CN-122022372-A - Dynamic plan scheduling system for intelligent counting factory

CN122022372ACN 122022372 ACN122022372 ACN 122022372ACN-122022372-A

Abstract

The invention relates to the technical field of data management, and particularly discloses a dynamic scheduling system of a digital intelligent factory, which comprises a real-time state identification module, a degradation generation module, a health degradation module, a scheduling preset module, a redistribution module and an instruction execution module, wherein the real-time state tensor of the factory is generated by combining equipment real-time working condition parameters, material real-time stock water level and order delivery priority; the method comprises the steps of continuously analyzing equipment working condition parameters in tensors in the time domain to obtain characteristic degradation curves of key moving parts of equipment, predicting sub-health precursor time stamps of the equipment according to curve inflection points, collecting the characteristic degradation curves into a dynamic constraint matrix, combining the matrix with order priority, identifying space-time resource conflict nodes of a process and equipment limited time period, screening and adapting substitution equipment for the conflict process to form a reassignment scheme, and finally disassembling the scheme into three types of instructions and synchronously issuing the three types of instructions to corresponding execution ends.

Inventors

ZHU WEIJUN
ZHU TAO
YU JIANING
LIU LINSHENG
Xu Panbin
Xi Yuanyang

Assignees

浙江伏特宝科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260226

Claims (10)

1. The intelligent factory dynamic plan scheduling system is characterized by comprising a real-time state identification module, a degradation generation module, a health degradation module, a plan presetting module, a reassignment module and an instruction execution module, wherein: the real-time state identification module is used for generating a real-time state tensor of the factory according to real-time working condition parameters of equipment in the factory, real-time stock water level of materials and order delivery priority; The degradation generation module is used for carrying out continuous time domain analysis on the equipment real-time working condition parameters in the real-time state tensor to obtain a characteristic degradation curve of the equipment key moving parts in the factory; the health degradation module is used for predicting a precursor time stamp of the equipment in the factory to enter a sub-health state based on the inflection point of the characteristic degradation curve, and collecting the precursor time stamp into a dynamic constraint matrix; the scheduling preset module is used for identifying space-time resource conflict nodes generated by overlapping the precursor time stamp and the processing time of a given process in a scheduling execution period of the factory based on the dynamic constraint matrix and the order delivery priority in the real-time state tensor; The redistribution module is used for screening substitute equipment which has the same processing function and has a current load rate lower than a preset elastic threshold value for the collision procedure of the factory according to the space-time resource collision node to obtain a redistribution scheme of the factory; the instruction execution module is used for disassembling the redistribution scheme into a material offset instruction, a process parameter switching instruction and a job guidance updating instruction, and synchronously issuing the instructions to the corresponding execution ends.
2. The intelligent factory dynamic planning and scheduling system of claim 1, wherein the real-time status recognition module is configured to, when executing the generation of the factory real-time status tensor according to the factory equipment real-time operating condition parameters, the material real-time stock level and the order delivery priority: Collecting real-time working condition parameters of equipment including load, temperature and vibration from an equipment controller of a factory, and generating a working condition time sequence vector of the factory by taking equipment numbers as indexes; Acquiring the types and the usable quantity of the stock materials of the factory, and generating real-time stock water levels of the materials of the factory by combining the line side bin consumption rate of the factory so as to obtain the stock state vector of the factory; extracting delivery time and customer grade of an unbounded order in the factory, and sequencing according to the degree of urgency to generate an order delivery priority so as to form an order weight vector of the factory; And aligning the working condition time sequence vector, the inventory state vector and the order weight vector along the same time axis, and carrying out three-dimensional tensor reconstruction by taking equipment numbers, material codes and order numbers as dimensions to generate the real-time state tensor of the factory.
3. The intelligent plant dynamic plan scheduling system of claim 1, wherein the degradation generation module is configured to, when performing continuous time domain analysis of the device real-time operating condition parameters in the real-time state tensor to obtain a characteristic degradation curve of the device critical moving parts in the plant: extracting a time sequence sheet of the main shaft load rate and the feed shaft vibration amplitude from the real-time state tensor; Removing transient load peaks caused by abrupt change of workpiece materials or uneven cutting allowance in the time sequence of the spindle load rate, and obtaining a smooth load baseline representing basic energy consumption trend; separating out characteristic frequency components related to the rotation fundamental frequency and frequency multiplication of the moving part in the time sequence of the vibration amplitude of the feeding shaft to obtain a vibration energy spectrum density sequence reflecting the surface fatigue degree of the part; Fusing the smooth load baseline and the vibration energy spectrum density sequence according to a uniform time index; and fitting the fused data by taking time as an independent variable and the degradation degree as an independent variable to obtain a characteristic degradation curve of the key moving parts of the equipment in the factory.
4. The dynamic plan scheduling system of a digital intelligent factory according to claim 3, wherein the degradation generating module is specifically configured to, when executing the fitting of the fused data with time as an independent variable and degradation degree as a dependent variable, obtain a characteristic degradation curve of a key motion component of the device in the factory: Calculating a first derivative of the characteristic degradation curve at the current moment to obtain an instantaneous degradation rate of key moving parts of equipment in the factory; Comparing the instantaneous degradation rate with a preset health-stage typical rate interval, and judging that the equipment key moving part enters an early degradation stage from a normal operation stage when the instantaneous degradation rate exceeds the upper limit value of the health-stage typical rate interval; And recording the time node at which the judgment occurs, and marking the coordinate point corresponding to the time node on the characteristic degradation curve.
5. The system of claim 1, wherein the health degradation module, when executing a predictive timestamp based on the inflection point of the characteristic degradation curve to predict an impending sub-health status of a device in the plant, is specifically configured to: Performing second-order differential operation on the characteristic degradation curve, and identifying candidate positions with the curve concave-convex properties on the characteristic degradation curve; screening out extreme points with curvature values exceeding a preset curvature threshold value from the candidate positions, and marking the abscissa time value corresponding to the extreme points as the starting moment of the key moving part of the equipment deviating from the reference operating characteristic; Acquiring a degradation degree value corresponding to the initial moment on a characteristic degradation curve, and comparing the degradation degree value with a pre-calibrated sub-health state critical threshold; when the degradation level value reaches or exceeds the sub-health state critical threshold, directly determining the starting moment as the precursor timestamp; And when the degradation degree value does not reach the sub-health state critical threshold value, determining the intersection point moment as the precursor timestamp based on the intersection point moment when the degradation rate mean value extrapolation degradation curve after the starting moment extends to the sub-health state critical threshold value.
6. The intelligent plant dynamic plan scheduling system of claim 5, wherein the health degradation module, when executing the pooling of the precursor timestamps into a dynamic constraint matrix, is specifically configured to: Establishing a precursor time stamp sequence taking equipment identification as an index for equipment of the factory, and writing the precursor time stamp into the precursor time stamp sequence according to time sequence; extracting start-stop time of a processing period of the equipment which is already scheduled in a future planning period from the real-time state tensor; Overlapping a time point in the precursor timestamp sequence with the processing time period of the corresponding equipment, and marking the processing time period as a limited time period when the time point falls into the section of the processing time period; And constructing an occupied state table of the equipment by taking the equipment identifier as a row index and taking a time unit in a planning period as a column index, setting cells corresponding to the limited period as unavailable states, setting other cells as available states, and generating a dynamic constraint matrix of the equipment.
7. The intelligent plant dynamic scheduling system of claim 1, wherein the scheduling pre-setting module, when executing the order delivery priority based on the dynamic constraint matrix and the real-time state tensor, identifies a space-time resource conflict node of the plant during a scheduled execution period due to the overlapping of the precursor timestamp and a predetermined process processing time, is specifically configured to: Traversing the cells marked as unavailable states in the dynamic constraint matrix, and extracting corresponding equipment identifiers and starting points and ending points of a limited period; reading a given working procedure allocated to the equipment identifier in the current scheduling scheme from the real-time state tensor, and acquiring the planned starting time and the planned finishing time of the given working procedure; Performing intersection operation on the limited time period and a time interval formed from the planned starting time to the planned finishing time, and judging that space-time resource occupation conflicts exist between the established working procedure and the limited time period when an overlapping interval exists between the two time periods; and combining the procedure numbers, the corresponding equipment identifiers and the time boundaries of the conflict overlapping intervals, which are in conflict, into conflict event records, and identifying the conflict event records as space-time resource conflict nodes.
8. The intelligent plant dynamic scheduling system of claim 1, wherein the scheduling pre-setting module, when executing the order delivery priority based on the dynamic constraint matrix and the real-time state tensor, identifies a space-time resource conflict node of the plant during a scheduled execution period due to the overlapping of the precursor time stamp and the predetermined process processing time, is further specifically configured to: Extracting order delivery priorities associated with process numbers of the equipment from the real-time state tensors; converting the delivery remaining days and the customer grade weight in the order delivery priority into quantitative scores for representing the delay treatment risk of the order according to a preset emergency degree mapping rule; according to the quantized scores, the space-time resource conflict nodes in the same planning period are arranged in a descending order, and a conflict processing priority sequence of the space-time resource conflict nodes is generated; And adding the conflict processing priority sequence into the corresponding time-space resource conflict node attribute field, and taking the conflict processing priority sequence as an execution sequence basis of a subsequent reassignment scheme to sequentially handle conflict nodes.
9. The dynamic plan scheduling system of claim 1, wherein the reassignment module, when executing the process of screening the replacement equipment with the same processing function and with the current load rate lower than the preset elastic threshold for the process of the plant according to the space-time resource conflict node, is specifically configured to: Extracting processing function types required by a conflict procedure from the space-time resource conflict node, and searching all equipment with the processing function types in an equipment capacity configuration table to obtain a candidate equipment list of the factory; Reading the current load rate of each device in the candidate device list from the real-time state tensor, screening out devices with the load rate lower than a preset elastic threshold value, and obtaining an available device list of the factory; querying the available state of the available equipment list in the dynamic constraint matrix in a conflict period, and eliminating equipment marked as unavailable state in the conflict period to obtain a final optional equipment list of the factory; And selecting the equipment with the lowest load rate from the final selectable equipment list as a substitute target, combining the process identification of the conflict process, the original equipment identification, the equipment identification of the substitute target equipment and the adjusted planned processing period into a reallocation record, and writing the reallocation record into a reallocation scheme.
10. The system of claim 1, wherein the instruction execution module, when executing the splitting of the redistribution scheme into a material offset instruction, a process parameter switching instruction, and a job guidance update instruction, is specifically configured to: Extracting a process identifier of a conflict process, an equipment identifier of a substitute target equipment and an adjusted planned machining period from the redistribution scheme, and generating a material offset instruction for a material distribution unit; the technological parameter set required by the conflict procedure of processing of the substitute target equipment is called from an equipment parameter library, and a technological parameter switching instruction for an equipment controller is generated; Associating the operation drawing and the operation specification of the conflict procedure with the equipment identifier of the substitution target equipment to generate an operation guidance updating instruction aiming at an operation terminal; And respectively routing the material offset instruction, the process parameter switching instruction and the operation guidance updating instruction to a corresponding material distribution unit controller, a corresponding equipment controller and a corresponding operation terminal display screen, receiving an instruction receiving confirmation signal returned by an execution end, and completing synchronous issuing of the instructions.

Description

Dynamic plan scheduling system for intelligent counting factory Technical Field The invention relates to the technical field of data processing, in particular to a dynamic plan scheduling system of a digital intelligence factory. Background In the production operation process of the intelligent counting factory, the scheduling is used as a core link of production control, a key supporting effect is achieved on production efficiency and order delivery efficiency, the existing factory scheduling system is used for carrying out scheduling planning by depending on static equipment parameters, established process routes and fixed order information, dynamic production data such as equipment real-time working condition parameters, material real-time stock water level and order delivery priority cannot be collected and integrated in real time, and a monitoring and predicting mechanism of equipment health state is not established, so that the scheduling scheme is obviously disjointed from the actual production state of the factory, production resource problems caused by equipment state change are difficult to predict in advance, and the situation that procedure processing is not matched with equipment operation state is easy to cause. When the existing scheduling system is used for coping with production resource conflicts, a systematic conflict node identification and processing system is lacking, so that space-time resource conflicts generated by overlapping of sub-health precursor time of equipment and processing time of a given process cannot be accurately positioned, a scientific processing priority ordering rule is not established by combining order delivery requirements with conflict nodes, meanwhile, in a resource reallocation link of the conflict process, screening dimensions of replacement equipment are single, factors such as equipment load rate and dynamic operation constraint are not comprehensively considered, and the scheduled and adjusted instructions are issued in a lack of cooperativity, synchronous update of materials, processes and operation guidance cannot be realized, so that dynamic adjustment efficiency of a scheduling scheme is low, and flexible and efficient production requirements of a digital intelligent factory are difficult to adapt. Disclosure of Invention In order to achieve the above object, the present invention provides a dynamic plan scheduling system for a digital intelligent factory, which is characterized in that the system comprises a real-time status recognition module, a degradation generation module, a health degradation module, a plan presetting module, a reassignment module and an instruction execution module, wherein: the real-time state identification module is used for generating a real-time state tensor of the factory according to real-time working condition parameters of equipment in the factory, real-time stock water level of materials and order delivery priority; The degradation generation module is used for carrying out continuous time domain analysis on the equipment real-time working condition parameters in the real-time state tensor to obtain a characteristic degradation curve of the equipment key moving parts in the factory; the health degradation module is used for predicting a precursor time stamp of the equipment in the factory to enter a sub-health state based on the inflection point of the characteristic degradation curve, and collecting the precursor time stamp into a dynamic constraint matrix; the scheduling preset module is used for identifying space-time resource conflict nodes generated by overlapping the precursor time stamp and the processing time of a given process in a scheduling execution period of the factory based on the dynamic constraint matrix and the order delivery priority in the real-time state tensor; The redistribution module is used for screening substitute equipment which has the same processing function and has a current load rate lower than a preset elastic threshold value for the collision procedure of the factory according to the space-time resource collision node to obtain a redistribution scheme of the factory; the instruction execution module is used for disassembling the redistribution scheme into a material offset instruction, a process parameter switching instruction and a job guidance updating instruction, and synchronously issuing the instructions to the corresponding execution ends. In a preferred embodiment, the real-time status recognition module is specifically configured to, when executing the generation of the real-time status tensor of the factory according to the real-time operating condition parameters of the equipment in the factory, the real-time stock water level of the material and the order delivery priority: Collecting real-time working condition parameters of equipment including load, temperature and vibration from an equipment controller of a factory, and generating a working condition time sequence vecto