence sorted out previously one by one, and focus on these high-risk nodes to confirm whether it is necessary to adjust the task execution sequence or adopt additional personnel allocation strategies. If there are multiple nodes competing for the same key equipment and the risk index is high, first compare the scheduling time and restrict the resources that exceed the standard, converge the repeated occupancy, and then update the corresponding status data for each node to generate the final deadlock risk result.

[0141] The steps to obtain the adjusted resource allocation plan are:

[0142] Based on the deadlock risk results, correlation analysis is performed with the current task execution progress, risk tasks are identified and organized into a priority list of risk tasks, and a deadlock risk task correlation table is generated;

[0143] Based on the deadlock risk task association table, calculate the priority adjustment coefficient of the task. The calculation formula is:

[0144] ;

[0145] in, For the The priority adjustment coefficient of each task, Indicates the The resource usage of each task, and Respectively The minimum resource requirements and maximum resource limits of each task, The first in history The average delay time of each task, Indicates the estimated completion time of the current task. Indicates the The deadline for each task;

[0146] According to the priority adjustment coefficient, the tasks are sorted and resources are reallocated to generate an adjusted resource allocation plan.

[0147] Specifically, based on the deadlock risk results, in the process of correlation analysis with the current task execution progress, the number corresponding to the risk task and the current execution progress can be extracted from the previously obtained node risk index and task time scheduling data, and the status of each task in the most recent production schedule can be compared one by one. For example, check whether the task has been registered in the equipment occupancy list and exceeds the preset threshold (for example, continuous occupancy for more than 12 hours or queued for more than 3 times, etc.). If it does exceed, mark the task as a high-risk category, and then continue to search whether the subsequent available time period on the scheduling timeline can still meet the task deadline or whether additional resources are needed. Compare these marked tasks, and include those with high resource competition or high urgency index in the risk task list, and then confirm each task in the list one by one. Priority level, for example, read the equipment type requirement and material quantity requirement of the task. If the task's equipment operation time requirement is far higher than the experience threshold (for example, more than 24 hours) and the material requirement also significantly exceeds the normal range, then add an additional "resource shortage" label to this task entry. If it is found that the current task's recent completion time has been delayed several times compared with the previous period, and the average delay time exceeds the experience limit stipulated by management regulations (for example, statistics in the past month show an average delay of more than 2 hours), then mark "frequent delays" in the entry. Finally, make a comprehensive comparison of these marked risk tasks based on their urgency, resource utilization level, historical delays and other dimensions, and record the sorting results as a risk task priority list. Then, associate and mark the relative position of each task in this list and number it for subsequent tracing, to obtain a deadlock risk task association table.

[0148] The usefulness of this formula lies in that it combines multiple factors, such as resource utilization, resource demand differences, and historical delays, to quantify the priority. By taking the absolute value of the difference between the current task's estimated completion time and the deadline as the denominator, it can amplify or weaken the urgency of the task in the schedule.

[0149] The steps for obtaining it are: periodically summarize the resource call records of each task, count the actual equipment hours, manpower hours and material usage, etc., convert these different forms of resource occupancy values ​​into the same dimension and combine them for calculation, and obtain the resource occupancy degree in numerical form.

[0150] The steps to obtain the required resources are as follows: In the workshop production plan document or equipment list, find the minimum resource requirement allowed for the task. If it is the minimum equipment working hours requirement, refer to the minimum operating hours recorded in the previous rounds of production tests. If it is the minimum material usage requirement, refer to the minimum usage statistics of the same type of tasks. Finally, summarize it as .

[0151] The steps to obtain the value are as follows: filter the resource limit value of the task from the production safety manual and equipment upper limit data. If the equipment can only run for a certain time in a single schedule or there is a reserve limit for material use, these values ​​are mapped to the same dimension and summarized as .

[0152] The acquisition steps are as follows: extract the delay status of the task in the past multiple rounds of production scheduling from the historical records, average the delay time of each round, and if the task has different degrees of delay in each schedule, calculate the final average delay time and record it as .

[0153] The acquisition steps are as follows: in the current schedule, the estimated completion time of the task is recorded. The equipment occupancy table and the operator's time sheet can be extracted from the actual production management system for analysis. Then, the estimated completion time of the task is obtained by summing up these data and performing unified processing for the unit.

[0154] The steps to obtain the deadline are: find the deadline for the task to be completed in the production plan or order requirement document prepared in the early stage, and use the updated deadline as the final deadline if there is any change. .

[0155] Calculation process:

[0156] For example, let , , , , , (The unit can be determined according to the specific situation, such as hours or a comprehensive value of resource consumption);

[0157] First calculate the sum of squares:

[0158] ;

[0159] ;

[0160] ;

[0161] Recalculate :

[0162] ;

[0163] ;

[0164] Add the two together:

[0165] ;

[0166] Then calculate the denominator:

[0167] ;

[0168] Finally we get:

[0169] ;

[0170] The results show that in this example, the priority adjustment coefficient of the task is approximately 2.4972. A larger value indicates that the task has a higher priority in terms of resource usage, demand differences, and delays. A value lower than 1 may indicate that its urgency is relatively weak.

[0171] When sorting tasks and reconfiguring resources based on priority adjustment coefficients, the coefficient values ​​of each task can be recorded in a list first and compared with the resource demand information marked previously. If it is found that the coefficients of some tasks are high and the equipment consumption is significantly greater than the predetermined standards set by the workshop in advance (for example, a single operation shall not exceed the upper limit of equipment operation of 20 hours or continuous loading of more than 5,000 pieces of raw materials is not allowed), the task will be marked as a special processing item for subsequent comparison in specific scheduling. At the same time, if the coefficients of some tasks are higher than the predetermined threshold (for example, more than 3.0), they can be prioritized for production in the nearest period when the equipment is idle, and check whether the manpower allocation within this period is sufficient. For example, first compare the total number of human resources with the currently scheduled The tasks are calculated together to see whether the experience limit of 30 people in a single time period has been reached. If it is close to or exceeds 30 people, it is necessary to stagger the schedule or deploy additional manpower for the task. After these high-priority tasks are inserted into the schedule in a centralized manner, other tasks with lower coefficients are scanned and filled in the remaining idle time periods in turn, and the arrival time of material transportation is checked to see whether it matches the equipment maintenance period. If the equipment startup time is scheduled during the maintenance period, it must be adjusted again. After completing the scheduling of all items, a new resource allocation list is finally formed based on the matching relationship between tasks and resources, and a final comparison is made between the total equipment occupancy time and the manpower concentration to confirm that it does not violate management regulations and does not trigger new risks, and obtain the adjusted resource allocation plan.

[0172] The steps to obtain the task execution status results are:

[0173] Based on the adjusted resource allocation plan, the real-time operation data of production equipment and material flow data are extracted, including the equipment's operating status, material transmission speed, and arrival time at the node. This data is then integrated to form a data set of equipment operation and material flow.

[0174] Based on the equipment operation and material flow data set, the progress of each production task is tracked, the task start time, current progress status and stage completion rate are recorded, and real-time task progress tracking data is generated;

[0175] Based on the real-time task progress tracking data, the execution status of each task is analyzed to check whether there are delays, stagnation or abnormalities, and the analysis results are summarized to obtain the task execution status results.

[0176] Specifically, according to the adjusted resource allocation plan, in the process of extracting the real-time operation data and material flow data of the production equipment, the working status parameters of the corresponding equipment can be read item by item from the previously approved workshop equipment list, such as temperature, operating time and current load rate, and compared with the preset range defined previously, such as temperature comparison to If the temperature of a device has reached or exceeded the If the material arrival volume at a certain timestamp exceeds the warehouse's empirical threshold (for example, a single input must not exceed 500 units), additional resource usage statistics are performed. All of this working status information and the configuration information for material arrival times are collected in the same data table. During this process, it is necessary to match the equipment load rate, material transmission speed, and arrival time to check whether there is a material backlog due to idle equipment or a busy equipment due to excessive transmission speed. Each record is summarized by time period and finally integrated into a data set for equipment operation and material flow.

[0177] Based on the equipment operation and material flow data set, in the process of tracking the progress of each production task, you can first find the start time of each task and the corresponding required equipment number in the task schedule list, and compare the start time with the current working time axis of the equipment. If the time period found is still in the equipment maintenance or preheating period (for example, 30 minutes of maintenance must be reserved after 8 hours of continuous work, and 30 minutes is the mandatory inspection time mentioned in the equipment manual), the start time of the task will be postponed to the first full hour after the maintenance is completed, and then the "in progress" mark will be marked according to the actual time when the equipment enters operation, and the material of the task will be checked in combination with the material flow data. Whether it is delivered to the designated input point at the expected time. If it is found in the comparison that a material arrives more than 5 minutes later than expected, the delay status will be updated and accumulated records will be made. The stage completion rate can be counted based on the actual output of the machine. For example, the output of each hour can be compared with the expected output of the task to see whether the stage goal has been achieved. If the stage output has reached or exceeded 50% of the total demand, it will be marked as "half completed". If it exceeds 90%, it will be marked as "near completion". The cumulative output and time interval are recorded in each record to form a detailed progress tracking table. Finally, all the dot records and segmented statistical results are combined to obtain real-time task progress tracking data.

[0178] When analyzing the execution status of each task based on real-time task progress tracking data, the current progress level of each task is first compared against the established deadline based on the previously acquired start time, progress status, and stage completion rate. This comparison data is then used to screen for delays exceeding a set threshold (for example, a task experiencing a period of no output exceeding one hour is considered a stall, with the one-hour threshold derived from an average analysis of historical production data). Alternatively, a period of no activity coinciding with the time of the equipment error is identified as an anomaly. Delayed tasks and tasks with potential equipment failures are then listed in order, and the specific cause of the stall is further identified. For example, whether the downtime is due to material shortages, excessive equipment temperatures, or zero transmission speeds is determined. If material arrivals during the same period deviate significantly from actual demand, the original transportation data is reviewed to verify whether there is a vehicle scheduling mismatch or material allocation error. If the delay is limited to a specific production line, the start time and recovery period are tracked and summarized, and the analysis results are uniformly marked as delays or anomalies. Finally, the execution status data of all tasks is merged to obtain the task execution status results.

Claims

1. A remote monitoring and management system for MES equipment in rare earth production, characterized in that: The system comprises: The material scheduling remote monitoring module collects real-time information on each material interaction point, including loading and unloading, reaction equipment status, and storage conditions, and generates a real-time interaction information overview. Based on this real-time interaction information overview, it performs dynamic route planning, adjusts the priority and route of transportation tasks, and generates optimized transportation instructions. The production process control optimization module analyzes the real-time status of the reaction equipment and the expected material arrival time based on the optimized transportation instructions, adjusts the task start and pause time points in the production process, and generates an adjusted production task sequence; according to the optimized transportation instructions, the status of each production stage in the Petri net is synchronously updated to generate the current process state diagram; the deadlock detection and resolution module analyzes the state space of the Petri net based on the current process state diagram, detects deadlock risks, and generates deadlock risk results; based on the deadlock risk results, adjusts task priorities and generates an adjusted resource allocation plan; the remote monitoring and real-time feedback module monitors the real-time operating status and material flow of the production equipment based on the adjusted resource allocation plan, tracks the progress and execution of each task, and obtains the task execution status results; The steps for obtaining the deadlock risk result are: Based on the current process state diagram, collect the status data of nodes in each production stage, including resource occupancy, task queuing and completion status, and obtain a complete Petri net state description; According to the complete Petri net state description, the deadlock risk index is calculated using the formula: Among them, R i represents the deadlock risk index of the i-th node, S ij is the current state of the j-th type of resource at the i-th node, T ij is the threshold state of the same type of resources, and N is the total number of resource categories; Using the deadlock risk index of each node, we conduct risk assessment on the nodes, filter out nodes that exceed the risk threshold, and generate deadlock risk results; The steps for obtaining the adjusted resource allocation plan are: Based on the deadlock risk results, correlation analysis is performed with the current task execution progress, risk tasks are identified and sorted into a priority list of risk tasks, and a deadlock risk task correlation table is generated; Based on the deadlock risk task association table, the priority adjustment coefficient of the task is calculated using the following formula: Among them, P i is the priority adjustment coefficient of the i-th task, Q i Indicates the resource occupancy of the i-th task, L i and U i are the minimum resource requirement and maximum resource constraint of the i-th task, H i is the average delay time of the i-th task in the history, E i Indicates the estimated completion time of the current task, D i represents the deadline of the i-th task; According to the priority adjustment coefficient, the tasks are sorted, and resources are reconfigured to generate an adjusted resource allocation plan.

2. The MES equipment remote monitoring and management system for rare earth production according to claim 1, characterized in that: The steps for obtaining the real-time interaction information overview are: Sensors installed in loading and unloading areas, reactors, and warehouses collect real-time information, including the status of loading and unloading operations, current operating parameters and status of reactors, and real-time data on storage capacity, which are aggregated into raw real-time information datasets. Based on the original real-time information data set, invalid data and outliers are removed, and the loading and unloading operation status, the operating status data of the reaction equipment, and the storage capacity data are classified and labeled to form a structured and organized real-time information data set; Based on the structured and organized real-time information data set, the efficiency of loading and unloading operations, the operating efficiency of equipment, and storage usage are evaluated to obtain a real-time interactive information overview.

3. The MES equipment remote monitoring and management system for rare earth production according to claim 1, characterized in that: The steps for obtaining the optimized transportation instructions are: Comparing the real-time interactive information overview with past route data, analyzing the optimal route and potential bottlenecks under the current environment, and obtaining a preliminary transportation route plan; Based on the preliminary transport path plan, the adjusted path weight is calculated using the formula: Among them, C i is the adjusted path weight of the i-th task, R i represents the original path weight corresponding to the i-th task, E ik Score the efficiency of the i-th task and the k-th link, α, β and γ k are adjustment coefficients respectively, and m is the total number of evaluation links; Based on the adjusted path weights, resource allocation and scheduling plans are optimized, and the execution order and resource input of the transportation tasks are reconfigured in combination with transportation demand and resource availability to obtain optimized transportation instructions.

4. The MES equipment remote monitoring and management system for rare earth production according to claim 1, characterized in that: The steps for obtaining the adjusted production task timing are as follows: Based on the optimized transport instructions, the time difference between the arrival of materials and the availability of equipment is calculated, and the start and pause time points of the task are adjusted. The formula is: Among them, T adjust,i is the adjusted start time of the i-th task, T atrive,i is the estimated arrival time of the material, T ready,i is the estimated time the equipment is available, ZR i is the task urgency index; The adjusted start time is used to rearrange the production process to obtain an adjusted production task sequence.

5. The MES equipment remote monitoring and management system for rare earth production according to claim 1, characterized in that: The steps for obtaining the current process state diagram are: Based on the optimized transportation instructions, updating the nodes and edges in the Petri net, recording the state transition of each production stage, including the task start, in progress and completed status, and obtaining an updated Petri net state model; The production process is analyzed according to the updated Petri net state model, and the state of the Petri net is converted into a process state diagram through mapping to obtain the current process state diagram.

6. The MES equipment remote monitoring and management system for rare earth production according to claim 1, characterized in that: The steps for obtaining the task execution status result are: According to the adjusted resource allocation plan, extract the real-time operation data and material flow data of the production equipment, including the working status of the equipment, the material transmission speed and the time point of arrival at the node, and integrate the data to form the equipment operation and material flow data set; Based on the equipment operation and material flow data set, the progress of each production task is tracked, the task start time, current progress status and stage completion rate are recorded, and real-time task progress tracking data is generated; According to the real-time task progress tracking data, the execution status of each task is analyzed to check whether there are delays, stagnation or abnormalities, and the analysis results are summarized to obtain the task execution status results.

Images